Neurocritical Care.2018 Continuum

DECEMBER 2018
VOL. 24 NO. 6 Neurocritical Care

Guest Editor: Jose I. Suarez, MD, FNCS, FANA
1586 Editor’s Preface

Editor-in-Chief: Steven L. Lewis, MD, FAAN
REVIEW ARTICLES
1588 Cerebral Edema and Elevated Intracranial Pressure

Matthew A. Koenig, MD, FNCS
1603 Intracerebral Hemorrhage

Wendy C. Ziai, MD, MPH, FAHA, FNCS, FESO;
J. Ricardo Carhuapoma, MD, FAHA
1623 Subarachnoid Hemorrhage

Susanne Muehlschlegel, MD, MPH, FNCS, FCCM
1658 Management of Stroke in the Neurocritical Care Unit

Chethan P. Venkatasubba Rao, MD, FNCS; Jose I. Suarez, MD, FNCS, FANA
1683 Status Epilepticus, Refractory Status Epilepticus,

and Super-refractory Status Epilepticus
Sarah E. Nelson, MD; Panayiotis N. Varelas, MD, PhD, FNCS, FAAN
1708 Coma and Brain Death

Alejandro A. Rabinstein, MD, FAAN
1732 Management of Comatose Survivors of Cardiac Arrest

David B. Seder, MD, FCCP, FCCM, FNCS
1753 Critical Care of Neuromuscular Disorders

Diana Greene-Chandos, MD, FNCS;
DENOTES CONTINUUM
Michel Torbey, MD, MPH, FCCM, FAHA, FNCS , FAAN
AUDIO INTERVIEW
DENOTES SUPPLEMENTAL 1776 Multimodality Monitoring in the Neurocritical Care Unit

DIGITAL CONTENT Lucia Rivera Lara, MD, MPH; Hans Adrian Püttgen, MD
Copyright © American Academy of Neurology. Unauthorized reproduction of this article is prohibited.

ETHICAL AND MEDICOLEGAL ISSUES
1789 Ethical Considerations in End-of-life Care in the Face of

Clinical Futility
Joseph S. Kass, MD, JD, FAAN; Ariane Lewis, MD;
Michael A. Rubin, MD, MA
PRACTICE ISSUES
1794 End-of-life Considerations and Shared Decision Making in

Neurocritical Care
Christos Lazaridis, MD, EDIC
1800 Neurocritical Care Coding for Neurologists

Marc R. Nuwer, MD, PhD, FAAN; Paul M. Vespa, MD, FAAN
SELF-ASSESSMENT AND CME
1576 Learning Objectives and Core Competencies
1811 Instructions for Completing Postreading Self-Assessment and CME

Test and Tally Sheet
1813 Postreading Self-Assessment and CME Test
1827 Postreading Self-Assessment and CME Test—Preferred Responses
1838 Appendix A: Summary of Evidence-based Guideline for Clinicians:

Practice Guideline Update: Disorders of Consciousness
1847 Appendix B: AAN Summary of Evidence-based Guideline for

Clinicians: Update: Determining Brain Death in Adults
1849 Errata
1850 Index
List of Abbreviations (Back Cover)

CONTRIBUTORS
Jose I. Suarez, MD, FNCS, FANA, Diana Greene-Chandos,

Guest Editor MD, FNCS
Professor, Director of Director of Neuroscience Critical
Neurosciences Critical Care, Care for Education; Program
Departments of Anesthesiology Director, Neuroscience Critical
and Critical Care Medicine, Care Fellowship, Wexner Medical
Neurology, and Neurosurgery, Center, Arthur G. James Cancer
Johns Hopkins University School Hospital and Richard J. Solove
of Medicine, Baltimore, Maryland Research Institute; Assistant
Professor, Departments of
Relationship Disclosure: Dr Suarez has
received research/grant support from Neurology and Neurosurgery,
the National Institute of Neurological The Ohio State University College
Disorders and Stroke and as co-investigator
in the SETPOINT2 (Stroke-related Early
of Medicine, Columbus, Ohio
Tracheostomy Versus Prolonged Orotracheal
Relationship Disclosure: Dr Greene-Chandos
Intubation in Neurocritical Care Trial) study
has received personal compensation as a
from the Patient-Centered Outcomes
speaker for the Ecuador Neurocritical Care
Research Institute. Dr Suarez is the current
Society Annual Meeting and the Neurocritical
president and a member of the board of
Care Society Annual Meeting and has
directors of the Neurocritical Care Society.
received research/grant support from
Biogen, the Neuroemergency Treatment Trials
Unlabeled Use of Products/Investigational
Grant from the National Institutes of Health,
Use Disclosure: Dr Suarez reports
and the Stroke Network Grant from the
no disclosure.
National Institute of Neurological Disorders
and Stroke. Dr Greene-Chandos has worked
as a consultant for court cases.
J. Ricardo Carhuapoma, Unlabeled Use of Products/Investigational

MD, FAHA Use Disclosure: Dr Greene-Chandos reports
Associate Professor, no disclosure.
Departments of Neurology,
Neurosurgery, and
Anesthesiology and Critical
Care Medicine, Johns Hopkins
University School of Medicine,
Baltimore, Maryland
Relationship Disclosure: Dr Carhuapoma
reports no disclosure.

Use Disclosure: Dr Carhuapoma discusses
the unlabeled/investigational use of
alteplase for the treatment of intracerebral
hemorrhage and discusses the use of
several devices, including ultrasound
microcatheters, sonothrombolysis for
minimally invasive subcortical parafascicular
transsulcal access for clot evacuation
(MiSPACE), and the stereotactic mechanical
(suction/vibration/aspiration) thrombolytic
technique for minimally invasive evacuation
of intracerebral hemorrhage.
1578 D EC E M B ER 2 0 1 8

Joseph S. Kass, MD, JD, FAAN Christos Lazaridis, MD, EDIC
Associate Dean, Office of Associate Professor, Division
Student Affairs; Professor of Neurocritical Care,
of Neurology, Psychiatry, Departments of Neurology and
and Medical Ethics; Director, Neurosurgery, University of
Alzheimer’s Disease and Chicago, Chicago, Illinois
Memory Disorders Center,
Relationship Disclosure: Dr Lazaridis serves on
Baylor College of Medicine; the editorial board of Neurocritical Care and
Chief of Neurology, Ben Taub has received research/grant support from
Cheetah Medical, Inc.
General Hospital,
Houston, Texas Unlabeled Use of Products/Investigational
Use Disclosure: Dr Lazaridis reports
Relationship Disclosure: Dr Kass serves as no disclosure.
associate editor of ethical and medicolegal
issues for Continuum, as an associate editor
for Continuum Audio, as a neurology section
editor of Ferri’s Clinical Advisor for Elsevier, Ariane Lewis, MD
and as co-editor of Neurology Secrets,
Sixth Edition. Dr Kass has received personal Associate Professor,
compensation for CME lectures from Departments of Neurology
Pri-Med Medical Group and has received and Neurosurgery, New York
personal compensation as a medicolegal
consultant in legal cases involving criminal University Langone Medical
cases, malpractice, and personal injury. Center, New York, New York
Unlabeled Use of Products/Investigational Relationship Disclosure: Dr Lewis has
Use Disclosure: Dr Kass reports received personal compensation as
no disclosure. a workshop lecturer for the American
Academy of Neurology and for grand rounds
lectures from Mount Sinai Hospital, Mount
Sinai Beth Israel Hospital, and Newark
Beth Israel Medical Center, and for work
Matthew A. Koenig, MD, FNCS as a legal consultant for David A. Axelrod &
Associates, PC.
Medical Director of
Telemedicine, The Queen’s Unlabeled Use of Products/Investigational
Use Disclosure: Dr Lewis reports
Health Systems; Associate no disclosure.
Professor of Medicine,
John A. Burns School of
Medicine, University of Hawaii,
Honolulu, Hawaii
Relationship Disclosure: Dr Koenig has
received research/grant support as
principal investigator of a study for the
Hawaii Department of Health Neurotrauma
Special Fund and receives publishing
royalties from Rutgers University Press.

Use Disclosure: Dr Koenig reports
no disclosure.
C O N T I N U U M J O U R N A L .C O M 1579

CONTRIBUTORS (CONTINUED)
Susanne Muehlschlegel, MD, Sarah E. Nelson, MD

MPH, FNCS, FCCM Assistant Professor,
Associate Professor, Director Departments of Neurology and
of Neurocritical Care Research, Anesthesiology and Critical
Departments of Neurology, Care Medicine, Johns Hopkins
Anesthesiology, and Surgery, University School of Medicine,
University of Massachusetts Baltimore, Maryland
Medical School, Worcester,
Relationship Disclosure: Dr Nelson receives
Massachusetts grant support from the Johns Hopkins
Anesthesiology and Critical Care Medicine
Relationship Disclosure: Dr Muehlschlegel (ACCM) Stimulating and Advancing ACCM
has received research/grant support from Research (StAAR) program.
the National Institutes of Health/National
Institute of Child Health and Human Unlabeled Use of Products/Investigational
Development and the Prize for Academic Use Disclosure: Dr Nelson discusses
Collaboration and Excellence (PACE) from the unlabeled/investigational use of
the University of Massachusetts Memorial allopregnanolone, deep brain stimulation,
Medical Group. Dr Muehlschlegel receives desflurane, diazepam, electroconvulsive
partial research salary support as the site therapy, fosphenytoin, gabapentin,
principal investigator for the INTREPID isoflurane, IV immunoglobulin, ketamine,
(Impact of Fever Prevention in Brain Injured ketogenic diet, lacosamide, levetiracetam,
Patients) trial sponsored by C. R. Bard Inc. lidocaine, lorazepam, methylprednisolone,
Dr Muehlschlegel has received midazolam, pentobarbital, phenobarbital,
compensation for serving as a course phenytoin, plasma exchange, propofol,
director for the American Academy pyridoxine, thiopental, topiramate, vagal
of Neurology. nerve stimulation, and valproic acid for the
treatment of refractory status epilepticus.
Use Disclosure: Dr Muehlschlegel discusses
the unlabeled/investigational short-term
use of antifibrinolytics (ε-aminocaproic
acid and tranexamic acid) for the treatment
of early aneurysm bleeding, the use of
fludrocortisone for the treatment of
cerebral salt wasting syndrome after
subarachnoid hemorrhage, the use of
levetiracetam for seizure prophylaxis,
and the use of milrinone, nicardipine, and
verapamil as endovascular therapy using
intraarterial vasodilators for the treatment
of subarachnoid hemorrhage.
1580 D EC E M B ER 2 0 1 8

Marc R. Nuwer, MD, PhD, FAAN Alejandro A. Rabinstein, MD,
Professor, Vice Chair, FAAN
Department of Neurology, David Professor of Neurology, Mayo
Geffen School of Medicine; Clinic, Rochester, Minnesota
Department Head, Department Relationship Disclosure: Dr Rabinstein
of Clinical Neurophysiology, receives royalties from Elsevier and Oxford
Ronald Reagan University of University Press and has received research
support from DJO Global, Inc.
California Los Angeles Medical
Center, University of California Unlabeled Use of Products/Investigational
Los Angeles, Los Angeles, Use Disclosure: Dr Rabinstein reports
no disclosure.
California
Relationship Disclosure: Dr Nuwer has
received personal compensation for
Lucia Rivera Lara, MD, MPH
serving as an honorary editor for Clinical
Neurophysiology; for serving on the board Assistant Professor,
of directors of CortiCare, Inc; and for serving Departments of Neurology and
on the editorial boards of the Journal of
Clinical Neurophysiology and Neurology
Anesthesiology and Critical
Clinical Practice. Dr Nuwer receives royalties Care Medicine, Johns Hopkins
from Cambridge University Press and has University School of Medicine,
received research/grant funding from the
National Institutes of Health, Second Sight, Baltimore, Maryland
and the United States Army. Dr Nuwer has
given expert medicolegal testimony in a Relationship Disclosure: Dr Rivera Lara has
court deposition and a court hearing. received research/grant support from
Covidien Ltd/Medtronic; the Johns Hopkins
Unlabeled Use of Products/Investigational Anesthesiology and Critical Care Medicine
Use Disclosure: Dr Nuwer reports (ACCM) Stimulating and Advancing ACCM
no disclosure. Research (StAAR) award; and Ornim, Inc.

Use Disclosure: Dr Rivera Lara discusses the
unlabeled/investigational use of near-infrared
Hans Adrian Püttgen, MD spectroscopy as a surrogate of cerebral blood
Assistant Professor, flow to measure cerebral autoregulation.
Department of Neurology,
Johns Hopkins University
School of Medicine,
Baltimore, Maryland Michael A. Rubin, MD, MA
Assistant Professor,
Relationship Disclosure: Dr Püttgen reports
no disclosure.
Departments of Neurology
and Neurotherapeutics,
Unlabeled Use of Products/Investigational University of Texas
Use Disclosure: Dr Püttgen discusses the
unlabeled/investigational use of near-infrared Southwestern Medical Center,
spectroscopy as a surrogate of cerebral blood Dallas, Texas
flow to measure cerebral autoregulation.
Relationship Disclosure: Dr Rubin reports
no disclosure.

Use Disclosure: Dr Rubin reports
no disclosure.

David B. Seder, MD, FCCP, Panayiotis N. Varelas, MD, PhD,

FCCM, FNCS FNCS, FAAN
Chairman, Department of Professor of Neurology,
Critical Care Services, Wayne State University;
Maine Medical Center, Division Head, Neurosciences
Portland, Maine; Associate Critical Care Services,
Professor of Medicine, Tufts Director Neuro-Intensive
University School of Medicine, Care Unit, Henry Ford
Boston, Massachusetts Hospital, Detroit, Michigan
Relationship Disclosure: Dr Seder receives Relationship Disclosure: Dr Varelas serves on
grant support from the Patient-Centered the board of directors of the Neurocritical
Outcomes Research Institute Care Society, on the editorial board of
(CER-1602-34137). Neurocritical Care, and on an advisory board
of Portola Pharmaceuticals, Inc. Dr Varelas
Unlabeled Use of Products/Investigational has received personal compensation
Use Disclosure: Dr Seder discusses the for speaking engagements for Portola
unlabeled/investigational use of the Pharmaceuticals, Inc and UCB SA and
bispectral index in determining the severity of receives publishing royalties from Springer.
brain injury after cardiac arrest. Dr Varelas receives research/grant support
from Bard Pharmaceuticals Limited; Edge
Therapeutics, Inc; Marinus Pharmaceuticals,
Inc; the National Institutes of Health; the
Patient-Centered Outcomes Research
Institute; and Portola Pharmaceuticals, Inc.
Michel Torbey, MD, MPH,
FCCM, FAHA, FNCS , FAAN Use Disclosure: Dr Varelas discusses
Professor and Chairman, the unlabeled/investigational use of
Department of Neurology; allopregnanolone, deep brain stimulation,
desflurane, diazepam, electroconvulsive
University of New Mexico, therapy, fosphenytoin, gabapentin,
Albuquerque, New Mexico isoflurane, IV immunoglobulin, ketamine,
ketogenic diet, lacosamide, levetiracetam,
Relationship Disclosure: Dr Torbey has lidocaine, lorazepam, methylprednisolone,
received personal compensation as the past midazolam, pentobarbital, phenobarbital,
president of the Neurocritical Care Society phenytoin, plasma exchange, propofol,
and has received grant support from the pyridoxine, thiopental, topiramate, vagal
National Institute of Neurological Disorders nerve stimulation, and valproic acid for the
and Stroke. treatment of refractory status epilepticus.

Use Disclosure: Dr Torbey reports
no disclosure.
1582 D EC E M B ER 2 0 1 8

Chethan P. Venkatasubba Rao, Wendy C. Ziai, MD, MPH, FAHA,
MD, FNCS FNCS, FESO
Section Head, Vascular Associate Professor,
Neurology and Neurocritical Departments of Neurology,
Care; Program Director, Neurosurgery, and
Neurocritical Care Fellowship; Anesthesiology and Critical
Assistant Professor, Baylor Care Medicine, Johns Hopkins
College of Medicine; Medical University School of Medicine,
Director, Neurocritical Care, Baltimore, Maryland
Baylor St. Luke’s Medical
Relationship Disclosure: Dr Ziai serves as
Center, Houston, Texas an associate editor for Neurocritical Care,
on an advisory board for C.R. Bard Inc,
Relationship Disclosure: Dr Venkatasubba and has received personal compensation
Rao has received personal compensation as a consultant for and research/grant
as an editorial board member of Brain support from HeadSense Medical Ltd. Dr
Disorders & Therapy. Ziai has received grant support from the
Johns Hopkins Anesthesiology and Critical
Unlabeled Use of Products/Investigational Care Medicine (ACCM) Stimulating and
Use Disclosure: Dr Venkatasubba Rao reports Advancing ACCM Research (StAAR) program
no disclosure. and from the National Institutes of Health
(5U01NS062851, 1U01NS08082).

Use Disclosure: Dr Ziai discusses the
Paul M. Vespa, MD, FAAN unlabeled/investigational use of alteplase
Assistant Dean of Critical Care for the treatment of intracerebral
hemorrhage and discusses the use of
Medicine, Gary L. Brinderson several devices, including ultrasound
Family Chair in Neurocritical microcatheters, sonothrombolysis for
Care , Neurology Director of minimally invasive subcortical parafascicular
transsulcal access for clot evacuation
Neurocritical Care, Professor (MiSPACE), and the stereotactic mechanical
of Neurosurgery and Neurology, (suction/vibration/aspiration) thrombolytic
technique for minimally invasive evacuation
David Geffen School of Medicine,
of intracerebral hemorrhage.
University of California
Los Angeles,
Los Angeles, California
Relationship Disclosure: Dr Vespa has
received personal compensation as an
editor for Critical Care Medicine and
Neurocritical Care. Dr Vespa has received
personal compensation as a consultant for
Sage Therapeutics, has received research/
grant support from the National Institutes
of Health, and holds stock options in
InTouch Health.

Use Disclosure: Dr Vespa reports
no disclosure.

Self-Assessment and CME Test Writers
Adam G. Kelly, MD James W. M. Owens Jr, MD, PhD

Associate Professor Associate Professor of
of Neurology; Chief, Neurology, Adjunct Associate
Neurovascular Division, Professor of Pediatrics,
University of Florida, University of Washington
Gainesville, Florida School of Medicine,
Seattle, Washington
Relationship Disclosure: Dr Kelly has
received personal compensation as CME Relationship Disclosure: Dr Owens serves as
editor of Neurology. CME co-editor for Neurology and receives
publishing royalties from UpToDate, Inc.
Use Disclosure: Dr Kelly reports no disclosure. Unlabeled Use of Products/Investigational
Use Disclosure: Dr Owens reports
no disclosure.
Peer Reviewers
Continuum is a peer-reviewed journal. For each review article, at least two peer
reviewers anonymous to the authors are selected to assess quality, balance,
and clarity of articles. Critical reviews of the material at other stages of the
development process may also be conducted. Upon receipt of final manuscripts,
the editor-in-chief makes the final decision as to whether material is suitable for
publication in Continuum.
We are thankful to the following peer reviewers of articles published during 2018
for the immense work they put in and their excellent assessment of manuscripts
to help ensure that only the highest-quality information is published in Continuum:
Michael J. Bradshaw, MD Aarti Sarwal, MD, FAAN

Hannah R. Briemberg, MD, FRCPC Divya Singhal, MD
Claude Hemphill III, MD, FAAN Erik K. St. Louis, MD, FAAN, FAASM
Karlo J. Lizarraga, MD, MS Jack W. Tsao, MD, DPhil, FAAN
Scott Lucchese, MD Nagagopal Venna, MBBS, FAAN
Robert W. McMahon Jr, MD, FAAN Allison L. Weathers, MD, FAAN
Jonathan W. Mink, MD, PhD, FAAN Max Wiznitzer, MD, FAAN
Adam Quick, MD Edward Y. Zamrini, MD
Richard A. Rison, MD, FAAN
1584 D EC E M B ER 2 0 1 8

EDITOR’S PREFACE
A Critical Look Back and Ahead

This issue marks the end of my first term and sixth year as
editor-in-chief of Continuum and, I am pleased to announce, the
start of my final 4-year term in this role. I am delighted that this
juncture is marked by this well-organized and remarkably thorough
issue on neurocritical care. Guest Editor Jose I. Suarez, MD, FNCS,
FANA, has brought together many renowned experts in the field to inform us of the
state of the art of diagnosis and management of our critically ill neurologic patients.
The issue begins with an overview by Dr Matthew Drs Diana Greene-Chandos and Michel Torbey
A. Koenig on the pathophysiology and management discuss the critical care of patients with neuromuscular
of cerebral edema, elevated intracranial pressure, disorders, including those with primary neuromuscular
and cerebral herniation syndromes, life-threatening disorders admitted to the critical care unit and those
processes for which diagnosis and management are with neuromuscular disorders that arise because of
integral to many of the disorders featured in the critical medical illness. In the final review article,
articles that follow. Next, Drs Wendy C. Ziai and Drs Lucia Rivera Lara and Hans Adrian Püttgen
J. Ricardo Carhuapoma provide an up-to-date review provide an overview of the status of multimodality
of the management of spontaneous intracerebral monitoring in the neurocritical care unit and a
hemorrhage. Dr Susanne Muehlschlegel then reviews glimpse into its future.
the diagnosis and management of subarachnoid In the Ethical and Medicolegal Issues article,
hemorrhage, a highly morbid disorder in which rapid Drs Joseph S. Kass, Ariane Lewis, and Michael A. Rubin
recognition and management are key components of discuss the ethical considerations in end-of-life care
neurologic practice. Moving on to ischemic stroke, in the face of clinical futility. This article segues
Drs Chethan P. Venkatasubba Rao and Jose I. Suarez nicely into the Practice Issues article in which
discuss the management of acute ischemic stroke and Dr Christos Lazaridis discusses the importance of
the specific issues related to management of patients shared decision making in the neurocritical care unit.
with stroke in the neurocritical care unit. Next, Drs Marc R. Nuwer and Paul M. Vespa provide
Drs Sarah E. Nelson and Panayiotis N. Varelas a practical overview of the important nuances of
discuss the current definitions and management coding neurocritical care.
of status epilepticus, refractory status epilepticus, After reading the issue and taking the Postreading
and super-refractory status epilepticus. Next, Self-Assessment and CME Test written by Drs Adam
Dr Alejandro A. Rabinstein reviews the systematic G. Kelly and James W. M. Owens Jr, you may earn
evaluation of patients in coma and provides the up to 20 AMA PRA Category 1 CreditsTM toward
approach to, and avoidance of pitfalls in, the self-assessment CME or, for Canadian participants,
determination of brain death, followed by the article a maximum of 20 hours toward the Self-Assessment
by Dr David B. Seder, who discusses the management Program (Section 3) of the Maintenance of Certification
and prognostication of comatose survivors of Program of the Royal College of Physicians and
cardiac arrest, in which neurologists are critically Surgeons of Canada. Additional credit can be
involved. obtained by listening to Continuum Audio interviews
Moving from disorders of the central nervous associated with this and other Continuum issues,
system to disorders of the peripheral nervous system, available to all subscribers, and completing tests on
1586 DECEMBER 2018

the Continuum Audio web platform or app. curricular cycle as well as timeliness. We look
Continuum Audio is also accredited by the Royal forward to continuing to hone the curriculum so that
College of Physicians and Surgeons of Canada. all neurologists have a one-stop trusted source to
My sincerest thanks to Dr Suarez and all the inform the diagnosis and management of the
experts in neurocritical care who have provided such disorders neurologists encounter in practice.
well-written contributions to this issue to inform all The print version of Continuum underwent a
neurologists about the diagnosis, management, tremendous redesign in 2018, including the striking
prognostication, and counseling of our critically ill artwork on the cover by British artist Peter Grundy.
neurologic patients. Although this redesign may have come as a shock,
As 2018 comes to a close along with my first term I am thankful for your open-mindedness and for
as editor-in-chief, I thought I would provide a brief the nice comments we’ve received from so many
update on my editorial vision. As most readers know, of you.
Continuum is designed as a curriculum, with 15 core What might be less on your radar is that our
topics covered in single issues over 3-year cycles. website (ContinuumJournal.com) has also been
Each cycle includes a completely new take on each redesigned, allowing for easier access to both the
of the core topics; when I discuss the planning for current issue and previous issues, given the new
each issue with guest editors, I ask them to cover ability to more easily search for back issues by both
“everything a practicing neurologist needs to know year and topic. The website also integrates
now” about the topic, regardless of what was covered Continuum Audio, which can now be accessed by a
in the previous curriculum cycle. This ensures that each direct click from each Continuum article. These
new issue provides readers with a state-of-the-art, interviews represent a remarkable added value to
comprehensive review of each category of disorders, your subscription and another way of learning from
so that you can remain up-to-date in the diagnosis the topic expert authors. We hope that readers will
and management of those disorders and access a continue to explore the new website as a tool to
single print and digital resource on these conditions access all the rich content of Continuum from any
whenever and wherever you need it. My philosophy location, including at the point of care. The website
is to be as inclusive as possible within each topic and is mobile responsive and easily accessible on your
avoid holes in the curriculum, even at the expense of cell phones. Please report your feedback so that
often physically larger issues. We have also worked we can further enhance this important aspect of
to ensure that the three noncore topics in each cycle Continuum’s accessibility. Although print is here to
are carefully chosen based on both gaps in the stay for Continuum, the ability to use Continuum at
the point of care (or anywhere) is an important and,
as yet, underutilized function of this rich resource
and critical to our vision in the next 4 years.
I feel incredibly privileged and am thankful to
the American Academy of Neurology to be able to
continue to provide editorial leadership for the next
Dr Jose I. Suarez has brought 4 years and look forward to positioning the journal
together many renowned experts and its future leadership for further success for the
in the field to inform us of the state next many years.
of the art of diagnosis and —STEVEN L. LEWIS, MD, FAAN
management of our critically ill EDITOR-IN-CHIEF
neurologic patients. © 2018 American Academy of Neurology.
CONTINUUMJOURNAL.COM 1587

REVIEW ARTICLE

Cerebral Edema
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
and Elevated Intracranial
Pressure
By Matthew A. Koenig, MD, FNCS
ABSTRACT
PURPOSE OF REVIEW: This article reviews the management of cerebral edema,
elevated intracranial pressure (ICP), and cerebral herniation syndromes
in neurocritical care.
RECENT FINDINGS: While corticosteroids may be effective in reducing vasogenic

edema around brain tumors, they are contraindicated in traumatic cerebral
edema. Mannitol and hypertonic saline use should be tailored to patient
characteristics including intravascular volume status. In patients with
traumatic brain injury who are comatose, elevated ICP should be managed
with an algorithmic, multitiered treatment protocol to maintain an ICP of
22 mm Hg or less. Third-line ICP treatments include anesthetic agents,
induced hypothermia, and decompressive craniectomy. Recent clinical
trials have demonstrated that induced hypothermia and decompressive
craniectomy are ineffective as early neuroprotective strategies and should
be reserved for third-line management of refractory ICP elevation in
CITE AS: severe traumatic brain injury. Monitoring for cerebral herniation should
CONTINUUM (MINNEAP MINN)
2018;24(6, NEUROCRITICAL CARE):
include bedside pupillometry in supratentorial space-occupying lesions
1588–1602. and recognition of upward herniation in patients with posterior fossa
lesions.
Address correspondence to
Dr Matthew A. Koenig, The
Queen’s Health Systems, 1301 SUMMARY: Although elevated ICP, cerebral edema, and cerebral herniation
Punchbowl St, Neuroscience are interrelated, treatments should be based on the distinct
Institute QET5, Honolulu, HI
96816, mkoenig@queens.org.
pathophysiologic process. Focal lesions resulting in brain compression
are primarily managed with surgical decompression, whereas global or
RELATIONSHIP DISCLOSURE: multifocal brain injury requires a treatment protocol that includes medical
Dr Koenig has received
research/grant support as and surgical interventions.
principal investigator of a study
for the Hawaii Department of
Health Neurotrauma Special
Fund and receives publishing INTRODUCTION
E
royalties from Rutgers levated intracranial pressure (ICP), cerebral edema, and cerebral
University Press.
herniation syndromes are distinct but overlapping processes in
UNLABELED USE OF neurocritical care. Management of elevated ICP and cerebral edema is
PRODUCTS/INVESTIGATIONAL
USE DISCLOSURE:
heavily dependent on the underlying mechanism and clinical context.
Dr Koenig reports no disclosure. In patients with cerebral edema, determination of whether the patient
has vasogenic edema, cytotoxic edema, or hydrostatic edema is a critical first
© 2018 American Academy
step in identifying the most effective management strategy. Determining
of Neurology. whether elevated ICP is caused by global elevation in intracranial volume or focal
1588 DECEMBER 2018

injury that results in displacement of the brain is also crucial in choosing
appropriate treatments. This article focuses on the etiology and treatment of
cerebral edema in the neurocritical care setting, current concepts in the
treatment of intracranial hypertension, and cerebral herniation syndromes.
CEREBRAL EDEMA
Cerebral edema results from the pathologic accumulation of excess water within
the brain parenchyma. Vasogenic edema results from increased permeability
of the blood-brain barrier with extravasation of proteins, electrolytes, and water
into the parenchymal extracellular compartment. Common etiologies of
vasogenic edema include intraaxial and extraaxial brain tumors and cerebral
abscess. Vasogenic edema disproportionately affects subcortical white matter
with relative sparing of the cerebral cortex and subcortical gray matter. Cytotoxic
edema is caused by disruption of cell membranes within the brain parenchyma,
resulting in water shifts from the extracellular to the intracellular compartment.
The most common cause of cytotoxic edema is ischemic stroke. Less common
etiologies include hepatic encephalopathy and Reye syndrome. Traumatic brain
injury (TBI) and intracerebral hemorrhage (ICH) result in a combination of
cytotoxic and vasogenic edema. Cytotoxic edema affects both gray matter and
white matter structures, resulting in loss of cortical-subcortical distinction on
imaging studies. Hydrostatic cerebral edema results from transependymal
displacement of CSF from the ventricular compartment into the brain parenchyma,
typically due to obstructive hydrocephalus. Cerebral edema contributes to an
increase in intracranial volume. Global cerebral edema primarily results in a
global rise in ICP, while focal cerebral edema can result in cerebral herniation
syndromes with or without ICP elevation.
Treatment of Cerebral Edema

Treatment strategies for cerebral edema are heavily contingent on the underlying
etiology and type of cerebral edema. The mainstay of treatment of hydrostatic
edema due to obstructive hydrocephalus is CSF diversion, typically by placement
of an external ventricular drain (EVD). Depending on the underlying etiology,
vasogenic cerebral edema is treated with corticosteroids, osmotic agents, and
surgical decompression. Treatment options for cytotoxic edema are much more
limited. While osmotic agents have been used as a temporizing measure, the
evidence for efficacy is poor, and surgical decompression may be considered in
the appropriate clinical context.1
CORTICOSTEROIDS. Dexamethasone has been a mainstay of treatment for

peritumoral vasogenic edema for both intraaxial and extraaxial brain tumors
since the 1960s. The role of corticosteroids in the treatment of vasogenic edema
from cerebral abscess is less clear. Despite widespread use, few clinical trials have
been conducted to determine the efficacy, optimal dose, and appropriate
duration of corticosteroids for vasogenic edema.2 Dexamethasone is typically
started at a dose of 4 mg every 6 hours, and subsequent dose adjustments are
based on the clinical course. For brain tumors that are amenable to surgical
resection and radiation, dexamethasone may be tapered off over several weeks.
For patients with untreatable brain tumors, the dose of dexamethasone may need
to be increased over time as a palliative measure. In the acute treatment of
peritumoral edema, dexamethasone use should be limited to patients who have

CEREBRAL EDEMA AND ELEVATED INTRACRANIAL PRESSURE
significant symptoms attributable to cerebral edema rather than focal neurologic

involvement from the tumor itself.2 These symptoms include severe headache
and depressed mental status from displacement of the brain. Caution should be
applied in patients with newly discovered brain tumors awaiting tissue diagnosis
when central nervous system lymphoma is a consideration; although uncommon,
early initiation of corticosteroids can result in nondiagnostic biopsy specimens
due to tumor necrosis.3 Outside of central nervous system lymphoma treatment,
early initiation of corticosteroids has not been proven to alter the clinical
course of brain tumors.
Corticosteroids have not demonstrated efficacy in the treatment of cytotoxic
cerebral edema, and routine use of dexamethasone was not recommended in
the latest American Heart Association (AHA) guideline for the management
of cerebral and cerebellar infarction with swelling.1 Similarly, routine use of
corticosteroids is not recommended for cerebral edema related to spontaneous
ICH.4 For patients with TBI, the CRASH (Corticosteroid Randomisation After
Significant Head Injury) trial was conducted to compare mortality rates of
patients treated with 48 hours of methylprednisolone compared to placebo.
This clinical trial showed significantly increased mortality among patients with
severe TBI who were treated with corticosteroids.5 The 2017 Brain Trauma
Foundation guidelines state, “In patients with severe TBI, high-dose
methylprednisolone was associated with increased mortality and is
contraindicated.”6
OSMOTIC AGENTS. The mainstays of osmotic therapy in the treatment of cerebral

edema in neurocritical care are mannitol and hypertonic saline. Hypertonic
saline can be administered in several different concentrated solutions depending
on institutional practices, ranging from 2% to 23.4%; 3% saline is commonly
administered in 250 mL to 500 mL boluses either on an as-needed or standing
basis (eg, 250 mL every 6 hours), while 23.4% is typically administered as a
30 mL bolus over 10 to 15 minutes. Faster administration of hypertonic saline
may cause hypotension. A central line is required for administration of
hypertonic saline more concentrated than 3% to avoid peripheral vascular injury.
Hypertonic saline can also be administered as a continuous infusion, particularly
in patients who have cerebral edema in the setting of serum hyponatremia.
Most neurointensivists advise against increasing the serum sodium level above
160 mmol/L, and the safety and efficacy of iatrogenic hypernatremia beyond this
value is not well studied.7 Frequent administration of hypertonic saline may
result in hyperchloremic metabolic acidosis, which has been associated with
higher mortality in patients with ICH.8 Buffering hypertonic saline with acetate
may reduce the risk of metabolic acidosis in this circumstance.
Mannitol is a potent osmotic diuretic most commonly delivered as a 20%
concentrated solution in a dose range of 0.5 g/kg IV to 2 g/kg IV either as a scheduled
or as-needed bolus. Administration of mannitol requires use of an in-line filter
because mannitol may crystallize, especially at lower storage temperatures. For
this reason, mannitol administration is not recommended through temperature
exchange catheters. Continuous infusion of mannitol is not recommended
because mannitol may permeate across a disrupted blood-brain barrier and
cause rebound cerebral edema.9 Most neurointensivists recommend routine
monitoring of serum osmolality in patients being treated with frequent doses of
mannitol with avoidance of routinely increasing the serum osmolality to more
1590 DECEMBER 2018

than 320 mOsm/kg or an osmolar gap of more than 20 mOsm/kg. The osmolar KEY POINTS
gap is the difference between the measured and calculated osmolality, where
● Corticosteroids are
osmolality is calculated as: ineffective for the treatment
of cytotoxic edema and are
2ðNaÞ þ blood urea nitrogen=2:8 þ glucose=18
contraindicated in the
However, observational studies have shown a low incidence of deleterious treatment of patients with
severe traumatic brain
effects, including acute kidney injury, in patients with inadvertent elevation in
injury.
osmolality to 340 mOsm/kg.10
The choice of osmotic agent for a particular patient should be tailored to ● Frequent administration
volume status, serum sodium concentration, and other patient-specific factors. of hypertonic saline may
As a rule of thumb, hypertonic saline is preferred in patients who would benefit cause hyperchloremic
metabolic acidosis, which is
from volume expansion (eg, patients with the combination of hypovolemic associated with higher
shock and cerebral edema), while mannitol is preferred in patients who would mortality in neurocritical
benefit from the diuretic effect. care. Buffering hypertonic
The evidence for effectiveness of osmotic therapy in patients with cytotoxic solutions with acetate may
lower the chance of
edema from ischemic stroke is relatively weak.11 Because cytotoxic edema is developing hyperchloremic
caused by disruption of cell membrane and blood-brain barrier integrity, osmotic metabolic acidosis.
agents can permeate into infarcted brain tissue rather than remaining within the
intravascular compartment. Osmotic agents may accumulate over time, resulting ● Serum osmolality should
be monitored in patients
in rebound cerebral edema. Nevertheless, mannitol and hypertonic saline are
treated with frequent
often used as a temporizing measure in both ischemic stroke and ICH with mass doses of mannitol. Mannitol
effect as a bridge to more definitive surgical management. For patients near the should be held when serum
period of peak edema, scheduled doses of mannitol or hypertonic saline alone osmolality exceeds
320 mOsm/kg to
may provide enough time for spontaneous resolution of cytotoxic edema. In the
340 mOsm/kg or when
AHA guideline for cerebral and cerebellar infarction with swelling, there was the osmolar gap exceeds
Class IIa, Level C evidence that osmotic therapy is reasonable for patients with 20 mOsm/kg.
stroke with clinical deterioration due to cerebral edema.1 Routine use of osmotic
agents in patients with stroke or ICH without clinical deterioration due to ● Osmotic agents can be
used as a temporizing
cerebral edema is not indicated. measure to treat mass effect
from cytotoxic edema
INTRACRANIAL PRESSURE ELEVATION related to stroke and
The Monro-Kellie doctrine states that the intracranial compartment contains a intracerebral hemorrhage,
but evidence for efficacy
fixed total volume determined by the rigid skull. The intracranial volume is is weak. These lesions
determined by the relative volume of three primary compartments: blood, brain, may require surgical
and CSF. A transient increase in volume of one of these compartments results in decompression.
a transient rise in ICP that is subsequently buffered by displacement of one of
the other compartments. In normal physiology, CSF is the lowest pressure
compartment and acts as the primary buffer for expanding space-occupying
lesions. This concept is best evidenced by displacement of CSF from the
subarachnoid space and intraventricular compartment with an enlarging brain
mass. The relationship between ICP and intracranial volume is described by the
property of compliance. In conditions leading to poor brain compliance, small
changes in intracranial volume result in relatively large changes in ICP.
In normal physiology, as CSF is eluted from the choroid plexus into the
ventricle, a transient and measurable rise in ICP occurs, which is subsequently
buffered by displacement of CSF from the subarachnoid and ventricular
compartments. This transient ICP elevation and buffering results in the
characteristic ICP waveform, which is composed of the percussion wave
(P1, cardiac systole), tidal wave (P2, brain parenchymal displacement restricted
by the dura), and the dicrotic wave (P3, closure of the aortic valve).

As demonstrated in FIGURE 1-1,

the CSF waveform can be used
as a subjective indicator of
intracranial compliance. In the
setting of global cerebral edema,
in which the CSF buffer is
displaced from the intracranial
compartment and the brain
FIGURE 1-1 parenchyma abuts the rigid dura,
Intracranial pressure waveform. Graphs shows further CSF production in the
intracranial pressure waveform contours in normal ventricle results in exaggeration
physiology (A) and poor intracranial compliance
of the P2 waveform relative to
(B). Note the relative elevation of the P2 wave in B,
representing loss of CSF buffering with diffuse P1.12 This finding indicates poor
brain edema. intracranial compliance and
suggests caution should be
exercised to ensure that even
relatively small changes in intracranial volume are avoided. In a state of poor
compliance, physiologic changes that could be expected to increase intracranial
volume include: flat head positioning, obstruction of venous return from the
external jugular veins (restrictive tape or cervical collar, internal jugular vein
catheter placement), hypercarbia, hyperemia, fever, pain and agitation, and
increased intrathoracic or intraabdominal compartment pressure.
Slower oscillations in ICP also occur during normal and pathologic states.
These Lundberg waves can be tracked by changing the time scale of physiologic
ICP monitoring in order to evaluate trends over minutes to hours.13 Lundberg C
waves can be seen in normal physiology and most likely represent interactions
between the cardiac and respiratory cycles. These oscillations occur 4 to 8 times
per minute and generally do not exceed 25 mm Hg. Lundberg B waves are an
indicator of poor intracranial compliance. These oscillations occur 0.5 to 2 times
per minute and generally do not exceed 30 mm Hg. Lundberg A waves are
also called “plateau waves.” These waves are always pathologic and may be a
harbinger of cerebral herniation. Lundberg A waves represent steep increases
in ICP and may be as high as 40 mm Hg to 50 mm Hg and last for 5 to 10 minutes.
Implicit in the discussion of Lundberg waves is that these spontaneous
oscillations in ICP are self-limited and do not necessarily require urgent
treatment. Although Lundberg A waves may be an indicator of impending
cerebral herniation, each ICP plateau does not require treatment per se.14
Well-designed ICP treatment algorithms reserve interventions for sustained
elevation of ICP that lasts more than 10 to 15 minutes. ICP treatment strategies
that require an intervention every time the ICP exceeds 20 mm Hg will likely
result in overtreatment of spontaneous ICP oscillations that would otherwise be
self-limited.
Cerebral Perfusion Pressure

Another important principle of ICP management is the relationship between
ICP and cerebral perfusion pressure (CPP). CPP is measured as the difference
between the mean arterial pressure (MAP) and the ICP (in mm Hg), which
determines the pressure gradient of cerebral perfusion as a global measure. The
normal range of CPP in adults is 50 mm Hg to 70 mm Hg, but these values can be
impacted by chronic hypertension, hydrocephalus, and other conditions. In the
1592 DECEMBER 2018

cerebral autoregulation graph KEY POINTS
depicted in FIGURE 1-2, cerebral
● In states of poor
blood flow is tightly maintained intracranial compliance due
along a wide range of CPP to global cerebral edema,
through regulation of cerebral small changes in intracranial
arterial vasoconstriction and volume related to flat head
position, hyperemia,
vasodilation. This physiologic
hypercarbia, fever, or pain
response creates a complex may result in exaggerated
relationship between ICP, increases in intracranial
cerebral perfusion, and pressure.
intravascular volume.
FIGURE 1-2 ● Spontaneous oscillations
In states of cerebral Cerebral autoregulation curve. Graph shows in intracranial pressure
hypoperfusion, the normal cerebral autoregulation curve demonstrating called Lundberg A and
physiologic response is cerebral relatively constant cerebral blood flow within a B waves may cause
wide range of systemic mean arterial pressures. At self-limited increases in
vasodilation. In patients
lower blood pressure, cerebral blood flow is intracranial pressure that
with elevated ICP and poor maintained by cerebral vasodilation, whereas last several minutes. These
intracranial compliance, cerebral cerebral vasoconstriction mitigates against high oscillations are an indicator
vasodilation increases the volume blood pressure–induced hyperperfusion. In this of poor intracranial
of intracranial blood (most of theoretical curve, constant cerebral blood flow is compliance, but they
maintained between mean arterial pressures of typically resolve
which is in the venous 50 mm Hg and 150 mm Hg. spontaneously without
compartment), further treatment.
increasing ICP, which can
paradoxically worsen cerebral ischemia by reducing CPP. In this situation, ● Augmentation of cerebral
perfusion pressure with
treatment with systemic vasopressors may be needed to augment the CPP to
systemic vasopressors may
supraphysiologic values to allow subsequent cerebral vasoconstriction with lower intracranial pressure
reduction of intracranial blood volume and ICP.15 On the other hand, in brain by causing reflex cerebral
regions with a disrupted blood-brain barrier, excessive CPP may further vasoconstriction, thereby
contribute to vasogenic edema, which will increase the parenchymal component lowering the intracranial
volume of blood. Excessive
of intracranial volume and further raise the ICP. Management of ICP in patients cerebral perfusion,
with regional hypoperfusion and cerebral edema, typical of patients with severe however, may contribute to
TBI, may require a trial-and-error process to determine whether hemodynamic vasogenic edema in regions
augmentation with vasopressors results in ICP reduction or ICP elevation. with a disrupted blood-brain
barrier.
Treatment of Elevated Intracranial Pressure

The first step in ICP treatment should be to determine whether ICP-based
treatment is actually justified based on the etiology of the neurologic injury. For
patients with focal cerebral edema or compression of brain structures from a
space-occupying lesion, ICP monitoring may not be indicated in the first place.
Current guidelines do not recommend routine ICP monitoring or ICP-based
treatment algorithms in patients with spontaneous ICH, brain tumors, meningitis,
or ischemic stroke. ICP monitoring and treatment continues to be recommended
in patients with severe TBI (initial Glasgow Coma Scale score of ≤8).6 Implicit in
this recommendation, however, is that patients with TBI who are noncomatose
do not necessarily require ICP monitoring, and ICP-based treatment may be
unhelpful or detrimental in these patients. In neurocritical care, this situation
commonly arises when patients with TBI who are initially comatose are
subsequently recovering and develop periods of intracranial hypertension
associated with pain, agitation, or other causes. In this situation, the risk of
overtreatment is high, and providers should consider removing the ICP monitor.
Current Brain Trauma Foundation guidelines recommend maintaining an ICP of

22 mm Hg or less and a CPP of at least 50 mm Hg to 60 mm Hg.6 The recent

BEST:TRIP (Benchmark Evidence From South American Trials: Treatment of
Intracranial Pressure) clinical trial, however, failed to show improved outcomes
in patients with severe TBI treated with an ICP-based protocol compared to
patients treated with a protocol based on imaging and clinical findings without
ICP monitoring.16
Best practice for ICP-based management is the creation and institution of a
multimodality treatment algorithm that reinforces consistent, evidence-based
practices for neurocritical care nurses and physicians. Although no single
algorithm can account for every patient, ICP-based treatment algorithms
improve the consistency of care, unburden physicians and nurses from focusing
on individual treatment decisions, and improve patient outcomes.17 Institutional
treatment protocols should be formulated by a multidisciplinary team that
includes nurses, intensive care providers, neurologists, and neurosurgeons.
These protocols should be reviewed and revised annually and on an ad hoc basis.
Most ICP-based treatment protocols include first-line therapies (noninvasive
maneuvers such as repositioning, ventilator changes, sedation, analgesia),
second-line therapies (osmotic agents, hyperventilation, CSF diversion), and
third-line therapies (metabolic suppression with anesthetic agents, induced
hypothermia, surgical decompression).18 See FIGURE 1-3 for the Emergency
Neurological Life Support (ENLS) ICP treatment algorithm.
ANESTHETIC AGENTS AND METABOLIC SUPPRESSION. Although propofol and

high-dose benzodiazepine infusions can be used for metabolic suppression,
the mainstay for induction of “pharmacologic coma” in refractory intracranial
hypertension is pentobarbital. Although strong evidence indicates that
pentobarbital lowers ICP, there is a paucity of high-quality evidence for
improvement of neurologic outcomes.19 Pentobarbital is typically delivered
as a 20 mg/kg IV bolus followed by continuous infusion at 0.5 mg/kg/h to
5 mg/kg/h. Although the primary titration parameter should be targeted to ICP
control, EEG monitoring with at least a limited electrode montage should also
be undertaken to trend the suppression ratio. Pentobarbital is typically titrated
to achieve burst-suppression anesthesia with a suppression ratio of 0.8 to
0.9, where suppression ratio is the ratio of the duration of EEG suppression
to the duration of EEG bursts. Suppression ratio may be monitored with a
commercially available quantitative EEG product typically used to measure
depth of anesthesia that also displays raw EEG on bedside monitors.
Pentobarbital includes propylene glycol, which can accumulate to cause fatal
lactic acidosis with elevated osmolar gap, acute kidney failure, and circulatory
collapse, so the osmolar gap should be monitored in patients being treated with
pentobarbital.20 Other toxicities include suppression of gastrointestinal motility,
immunosuppression, bone marrow suppression, and distributive shock. The
half-life of pentobarbital ranges from 15 to 50 hours, so patients may not recover
consciousness for several days after medication discontinuation, and caution
should be exercised in brain death declaration for patients who were treated with
pentobarbital. Very high-dose pentobarbital can mimic every clinical sign of
brain death, including bilateral fixed and dilated pupils and diabetes insipidus.
After prolonged use of pentobarbital, rapid discontinuation may produce
withdrawal symptoms including seizures and rebound ICP elevation, so
phenobarbital can be administered and tapered during pentobarbital weaning.
1594 DECEMBER 2018

INDUCED HYPOTHERMIA. Ample KEY POINTS
evidence indicates induced
● Current Brain Trauma
hypothermia lowers ICP, but Foundation guidelines
the evidence that it improves recommend maintaining an
outcomes is lacking. The recent intracranial pressure of
Eurotherm3235 trial tested 22 mm Hg or less and a
cerebral perfusion pressure
whether an ICP treatment
of at least 50 mm Hg to
protocol that included early 60 mm Hg in patients with
induction of hypothermia severe traumatic brain
resulted in improved outcomes injury.
and reduced the need for other
● A recent clinical trial of an
third-line therapies in patients intracranial pressure–based
with severe TBI.21 Although treatment protocol for
fewer ICP-directed severe traumatic brain injury
interventions were required in compared to a treatment
protocol based on clinical
the hypothermia group, examination and imaging
neurologic outcomes were FIGURE 1-3 without intracranial
actually worse. This clinical trial Intracranial pressure treatment algorithm. pressure monitoring showed
Emergency Neurological Life Support (ENLS) no difference in outcomes
demonstrated that early treatment protocol for elevated intracranial between the two groups.
induced hypothermia does not pressure demonstrating an algorithmic,
improve outcomes in TBI as a multitiered approach to escalating treatment ● Pentobarbital infusion for
neuroprotective strategy, but modalities. Third-line treatments include intracranial pressure
anesthetic coma, induced hypothermia, and reduction can result in
hypothermia continues to be decompressive craniectomy. severe medication side
employed as a third-line CSF = cerebrospinal fluid; CT = computed tomography; effects such as propylene
therapy for refractory ICP ICP = intracranial pressure; MAP = mean arterial pressure;
glycol toxicity,
NaCl = sodium chloride; PaCO2 = partial pressure of
elevation. Most ICP treatment immunosuppression,
carbon dioxide, arterial.
protocols that include a
This example of an approach to treating intracranial
impaired gastrointestinal
hypothermia target a core motility, and distributive
pressure elevation includes hypertonic saline (denoted
NaCl in this figure) in tier 2 as a second-line therapy for
shock.
temperature of 32°C to 34°C
patients who were already treated with mannitol in tier 1. In
(89.6°F to 93.2°F) using either some cases, however, patients may have been treated ● A recent clinical trial of
surface cooling or intravascular with hypertonic saline instead of mannitol in tier 1. This is early induced hypothermia
cooling devices with a often the case for patients with low intravascular volume. as a neuroprotective
Modified with permission from Stevens RD, et al, Neurocrit strategy in severe traumatic
continuous automated feedback Care.18 © 2015 Springer. brain injury showed
mechanism. Successful reduction of intracranial
induction and maintenance of pressure–directed
hypothermia, especially in interventions, but
neurologic outcomes were
patients not concurrently treated with anesthetic agents, requires institution
worse in patients treated
of a multimodality shivering protocol.22 Shivering causes failure of with hypothermia.
temperature maintenance and paradoxical ICP elevation from a combination
of cerebral hypermetabolism and systemic hypercarbia. Antishivering ● Induced hypothermia
protocols typically include a combination of surface counterwarming, continues to be a useful
third-line intervention for
magnesium infusion, buspirone, meperidine, and sedative infusion.23 During refractory intracranial
the induction phase, hypothermia can produce peripheral vasoconstriction pressure elevation, but an
leading to skin ischemia, severe hypokalemia, and diuresis from shunting effective hypothermia
blood flow to the kidneys. Electrolytes should be monitored frequently during protocol that includes
multimodality treatment of
the hypothermia induction and rewarming phases.24 Rewarming should occur shivering is required.
in a tightly controlled fashion to avoid severe rebound hyperkalemia and
distributive shock from peripheral vasodilation. For this reason, rewarming
should be undertaken at a rate of 0.1°C (32.2°F) per hour or slower in patients
who have been cooled for long intervals.

DECOMPRESSIVE SURGERY. For patients with ICP elevation due to obstructive

hydrocephalus, placement of an EVD for CSF diversion is considered a first-line
treatment. For patients with elevated ICP due to a focal compressive brain lesion,
decompressive surgery is considered a first-line treatment rather than global
ICP-directed treatments (CASE 1-1). For patients with elevated ICP due to global
or multifocal brain injuries such as severe TBI, however, decompressive
craniectomy should be considered as a third-line treatment for refractory ICP
elevation. The appropriate sequence of ICP treatment protocols has come into
better focus with the recent DECRA (Decompressive Craniectomy in Diffuse
Traumatic Brain Injury) and RESCUEicp (Randomised Evaluation of Surgery
With Craniectomy for Uncontrollable Elevation of Intracranial Pressure) clinical
CASE 1-1 A 47-year-old man presented to the neurocritical care unit because of a
gunshot wound to the left occipital lobe that was initially managed with
surgical debridement and placement of an intracranial pressure (ICP)
monitor. The patient initially had purposeful movements of the left side,
but he had aphasia and right hemiplegia.
On hospital day 3, he developed
refractory ICP elevation that was
treated with sedation and analgesia
followed by hypertonic saline boluses
according to the institutional ICP
management protocol. Despite these
interventions, he developed a
sustained ICP of more than 30 mm Hg
and a worsening neurologic
examination with extensor posturing.
Repeat head CT demonstrated an
evolving infarct of the left occipital
lobe with hemorrhagic transformation
and 8-mm midline shift. He was
treated with mannitol 1 g/kg IV as a
FIGURE 1-4 temporizing measure and then was
Head CT of the patient in CASE 1-1.
taken for decompressive
CT shows decompressive craniectomy
to allow room for the hemorrhagic craniectomy. After surgery, the ICP
infarction to swell outside of the normalized, and he began to have
skull as a treatment for refractory purposeful movements of the left side
intracranial pressure elevation and again. The postoperative head CT is
brain compression.
shown (FIGURE 1-4).
COMMENT This case illustrates the importance of tailoring interventions for ICP
elevation to the underlying pathophysiology. In this case, rather than
escalating to third-line medical management of ICP elevation with
pentobarbital coma or induced hypothermia, the team recognized that ICP
elevation was caused by an enlarging focal brain lesion that required
surgical decompression.
1596 DECEMBER 2018

trials.25,26 DECRA tested whether early bifrontal decompressive craniectomy KEY POINTS
would improve outcomes and lessen the need for other ICP-directed
● A recent clinical trial
interventions in patients with severe TBI.25 Although surgical decompression demonstrated that
lowered ICP, patients had worse neurologic outcomes in the early surgery arm. decompressive craniectomy
The RESCUEicp trial randomly selected patients with severe TBI with refractory for refractory intracranial
ICP elevation to undergo decompressive craniectomy (unilateral or bilateral) pressure elevation in severe
traumatic brain injury
as a third-line treatment or continued medical management with anesthetic
improves survival and
agents.26 This clinical trial demonstrated modest benefits to decompressive reduces the chance of
craniectomy with higher odds of survival and lower odds of severe disability, but severe disability, but more
there was no effect on the odds of good recovery, and more patients survived patients survived in a
vegetative state compared
in the vegetative state in the surgical arm. These study results clarify that
to medical management
decompressive surgery should be reserved for third-line management of alone.
refractory ICP elevation, and careful discussion of expected outcomes should be
undertaken with surrogate decision makers, including the higher chance of ● With transtentorial
survival in a vegetative state. herniation, the pupil dilation
is ipsilateral to the mass
lesion, but hemiplegia may
CEREBRAL HERNIATION SYNDROMES be contralateral because of
Cerebral herniation occurs when brain structures are displaced into an adjacent direct corticospinal tract
compartment with compression of surrounding neurologic structures. Cerebral involvement or ipsilateral
because of compression of
herniation typically occurs when enlarging space-occupying lesions result in the contralateral cerebral
displacement or expansion of brain tissue. Herniation is often but not universally peduncle against the
associated with elevated ICP. In some cases, global ICP values are normal, but tentorial edge (Kernohan
compartmentalized ICP elevation contributes to pressure gradients that displace notch phenomenon).
brain tissue. For this reason, global ICP monitoring is not always a sensitive
indicator of impending cerebral herniation. This is especially true of posterior
fossa lesions, which may result in upward herniation with little change in
ICP. Clinical monitoring for impending herniation relies on serial
neurologic examinations.
Transtentorial (Uncal) Herniation

Transtentorial herniation occurs when an expanding supratentorial mass lesion
displaces the medial temporal lobe (uncus) through the tentorial incisura with
compression of the ipsilateral oculomotor nerve followed by the midbrain. The
clinical hallmark of transtentorial herniation is ipsilateral dilation of the pupil
and, less commonly, deviation of the eye laterally and inferiorly. As herniation
progresses, patients develop extensor posturing and depressed mental status due
to dysfunction of the corticospinal tract and reticular activating system in the
midbrain. Bilateral transtentorial herniation, also known as central herniation,
typically occurs with global cerebral edema and presents with coma, extensor
posturing, and bilateral fixed and dilated pupils. With expanding extraaxial
lesions (typically subdural hematoma), a characteristic pattern of ipsilateral
pupil dilation and ipsilateral hemiplegia may develop. This so-called Kernohan
notch phenomenon occurs with lateral displacement of the midbrain such that
the contralateral cerebral peduncle is compressed against the tentorial edge at
Kernohan notch with relative sparing of the ipsilateral cerebral peduncle.27 For
patients with supratentorial space-occupying lesions, the side of the pupil
dilation is ipsilateral to the source of mass effect, while the side of hemiplegia
may be either ipsilateral or contralateral (ie, “the pupil doesn’t lie”).
Pupil dilation is the sine qua non manifestation of transtentorial herniation,
so clinical monitoring for pupillary abnormalities is the hallmark of bedside

neuromonitoring in patients with supratentorial mass lesions at risk for herniation.

ICP elevation is often a late finding. Prior to dilation of the pupil, however,
the velocity of pupillary constriction diminishes with early compression of the
oculomotor nerve. This can be monitored with serial bedside pupillometry with
quantitative measurement of the pupillary constriction velocity using commercially
available bedside monitors.28 Observational studies have demonstrated that
the diminution of the constriction velocity may precede frank clinical signs of
transtentorial herniation by hours.29 In some cases, early recognition of
impending herniation may provide time for temporizing medical interventions
and definitive surgical decompression.
Transtentorial herniation is caused by focal displacement and compression
of brain structures rather than global elevation of ICP or cerebral edema.
For this reason, therapies directed purely at reduction of cerebral edema
(eg, corticosteroids) or global reduction in ICP (eg, anesthetic agents,
hypothermia) are ineffective in reversing clinical signs of transtentorial herniation.
Without rapid treatment to reverse clinical signs of transtentorial herniation, the
natural history is progression to central herniation, which is nearly universally
fatal. Although observational studies have demonstrated restoration of the
pupillary light reflex—and in some cases survival—after treatment with
hypertonic saline and mannitol, radiographic evidence of reversal of herniation
is lacking.30,31 For this reason, osmotic therapy and hyperventilation should be
considered temporizing measures to reverse clinical evidence of transtentorial
herniation prior to definitive surgical decompression. Decompressive surgery
requires either excision of the space-occupying lesion (extraaxial hematoma or
temporal lobectomy) or decompressive craniectomy.
Transtentorial herniation used to be considered a universally fatal event.
Depending on the etiology of herniation and availability of definitive surgical
treatment, patients with unilateral transtentorial herniation have been reported
to survive and recover independent function in some series.32 Even if clinical
signs of herniation are reversed with medical or surgical interventions, three
common sequelae of the herniation event may occur. Compression of venous
drainage from the central midbrain and pons may lead to venous congestion and
hemorrhagic infarction within the medial brainstem structures, which is termed
Duret hemorrhage. Compression of the contralateral posterior cerebral artery
against the tentorial edge may cause posterior cerebral artery territory stroke.33
Compression of the ipsilateral anterior cerebral artery against the inferior edge of
the falx cerebri due to associated subfalcine herniation may cause anterior
cerebral artery territory stroke.
Posterior Fossa Herniation Syndromes

Expanding posterior fossa lesions such as cerebellar stroke can result in
compression of the fourth ventricle with acute obstructive hydrocephalus,
compression of the brainstem and cranial nerves, and two distinct
herniation syndromes.
Cerebellar tonsillar herniation occurs with downward displacement of the
cerebellar tonsils through the foramen magnum with compression of the
cervicomedullary junction. This herniation syndrome may not be accompanied
by elevated ICP or pupillary changes until obstructive hydrocephalus develops
as a late event. For this reason, tonsillar herniation is commonly missed
with neuromonitoring that is focused on ICP and pupillary changes. Tonsillar
1598 DECEMBER 2018

herniation typically causes quadriparesis due to compression of the pyramidal KEY POINTS
decussation, respiratory insufficiency due to bulbar dysfunction and
● Serial bedside
compression of the upper cervical spinal cord, and depressed level of quantitative pupillometry
consciousness.34 As with transtentorial herniation, osmotic therapy and may detect reduction in
hyperventilation are temporizing measures, but definitive treatment requires the pupillary constriction
surgical decompression, either decompressive suboccipital craniectomy, partial velocity hours prior to
frank clinical signs of
cerebellectomy, or hematoma evacuation.1,35
transtentorial herniation,
Upward herniation occurs with upward displacement of the cerebellum which may afford time for
through the tentorial incisura with compression of the dorsal midbrain. This preemptive treatment.
uncommon herniation syndrome typically occurs in patients with expanding
posterior fossa mass lesions in combination with an EVD. Excessive CSF ● Although patients can
survive and recover from
diversion from the lateral ventricle creates a pressure gradient resulting in transtentorial herniation in
upward displacement of the cerebellum.36 Patients typically present with some cases, the sequelae of
new-onset vertical gaze palsy and depressed level of consciousness. Treatment herniation can include Duret
requires clamping the EVD in addition to treatment with osmotic agents and brainstem hemorrhage,
ipsilateral anterior cerebral
hyperventilation. As with tonsillar herniation, upward herniation is not artery stroke, and
commonly associated with pupil dilation. Because of the mechanism of contralateral posterior
herniation, the ICP is typically normal or low rather than elevated.37 For these cerebral artery stroke.
reasons, diagnosis of upward herniation is often delayed or missed (CASE 1-2).
● Posterior fossa mass
Because of the risk of upward herniation, current AHA guidelines recommend
lesions can cause cerebellar
against placement of an EVD as the primary intervention for obstructive tonsillar herniation or
hydrocephalus caused by cerebellar strokes.1 The current recommendation is to upward herniation of the
consider decompressive surgery prior to EVD insertion in this setting.38 cerebellum through the
tentorial incisura, which may
not be accompanied by
intracranial pressure
CONCLUSION elevation or pupillary
Cerebral edema, elevated ICP, and cerebral herniation syndromes are a major changes.
cause of morbidity and mortality in neurocritical care. Although these conditions
frequently overlap, they are distinct pathophysiologic entities that require
tailored neurologic monitoring and treatment. For cerebral edema, both the
mechanism and context are important to determine best practices for treatment.
While vasogenic edema surrounding brain tumors may respond to
corticosteroids, this same treatment strategy is contraindicated in treatment of
cerebral edema from TBI. Our understanding of ICP as a global measure and
treatment target is changing. Management strategies are typically directed to
global reduction of ICP, including osmotic therapies, metabolic suppression, and
hypothermia, whereas in many focal neurologic conditions, ICP can be
compartmentalized and contribute to displacement of brain structures. In these
situations, surgical decompression should be the mainstay of treatment rather
than global ICP reduction strategies.
Although many observational studies in TBI have shown that sustained ICP
elevation is associated with poor outcomes, a recent clinical trial showed that
ICP-targeted management resulted in outcomes that were no better than
management based on imaging and clinical examination, a result that could
almost be considered heretical to the last few decades of TBI management. In
addition, improved understanding of the physiology of ICP-volume relationships
and cerebral autoregulation have demonstrated that acceptable ICP thresholds
at the level of an individual patient are not as clear-cut as we imagined. ICP
thresholds may be best understood in the context of accompanying brain
perfusion and tissue oxygenation.

CASE 1-2 A 55-year-old man presented to the emergency department after he fell
down a flight of stairs, during which he struck his head on the pavement
and experienced brief loss of consciousness. The initial head CT showed
convexity subarachnoid hemorrhage, small bifrontal cerebral contusions,
and a 20-mL contusion of the cerebellar vermis. He was admitted to the
neurocritical care unit.
On hospital day 2, he became more lethargic with new onset of
quadriparesis and acute respiratory insufficiency requiring intubation.
Repeat head CT (FIGURE 1-5) demonstrated increasing brainstem
compression, cerebellar tonsillar herniation, and acute obstructive
hydrocephalus. The
neurosurgeon placed an
external ventricular drain
(EVD), but the patient had
no significant clinical
improvement after CSF
diversion. The patient was
subsequently noted to be
obtunded with vertical
ophthalmoplegia characterized
by bilateral inferior-medial
eye deviation (“sunset eyes”).
Suspecting upward herniation,
the neurocritical care team
FIGURE 1-5 clamped the EVD, and the
Head CT of the patient in CASE 1-2. CT shows patient was taken for
compression of the brainstem and cerebellar decompressive suboccipital
tonsillar herniation in a patient with cerebellar craniectomy. He had gradual
vermis contusion and frontal subarachnoid
hemorrhage. The patient also has evidence of mild neurologic improvement,
hydrocephalus. After an external ventricular drain and the EVD was removed
placement, he developed upward herniation. during the subsequent
hospital course.
COMMENT This case demonstrates the clinical presentation and treatment of posterior
fossa herniation syndrome, which is distinct from the more commonly
recognized transtentorial herniation syndrome. Because upward herniation
is uncommon and the progression is insidious, signs are often overlooked.
Current American Heart Association guidelines recommend surgical
decompression prior to or concomitant with placement of an EVD for
posterior fossa lesions with mass effect in order to lower the chances of
upward herniation.1 This case illustrates the potential to worsen upward
herniation when CSF diversion is undertaken without surgical decompression.
1600 DECEMBER 2018

Monitoring for and treating cerebral herniation syndromes have also
evolved over time. What was once considered a uniformly fatal event may now
be reversible in the right clinical context. Routine bedside pupillometry in the
neurocritical care setting has demonstrated the ability to detect impending
transtentorial herniation and may buy time for definitive surgical management
in some cases.
REFERENCES
1 Wijdicks EF, Sheth KN, Carter BS, et al. 10 Gondim Fde A, Aiyagari V, Shackleford A,
Recommendations for the management of Diringer MN. Osmolality not predictive of
cerebral and cerebellar infarction with swelling: a mannitol-induced acute renal insufficiency.
statement for healthcare professionals from the J Neurosurg 2005;103(3):444–447. doi:10.3171/
American Heart Association/American Stroke jns.2005.103.3.0444.
Association. Stroke 2014;45(4):1222–1238.
11 Misra UK, Kalita J, Ranjan P, Mandal SK. Mannitol
doi:10.1161/01.str.0000441965.15164.d6.
in intracerebral hemorrhage: a randomized
2 Ryken TC, McDermott M, Robinson PD, et al. The controlled study. J Neurol Sci 2005;234(1):41–45.
role of steroids in the management of brain doi:10.1016/j.jns.2005.03.038.
metastases: a systematic review and evidence-
12 Oshio K, Onodera H, Uchida M, et al. Assessment
based clinical practice guideline. J Neurooncol
of brain compliance using ICP waveform analysis
2010;96(1):103–114. doi:10.1007/s11060-009-0057-4.
in water intoxication rat model. Acta Neurochir
3 Binnahil M, Au K, Lu JQ, et al. The influence of Suppl 2013;118:219–221. doi:10.1007/978-3-7091-
corticosteroids on diagnostic accuracy of biopsy 1434-6_41.
for primary central nervous system lymphoma.
13 Kasprowicz M, Lalou DA, Czosnyka M, et al.
Can J Neurol Sci 2016;43(5):721–725. doi:10.1017/
Intracranial pressure, its components and
cjn.2016.255.
cerebrospinal fluid pressure-volume
4 Hemphill JC 3rd, Greenberg SM, Anderson CS, compensation. Acta Neurol Scand 2016;134(3):
et al. Guidelines for the management of 168–180. doi:10.1111/ane.12541.
spontaneous intracerebral hemorrhage: a
14 Castellani G, Zweifel C, Kim DJ, et al. Plateau
guideline for healthcare professionals from the
waves in head injured patients requiring
American Heart Association/American Stroke
neurocritical care. Neurocrit Care 2009;11(2):
143–150. doi:10.1007/s12028-009-9235-7.
doi:10.1161/STR.0000000000000069.
15 Bohman LE, Heuer GG, Macyszyn L, et al. Medical
5 Roberts I, Yates D, Sandercock P, et al. Effect of
management of compromised brain oxygen in
intravenous corticosteroids on death within
patients with severe traumatic brain injury.
14 days in 10008 adults with clinically significant
Neurocrit Care 2011;14(3):361–369. doi:10.1007/
head injury (MRC CRASH trial): randomised
s12028-011-9526-7.
placebo-controlled trial. Lancet 2004;364(9442):
1321–1328. doi:10.1016/S0140-6736(04)17188-2. 16 Chesnut RM, Temkin N, Carney N, et al. A trial of
intracranial-pressure monitoring in traumatic
6 Carney N, Totten AM, O’Reilly C, et al. Guidelines
brain injury. N Engl J Med 2012;367(26):2471–2481.
for the management of severe traumatic brain
doi:10.1056/NEJMoa1207363.
injury, 4th edition. Neurosurgery 2017;1(80):6–15.
doi:10.1227/NEU.0000000000001432. 17 Clayton TJ, Nelson RJ, Manara AR. Reduction in
mortality from severe head injury following
7 Ryu JH, Kahle KT, Sheth SA, et al. Induced and
introduction of a protocol for intensive care
sustained hypernatremia for the prevention and
management. Br J Anaesth 2004;93(6):761–767.
treatment of cerebral edema following brain
doi:10.1093/bja/aeh249.
injury. Neurocrit Care 2013;19(2):222–231.
doi:10.1007/s12028-013-9824-3. 18 Stevens RD, Shoykhet M, Cadena R. Emergency
neurological life support: intracranial hypertension
8 Riha HM, Erdman MJ, Vandigo JE, et al. Impact of
and herniation. Neurocrit Care 2015;23(suppl 2):
moderate hyperchloremia on clinical outcomes
S76–S82. doi:10.1007/s12028-015-0168-z.
in intracerebral hemorrhage patients treated
with continuous infusion hypertonic saline: a 19 Roberts I, Sydenham E. Barbiturates for acute
pilot study. Crit Care Med 2017;45(9):e947–e953. traumatic brain injury. Cochrane Database Syst
doi:10.1097/CCM.0000000000002522. Rev 2012;12:CD000033. doi:10.1002/14651858.
CD000033.pub2.
9 Palma L, Bruni G, Fiaschi AI, Mariottini A. Passage
of mannitol into the brain around gliomas: a 20 Bledsoe KA, Kramer AH. Propylene glycol toxicity
potential cause of rebound phenomenon. A complicating use of barbiturate coma. Neurocrit
study of 21 patients. J Neurosurg Sci 2006;50(3): Care 2008;9(1):122–124. doi:10.1007/s12028-008-
63–66. 9065-z.

21 Andrews PJ, Sinclair HL, Rodriguez A, et al. 30 Qureshi AI, Suarez JI, Bhardwaj A, et al. Use of
Hypothermia for intracranial hypertension after hypertonic (3%) saline/acetate infusion in the
traumatic brain injury. N Engl J Med 2015;373(25): treatment of cerebral edema: effect on
2403–2412. doi:10.1056/NEJMoa1507581. intracranial pressure and lateral displacement of
the brain. Crit Care Med 1998;26(3):440–446.
22 Urbano LA, Oddo M. Therapeutic hypothermia for
traumatic brain injury. Curr Neurol Neurosci Rep 31 Koenig MA, Bryan M, Lewin JL 3rd, et al. Reversal
2012;12(5):580–591. doi:10.1007/s11910-012-0304-5. of transtentorial herniation with hypertonic
saline. Neurology 2008;70(13):1023–1029.
23 Choi HA, Ko SB, Presciutti M, et al. Prevention of
doi:10.1212/01.wnl.0000304042.05557.60.
shivering during therapeutic temperature
modulation: the Columbia anti-shivering 32 Ishizaka S, Shimizu T, Ryu N. Dramatic recovery
protocol. Neurocrit Care 2011;14(3):389–394. after severe descending transtentorial
doi:10.1007/s12028-010-9474-7. herniation-induced Duret haemorrhage: a case
report and review of literature. Brain Inj 2014;
24 Soeholm H, Kirkegaard H. Serum potassium
28(3):374–377. doi:10.3109/02699052.2013.864422.
changes during therapeutic hypothermia after
out-of-hospital cardiac arrest—should it be 33 Namura S, Kang Y, Matsuda I, Kamijyo Y.
treated? Ther Hypothermia Temp Manag 2012; Magnetic resonance imaging of sequelae of
2(1):30–36. doi:10.1089/ther.2012.0004. temporal lobe herniation secondary to traumatic
acute subdural hematoma: Kernohan’s notch and
25 Cooper DJ, Rosenfeld JV, Murray L, et al.
posterior cerebral artery territory infarctions
Decompressive craniectomy in diffuse traumatic
contralateral to the supratentorial lesion—case
brain injury. N Engl J Med 2011;364(16):1493–1502.
report. Neurol Med Chir (Tokyo) 1997;37(1):32–35.
doi:10.1056/NEJMoa1102077.
34 St Louis EK, Wijdicks EF, Li H. Predicting
26 Hutchinson PJ, Kolias AG, Timofeev IS, et al. Trial
neurologic deterioration in patients with
of decompressive craniectomy for traumatic
cerebellar hematomas. Neurology 1998;51(5):
intracranial hypertension. N Engl J Med 2016;
1364–1369. doi:10.1212/WNL.51.5.1364.
375(12):1119–1130. doi:10.1056/NEJMoa1605215.
35 Kirollos RW, Tyagi AK, Ross SA, et al. Management
27 Yoo WK, Kim DS, Jang SH, et al. Kernohan’s notch
of spontaneous cerebellar hematomas: a
phenomenon demonstrated by diffusion tensor
prospective treatment protocol. Neurosurgery
imaging and transcranial magnetic stimulation.
2001;49(6):1378–1386; discussion 1386–1387.
J Neurol Neurosurg Psychiatry 2008;79(11):
doi:10.1097/00006123-200112000-00015.
1295–1297. doi:10.1136/jnnp.2007.138131.
36 Kase CS, Wolf PA. Cerebellar infarction: upward
28 Zafar SF, Suarez JI. Automated pupillometer for
transtentorial herniation after ventriculostomy.
monitoring the critically ill patient: a critical
Stroke 1993;24(7):1096–1098.
appraisal. J Crit Care 2014;29(4):599–603.
doi:10.1016/j.jcrc.2014.01.012. 37 Wang CY, Chee CP, Delilkan AE. Upward
transtentorial herniation of posterior fossa
29 Manley GT, Larson MD. Infrared pupillometry
structures. Eur J Anaesthesiol 1991;8(6):469–470.
during uncal herniation. J Neurosurg Anesthesiol
2002;14(3):223–228. 38 Gurol ME, St Louis EK. Treatment of cerebellar
masses. Curr Treat Options Neurol 2008;10(2):
138–150.
1602 DECEMBER 2018

Intracerebral REVIEW ARTICLE
Hemorrhage

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
By Wendy C. Ziai, MD, MPH, FAHA, FNCS, FESO;
CITE AS:
ABSTRACT CONTINUUM (MINNEAP MINN)
PURPOSE OF REVIEW: This article describes the advances in the management of 2018;24(6, NEUROCRITICAL CARE):
spontaneous intracerebral hemorrhage in adults. 1603–1622.
RECENT FINDINGS: Therapeuticintervention in intracerebral hemorrhage has Dr J. Ricardo Carhuapoma, Johns
continued to focus on arresting hemorrhage expansion, with large Hopkins Hospital, 1800 Orleans
St, Phipps 455, Baltimore, MD
randomized controlled trials addressing the effectiveness of rapidly 21287, jcarhua1@jhmi.edu.
lowering blood pressure, hemostatic therapy with platelet transfusion,
and other clotting complexes and clot volume reduction both of RELATIONSHIP DISCLOSURE:
Dr Ziai serves as an associate
intraventricular and parenchymal hematomas using minimally invasive editor for Neurocritical Care, on
techniques. Smaller studies targeting perihematomal edema and an advisory board for C.R. Bard
inflammation may also show promise. Inc, and has received personal
compensation as a consultant
for and research/grant support
SUMMARY: The management of spontaneous intracerebral hemorrhage, from HeadSense Medical Ltd.
Dr Ziai has received grant
long relegated to the management and prevention of complications, is
support from the Johns Hopkins
undergoing a recent evolution in large part owing to stereotactically guided Anesthesiology and Critical Care
clot evacuation techniques that have been shown to be safe and that Medicine (ACCM) Stimulating
and Advancing ACCM Research
may potentially improve outcomes. (StAAR) program and from the
National Institutes of Health
(5U01NS062851, 1U01NS08082).
Dr Carhuapoma reports no
INTRODUCTION disclosure.
I
ntracerebral hemorrhage (ICH) remains a cause of significant morbidity UNLABELED USE OF
and mortality and is associated with severe long-term disability. PRODUCTS/INVESTIGATIONAL
USE DISCLOSURE:
Furthermore, it comprises 10% to 15% of all strokes, with an incidence of 24.6
Drs Ziai and Carhuapoma discuss
per 100,000 person-years and with a growing incidence related to the use the unlabeled/investigational
of anticoagulation, antiplatelet drugs, and an aging population.1 Despite this, use of alteplase for the
treatment of intracerebral
ICH is the last form of stroke without specific therapy. The financial impact of hemorrhage and discuss the use
ICH is, in part, owing to its high mortality, with up to one-half of patients dying of several devices, including
30 days after experiencing ICH, often despite extensive stays in the intensive ultrasound microcatheters,
sonothrombolysis for minimally
care unit.2 Moreover, those who survive are left with a high degree of disability.1 invasive subcortical
Treatment of ICH ranges from best medical therapy to approaches involving parafascicular transsulcal
several different surgical techniques, most of which are at different levels of access for clot evacuation
(MiSPACE), and the stereotactic
experimental state.1,3 A lack of definitive evidence-based recommendations to mechanical (suction/vibration/
guide the care of patients with ICH has led to significant heterogeneity in current aspiration) thrombolytic
technique for minimally invasive
clinical practice.1,3–5 evacuation of intracerebral
hemorrhage.
PATHOPHYSIOLOGY
ICH occurs after a parenchymal arteriole in the brain ruptures. Common © 2018 American Academy
processes leading to ICH include amyloid angiopathy, tumors, hemorrhagic of Neurology.

INTRACEREBRAL HEMORRHAGE
KEY POINTS transformation of an ischemic stroke, cerebral venous thrombosis, vasculitis, and
vascular malformations such as cavernous malformations, arteriovenous
● Treatment for
intracerebral hemorrhage
malformations, and ruptured saccular aneurysms. In the case of spontaneous
revolves around acute ICH, where preexisting hypertension is ubiquitous, Charcot-Bouchard
management of blood aneurysms, believed to result from lipohyalinosis of small arterioles, are often
pressure, intracranial blamed for the rupture of small penetrating blood vessels implicated in ICH
pressure, and cerebral
involving the cerebellum, pons, thalamus, and basal ganglia.6
edema.
The primary nature of ICH is determined as a diagnosis of exclusion based on a
● Although the majority thorough investigation for secondary structural causes of ICH. Advanced age,
of intracerebral hemorrhage deep location (basal ganglia, thalami, or posterior fossa), or history of
is attributed to hypertensive hypertension are often interpreted as surrogates of primary ICH, although
small penetrating vessel
arteriopathy, vascular cerebral angiography studies suggest that these imaging and clinical features are
imaging (CT angiography, not always reliable indicators, and patients with these features may have
magnetic resonance coexisting vascular abnormalities.7,8
angiography, or catheter
angiography) and MRI are
essential to rule out other
MEDICAL MANAGEMENT
etiologies of intracerebral Medical management is aimed at treating intracranial pressure (ICP), controlling
hemorrhage. systemic hypertension, and preventing hematoma expansion. With the
exception of early blood pressure control, no successful phase 3 trials have been
● The Intracerebral
conducted at the time of this writing that have proven to improve survival or
Hemorrhage Score and the
Functional Outcome in neurologic outcomes after ICH.9 Surgical management is focused on clot
Patients With Primary extraction, removal of intraventricular blood products, and managing
Intracerebral Hemorrhage intracranial hypertension.
score allow rapid outcome
Initial management should focus on standard principles of critical care such
stratification of patients
with intracerebral as the stabilization of the patient’s airway, breathing, and circulation.10 Hourly
hemorrhage. Nevertheless, or more frequent neurologic examinations should be scheduled immediately
they should not be used thereafter. Neuroimaging recommended by the American Heart Association
independently to guide includes CT with CT angiography to look for a spot sign (associated with
goals of care discussions of
patients with intracerebral increased risk of ICH expansion; see below) and to assess for structural vessel
hemorrhage. pathology and MRI to rule out other etiologies of ICH.10 Characteristics that
increase the likelihood of finding a vascular abnormality include female sex,
● A spot sign, defined by younger than 65 years of age with lobar ICH, primary intraventricular
the presence of contrast
within the hematoma
hemorrhage (IVH), and an absent history of hypertension, smoking, or
visualized on CT coagulopathy.11 In properly selected cases, catheter cerebral angiography can
angiography or contrast- confirm if an underlying vascular lesion is present.11 If cerebral venous
enhanced CT, is associated thrombosis is suspected, CT venography or magnetic resonance venography
with a high risk of early
(MRV) should be performed.12
expansion. ICH grading scales are routinely used to assess initial neurologic severity,
facilitate communication between providers, and frame expectations of family
members. However, they should not be used in isolation when prognosticating
survival and functional outcomes after ICH. Overreliance on grading scales can
lead to self-fulfilling prophecies in which patients with predicted poor outcomes
can have subsequent limitations in the level of care provided to them.13,14 The
ICH Score allows rapid outcome stratification of patients with ICH and has been
widely adopted since several studies have validated its feasibility.5 It includes
five independent risk factors for 30-day mortality that are assigned weights to
derive a score from 0 to 6 (TABLE 2-1 and FIGURE 2-1). The Glasgow Coma Scale
(GCS) has more utility when evaluated after clinical stabilization than on initial
assessment.16 Another useful score for clinical decision making and potential
patient selection for clinical trials is the Functional Outcome in Patients With
1604 DECEMBER 2018

Primary Intracerebral Hemorrhage (FUNC) score, which estimates the likelihood
of functional independence at 90 days. (TABLE 2-2 and TABLE 2-3).17
HEMATOMA EXPANSION
Hematoma expansion is a common cause of secondary neurologic deterioration
and is a leading modifier of survival and functional outcomes in up to one-third
of patients after ICH occurs. Hematoma expansion generally occurs within
24 hours, although delayed expansion has been described (FIGURE 2-2).16 The
strong prognostic impact of hematoma expansion on ICH outcomes is
predominantly owing to its effect on midline shift and cerebral herniation,
although even relatively small expansions can cause neurologic deterioration.18,19
Extravasation of contrast within the hematoma, called the spot sign (FIGURE 2-3),20
can be visualized on CT angiography source images or contrast-enhanced CT and
Determination of the Intracerebral Hemorrhage Scorea,b TABLE 2-1
Intracerebral Hemorrhage
Component Score Points
Glasgow Coma Scale score
3–4 2
5–12 1
13–15 0
3
Intracerebral hemorrhage volume, cm
≥30 1
<30 0
Intraventricular hemorrhage
Yes 1
No 0
Infratentorial origin of intracerebral hemorrhage
Yes 1
No 0
Age in years
≥80 1
<80 0
Total intracerebral hemorrhage score 0–6
a
Reprinted with permission from Hemphill JC 3rd, et al, Stroke.5 © 2001 American Heart Association.
b
Glasgow Coma Scale score indicates score on initial presentation (or after resuscitation); intracerebral
hemorrhage volume indicates volume on initial CT calculated using the ABC/2 method (the ABC/2 method is
the mathematical formula using the greatest three-dimensional diameters [A = length, B = width, and
C = thickness] of the intracerebral hemorrhage, that provides an accurate representation of the hematoma
volume)15; and intraventricular hemorrhage indicates the presence of any intraventricular hemorrhage on
initial CT. The total score is used to stratify 30-day mortality.

KEY POINT is an independent predictor of

early hematoma growth.21
● Although hematoma
expansion is a complication
Although such predictive value
that negatively modifies has been repeatedly and
mortality and functional independently validated, the
outcome after intracerebral current utility of this information
hemorrhage, it lacks
is not well defined in the absence
specific interventions aimed
at ameliorating its impact. of therapies specifically targeted
to prevent this feared
complication. In the PREDICT
FIGURE 2-1
Impact of the Intracerebral Hemorrhage (ICH)
(Prediction of Hematoma
score on 30-day mortality. Hemphill and Growth and Outcome in Patients
5
colleagues reported on the association between With Intracerebral Hemorrhage
mortality and the ICH score studied in a cohort Using the CT Angiography Spot
of 152 participants. Thirty-day mortality increases
as the ICH score increases. No patient with an
Sign) trial, patients with the spot
ICH score of 0 died. All patients with an ICH score sign had significantly higher
of 5 died. No patient in the University of California, mortality at 3 months (43.4%) as
San Francisco ICH cohort had an ICH score of 6, compared to patients who were
although this would be expected to be associated
spot sign negative (19.6%).22
with mortality.
Reprinted with permission from Hemphill JC 3rd, et al, Ongoing bleeding during
Stroke.5 © 2001 American Heart Association. hematoma evacuation was also
predicted by the spot sign, and a
retrospective study found that
among patients with ICH who are spot sign positive, surgical treatment was
associated with lower mortality than conservative management.23,24
Several trials have attempted to arrest hematoma expansion by acutely
controlling hypertension or by supplementing clotting factors. INTERACT 1
(Intensive Blood Pressure Reduction in Acute Cerebral Hemorrhage Trial 1)
was a feasibility study that demonstrated the safety and feasibility of intensive
blood pressure reduction in acute cerebral hemorrhage.18 The follow-up study,
INTERACT 2, randomly assigned 2839 patients to receive either intensive
treatment of systolic blood pressure to less than 140 mm Hg or treatment of
systolic blood pressure to less than 180 mm Hg as recommended by the American
Heart Association guidelines.18 The primary outcome was death or major
disability at 90 days, as assessed by the modified Rankin Scale (mRS), and was
not statistically different between subjects in the intensive treatment group
compared to the group treated according to the American Heart Association
guidelines (52% versus 55.6%, respectively), and mortality was also similar
(11.9% versus 12.0%). No significant reduction in hematoma growth was found
in the intensive treatment group, although several secondary outcomes,
including ordinal analysis of the mRS score, did show a significant favorable shift
for intensive blood pressure lowering versus guideline management. ATACH-2
(Antihypertensive Treatment of Acute Cerebral Hemorrhage II), another large
randomized trial of acute blood pressure reduction, randomly assigned 1000
subjects with an ICH volume of <60 mL and a GCS score of ≥5 to either a systolic
blood pressure goal of 110 mm Hg to 139 mm Hg or to a standard treatment goal
of 140 mm Hg to 179 mm Hg.19 All subjects received IV nicardipine. This trial
again failed to show a significant difference in the primary outcome: death or
disability at 3 months (38.7% in the intensive group versus 37.7% of the
standard group).
1606 DECEMBER 2018

The FAST (Factor Seven for Acute Hemorrhagic Stroke) trial targeted ICH
expansion using recombinant activated factor VII, which enhances local hemostasis
after binding to exposed tissue factor, facilitating conversion of prothrombin to
thrombin.25 Preliminary studies had suggested that the administration of
recombinant activated factor VII within 4 hours of ICH onset could reduce
mortality and could be associated with improved functional outcomes after
3 months.26 The phase 3 multicenter trial randomly assigned 841 patients with ICH
within 4 hours after onset to placebo, 20 mcg/kg of recombinant activated factor
VII, or 80 mcg/kg of recombinant activated factor VII. The primary outcome,
death or severe disability at 90 days, was not different between the groups
(placebo: 24%; low-dose recombinant activated factor VII: 26%; high-dose
Determinants of the Functional Outcome in Patients With Primary TABLE 2-2

Intracerebral Hemorrhage Scorea
Functional Outcome in
Patients With Primary Intracerebral
Component Hemorrhage (FUNC) Score Points
Intracerebral hemorrhage volume, cm3
<30 4
30–60 2
>60 0
Age in years
<70 2
70–79 1
≥80 0
Intracerebral hemorrhage location
Lobar 2
Deep 1
Infratentorial 0
Glasgow Coma Scale score
≥9 2
≤8 0
Pre–intracerebral hemorrhage
cognitive impairment
No 1
Yes 0
Total FUNC score 0–11
a
Reprinted with permission from Rost NS, et al, Stroke.17 © 2008 American Heart Association.

recombinant activated factor VII: 29%). The absolute reduction in ICH growth
relative to the placebo group was statistically significant but clinically minor
(2.6 mL in the low-dose group and 3.8 mL in the high-dose group). Therefore, the
use of recombinant activated factor VII in acute ICH did not find clinical benefit
in a phase 3 trial and was associated with more arterial thrombotic events.
A different approach using the antifibrinolytic agent tranexamic acid is currently
nearing completion of a multicenter phase 3 randomized trial (TICH-2
[Tranexamic Acid for Hyperacute Primary Intracerebral Haemorrhage]), testing
the hypothesis that IV tranexamic acid reduces death and disability when given
within 8hours of ICH.27
Patients with abnormalities of primary or secondary hemostasis or who are
taking oral anticoagulants (up to 20% of patients with ICH) have an increased
likelihood of ICH expansion because of their inability to form stable clots.28
Recommended therapy is emergent replacement of deficient factors and includes
vitamin K for those on vitamin K antagonists. Additionally, such patients benefit
from prothrombin complex concentrate since this corrects the international
TABLE 2-3 Proportion of Patients Who Achieve Functional Independence at 90 Days

Stratified by Functional Outcome in Patients With Primary Intracerebral
Hemorrhage Scorea
Functionally Independent at 90 Days, n/N (%)
Functional Outcome in Patients

With Primary Intracerebral Development Validation Subset
Hemorrhage (FUNC) Score Subset (n = 418)b (n = 211)b
0 0/1 (0) 0/0 (0)
1 0/12 (0) 0/2 (0)
2 0/15 (0) 0/8 (0)
3 0/37 (0) 0/17 (0)
4 0/28 (0) 0/13 (0)
5 2/48 (4) 2/19 (10)
6 4/36 (11) 0/16 (0)
7 14/77 (18) 10/46 (22)
8 20/51 (39) 14/30 (47)
9 60/87 (69) 24/40 (60)
10 9/12 (75) 6/12 (50)
11 12/14 (86) 6/8 (75)
n/N = proportion of subjects functionally independent at 90 days.

a
Reprinted with permission from Rost NS, et al, Stroke.17 © 2008 American Heart Association.
b
The development subset represents the cohort of patients used to construct the logistic regression
analysis leading to the prognostication model. The validation subset represents the group of patients used
to test the prognostication model.
1608 DECEMBER 2018

FIGURE 2-2
Head CT of patient with left basal ganglia intracerebral hemorrhage. Images show initial
head CT (A) and follow-up head CT (B) 6 hours later after clinical deterioration, demonstrating
hematoma expansion with intraventricular hemorrhage.
normalized ratio (INR) rapidly and completely, although with a small

prothrombotic risk, but overall safety is superior to fresh frozen plasma (FFP).29,30
Direct oral anticoagulants, which include direct thrombin inhibitors and factor
Xa inhibitors, are used increasingly over warfarin because of more stable
pharmacokinetics. Currently, only dabigatran (direct thrombin inhibitor) has a
reversal agent (idarucizumab, a humanized monoclonal antibody fragment
against dabigatran). For the remaining agents in this group, current
anticoagulation reversal options include prothrombin complex concentrate,
antifibrinolytic agents, and prohemostatic therapies such as desmopressin;
dialysis is not useful because of low renal excretion of the factor Xa inhibitors.
Specific reversal agents such as andexanet alfa for factor Xa inhibitors and
aripazine (which also targets direct thrombin inhibitors, and unfractionated and
low-molecular-weight heparin) are expected to be available soon. Reversal of
factor Xa inhibitors is typically managed with prothrombin complex
concentrates at this time, although randomized prospective data in humans to
support this approach is lacking.31,32
In clinical practice, considerable controversy exists surrounding the best care
of patients on antiplatelet agents who develop ICH. In the recent past, several
reports described worse clinical outcomes and increased hematoma expansion in
this subset of patients with ICH.26,28 Recent studies show that rates of hematoma
expansion and clinical outcomes are independent of antiplatelet use.29,30 Despite
a number of small observational studies suggesting clinical benefit for platelet

KEY POINTS
● Currently, no proven
benefit has been shown
from recombinant activated
factor VII in spontaneous or
anticoagulation-associated
intracerebral hemorrhage.
● In patients with
FIGURE 2-3
Spot sign demonstrating extravasation and hematoma expansion. A, Unenhanced CT
with disorders of primary or
demonstrates left posterior putaminal and internal capsule hematoma with mild surrounding
secondary hemostasis, rapid
edema. An old parietooccipital infarct is seen posterior to this. B, A small focus of
reversal of coagulopathy is
enhancement is seen peripherally on CT angiography, consistent with the spot sign (yellow
indicated in an attempt to
arrow). C, Postcontrast CT demonstrates enlargement of the spot sign, consistent with
improve neurologic outcomes
extravasation (white arrow). D, Unenhanced CT 1 day after presentation reveals hematoma
and survival.
enlargement and intraventricular hemorrhage.
Reprinted with permission from Wada R, et al, Stroke.20 © 2007 Wolters Kluwer.
with a history of recent use
of antiplatelet agents, transfusion after ICH, a single randomized phase 3 trial (PATCH [Platelet
routine use of platelet Transfusion Versus Standard Care After Acute Stroke Due to Spontaneous
transfusions is not indicated.
Cerebral Haemorrhage Associated With Antiplatelet Therapy]) did not confirm
The only exception for this
recommendation is if clinical benefit and raised the possibility of additional harm from platelet
surgical interventions are transfusion.33 This small trial enrolled 190 patients with ICH within 6 hours of
anticipated in the symptom onset who were on antiplatelet therapy; patients were randomly
emergency care of these
assigned to platelet transfusion within 90 minutes of diagnostic imaging
patients.
compared to standard care. The primary end point, a shift in ordinal mRS scores
● Perihematomal edema, at 3 months, demonstrated worse outcomes in subjects who received platelet
which contributes to early transfusion compared to those who did not (odds ratio of 2.5; 95% confidence
neurologic deterioration and interval, 1.18–3.56; P=.01). Median ICH growth at 24 hours was not different
poor outcome, develops
rapidly following
between groups, and serious adverse events were more frequent in the
intracerebral hemorrhage, transfused versus standard care group (42% versus 29%, respectively). Platelet
reaching maximal volume by transfusion is therefore not currently indicated in conservatively managed
2 weeks. There is currently patients with ICH but should be considered in patients who undergo
no proven clinical therapy
that both reduces
neurosurgical intervention.34,35
perihematomal edema and
improves outcomes. CEREBRAL EDEMA
Perihematomal edema evolves during the first 2 weeks after ICH and most
● The routine use of
rapidly evolves over the first 48 to 72 hours.36 It is associated with activation of
antiepileptic drugs following
intracerebral hemorrhage is inflammatory pathways by the toxic biochemical and metabolic effects of clot
currently not recommended. products.37 Although the association between perihematomal edema and ICH
There should be, however, outcomes has conflicting results in the literature, perihematomal edema growth
a high degree of suspicion of appears to be a biomarker for secondary injury, and a number of clinical early
electrographic seizures or
status epilepticus in patients phase studies have evaluated pharmacologic agents targeting inflammatory
with intracerebral responses with edema reduction as a clinical end point. These agents include
hemorrhage with decreased fingolimod, which significantly reduced perihematomal edema volume in a small
level of consciousness. proof-of-concept study in 23 patients with ICH treated within 72 hours of
symptom onset,38 and the iron chelator deferoxamine mesylate, studied in a
multicenter, randomized, placebo-controlled, phase 2 clinical trial to determine
whether deferoxamine reduces perihematomal edema and improves outcomes
when administered within 24 hours of ICH.39 There is currently no proven clinical
therapy that both reduces perihematomal edema and improves outcomes.
1610 DECEMBER 2018

SEIZURES
During the first 7 days after the ICH, up to 16% of patients experience clinical
seizures and up to 31% develop electrographic seizures.40–42 Seizures are the
product of cortical irritation and are most likely to occur in patients with cortical
extension of the ICH.43,44 However, while a large single-center study has shown
that prophylaxis decreases seizure risk, prospective and population-based studies
have not shown an association between clinical seizures and mortality and
outcome. Seizure prophylaxis in ICH is therefore not recommended in current
guidelines. Moreover, antiepileptic agents such as phenytoin have been
associated with worse neurologic outcomes and increased mortality,45,46 while
levetiracetam, which is increasingly used for prophylaxis, has significantly
less data regarding impact on outcomes.47 This is a point well demonstrated
by CASE 2-1.
INTRAVENTRICULAR HEMORRHAGE
IVH complicates ICH in approximately 40% of cases and is a well-established
independent predictor of increased mortality, with estimates ranging from 50%
to 80%.34,35,49–51 The surgical management of IVH has also received considerable
attention over the past few years, although it is likely that external ventricular
drains (EVDs), for treatment of obstructive hydrocephalus and to reduce the
neurotoxic effects of intraventricular blood, remain underutilized, and the
clinical and radiographic threshold for EVD insertion after IVH varies
considerably.35,52–55 The American Heart Association guidelines currently state
A 61-year-old woman presented to the emergency department for CASE 2-1

evaluation of acute-onset right-sided weakness. She had a history of
hypertension. Head CT without contrast revealed a 32 cc left basal
ganglia hemorrhage.
The following day in the neurointensive care unit, the patient
developed speech difficulties and altered level of consciousness. A
repeat head CT was unchanged. An initial spot EEG did not reveal
significant abnormalities. The patient developed a short-lasting clinical
seizure that manifested as generalized tonic-clonic convulsions and loss
of consciousness, and she remained somnolent while on an antiepileptic
drug. Continuous EEG monitoring revealed electrographic status
epilepticus. The patient required dual antiepileptic drug therapy and
benzodiazepine IV infusion to control electrographic seizure activity.
As described by Vespa and colleagues,48 the occurrence of electrographic COMMENT

seizure activity can occur in up to 28% of patients with intracerebral
hemorrhage. Although prophylactic antiepileptic drug administration
following intracerebral hemorrhage is discouraged, demonstration of ictal
activity requires a high degree of suspicion and the use of continuous EEG
to best guide the medical therapy of status epilepticus after intracerebral
hemorrhage.

that an EVD is reasonable in the management of ICH and IVH, especially in

patients with decreased level of consciousness and obstructive hydrocephalus
(Class IIa/level of evidence B).10
The brief history of intraventricular thrombolysis originates from
experimental models (canine and porcine) demonstrating that thrombolysis of
blood injected into the ventricles accelerates the resolution of the intraventricular
clot, promotes rapid return of consciousness, and improves neurologic outcome,56,57
and as a result of clinical observations that EVD placement does not immediately
clear the ventricular clot and is often complicated by obstruction from blood
clots and debris, requiring replacement. Building on the underlying biological
premise that thrombolytic therapy can decrease white blood cell infiltrates, reduce
periventricular and generalized edema,45–47,58–60 and reverse or prevent ventricular
enlargement, herniation, and coma, a large number of small clinical trials and
retrospective studies have been reported, and several systematic reviews and
meta-analyses have been performed.61–65 These reports have consistently
demonstrated mortality reduction with intraventricular thrombolysis by nearly half
(compared to EVD alone), increased the odds of good functional outcome, and
decreased the rate of shunt dependence. Ventriculitis and rehemorrhage rates were
not increased with intraventricular thrombolysis.
The CLEAR IVH (Clot Lysis: Evaluating Accelerated Resolution of
Intraventricular Hemorrhage) trials represent a multiphase clinical trial program
that sought to establish the effectiveness of intraventricular alteplase for the
treatment of obstructive IVH.66,67 These trials targeted subjects with a
spontaneous ICH volume of <30 mL, obstruction of the third or fourth ventricles
with IVH, who presented within 24 hours of symptom onset, and who
demonstrated stability of ICH, IVH, and any EVD tract hemorrhage prior to
72 hours from diagnostic CT scan.59,63
The CLEAR III trial randomly assigned patients to receive up to 12 doses of
alteplase or 0.9% saline every 8 hours via a pragmatically placed EVD until the
third and fourth ventricles were radiographically open.66 The primary outcome,
defined by mRS at 6 months (dichotomized at scores of 0 to 3 versus scores of
4 to 6), showed no statistically significant difference in good functional status
between the groups treated with saline or alteplase (saline group at 45% versus
the alteplase group at 48%, a difference of 3.5% [P=.42] after adjustment for IVH
volume and thalamic ICH location). A significant decrease in mortality occurred
at 180 days for the experimental group (18% versus 29% in the saline group;
hazard ratio of 0.60 [95% confidence interval, 0.41–0.86], P=.006), which
appeared to come at the expense of a significantly greater proportion of patients
with an mRS score of 5 (indicating patients with severe disability who are
bedridden, incontinent, and require constant nursing care and attention)
compared to the saline group (17% versus 9%; P=.007). No difference was
observed, however, in the proportion of patients in a vegetative state as
measured by the extended Glasgow Outcome Scale (3% in both groups), nor any
difference in survival in a long-term facility (alteplase 14% versus saline 12%;
P=.48). Safety events, including symptomatic hemorrhage and 30-day mortality,
were not significantly different between treatment groups, and the rate of
bacterial ventriculitis was significantly lower compared to intraventricular saline.
A significant decrease in neurologic, respiratory, and sudden deaths in the
alteplase group suggested that intraventricular alteplase may correct life-
threatening ventricular obstruction.
1612 DECEMBER 2018

In planned post hoc analyses, patients with initial IVH volumes of >20 mL and KEY POINT
those with >85% IVH removal during the active treatment phase demonstrated
● Use of an external
statistically significant increases in good functional outcome (mRS scores of 0 to ventricular drain is
3) at 180 days after adjustment for confounders. Because of a relatively small recommended in cases of
proportion (33%) of the alteplase group reaching the treatment end point of 80% intracerebral hemorrhage
IVH clot removal, a large number of subjects with small (<20 mL) IVH, and an with intraventricular
extension and obstructive
unexpectedly high proportion of patients treated with saline achieving good
hydrocephalus. External
outcomes (45%), the trial was inadequately powered to resolve narrow differences ventricular drains can
between the group treated with alteplase and the group treated with saline.66 facilitate relief of
Further investigations are planned to evaluate a targeted approach to IVH clot obstructive hydrocephalus
as well as reduce the
reduction with thrombolytic agents in both efficacy and effectiveness trials. Two
neurotoxic effects of
recent meta-analyses of thrombolytic therapy for IVH include the CLEAR III intraventricular blood.
data and largely reflect the results of the largest clinical trial. The first, with 17
studies, found a meaningful mortality benefit and possible functional outcome
benefits with different effect sizes depending on the functional outcome scale
used (mRS or Glasgow Outcome Scale).68 The second, with six studies, reported
that intraventricular thrombolysis reduced death from any cause by the end of
follow-up, but not the composite end point of death and poor functional
outcome, again suggesting an increased number of survivors with moderately
severe to severe disability.69
Intraventricular thrombolysis was deemed to be safe, neither increasing
ventriculitis nor rebleeding, but was not found to reduce the need for shunt
placement. The current status of intraventricular thrombolysis is unlikely to
significantly change the American Heart Association guidelines, which state
that intraventricular thrombolysis appears safe, while acknowledging that the
efficacy and safety of this treatment strategy remain uncertain (Class IIb/level
of evidence B).10
As with parenchymal hematoma reduction strategies, the next challenge will
be to test the hypothesis that greater ventricular clot reduction maximizes
functional outcomes consistent with reducing the biochemical effect of
ventricular blood on outcomes through reduction of hemoglobin toxicity and
inflammation, improving ventricular dilatation, and control of ICP.70,71 To this
end, newer minimally invasive approaches to IVH include endoscopic removal of
the ventricular clot and controlled lumbar drainage.72–75 The latter hypothesizes
that more rapid clearance of blood from the ventricles and removal of
inflammatory mediators using a lumbar drain may improve hydrocephalus and
CSF flow and absorption.
A randomized trial of combined intraventricular thrombolysis with lumbar
drainage versus intraventricular thrombolysis alone for prevention of permanent
shunt dependency after ICH with severe ventricular involvement was stopped
after treating 30 patients following interim analysis because of significant
efficacy of intraventricular thrombolysis plus lumbar drainage. Lumbar drainage
started after radiologic opening of the lower ventricular system was found to
be safe, avoided EVD exchange, and significantly reduced the need for
shunt surgery.74
Experience with minimally invasive endoscopic surgery for thalamic
hemorrhages with IVH and hydrocephalus in a small cohort (n = 48)
demonstrated shorter length of stay and reduced the need for shunt placement,
although without benefit in mortality or functional outcomes at 30 and 90 days.72
An earlier trial randomly assigned 42 patients with IVH to management with

EVD with intraventricular thrombolysis or to endoscopic evacuation. A

significant improvement in functional outcome was reported at 2 months in the
neuroendoscopy group.75
A meta-analysis comparing neuroendoscopic surgery combined with EVD for
IVH versus combination EVD plus intraventricular thrombolysis in 680 patients
observed significant benefits in mortality, good functional outcome, hematoma
evacuation rate, and reduced need for chronic ventriculoperitoneal CSF
diversion in favor of EVD plus intraventricular thrombolysis.73 This analysis
included small cohort studies, without protocolization of intraventricular
thrombolysis or safety event monitoring. These promising techniques will
require large-scale validation with evaluation of clot reduction, ICP control in the
acute phase, and prevention of chronic hydrocephalus.
SURGICAL MANAGEMENT
The allure of surgical reduction of hematoma volume is theoretically plausible
with advantages of correcting parenchymal displacement, decreasing ICP, and
potentially mitigating neurotoxic and inflammatory cascades.75 Although most
neurosurgeons operate on lobar or cerebellar hematomas larger than 3 cm in
patients who deteriorate clinically, as supported by current guidelines,76
uncertainty remains regarding deep hemorrhages where the parenchymal injury
required to access the hematoma appears to have been a limiting factor.54,77,78
The role of surgery for acute ICH is likely to increase significantly in the next
decade. Previously a relatively infrequent occurrence, with about one in
10 patients undergoing surgery for ICH in North America, new devices and a
focus on minimally invasive techniques are undergoing rigorous
investigation.3,54,79,80 These are intended to redefine surgical decision making
(patient selection, procedure selection, and timing) with evidence-based
models and will hopefully rescue these procedures from their most common
intended indication as a life-saving resort. Such thinking is largely influenced
and inspired by a small number of clinical trials that showed no overall benefit
for early craniotomy, ultra-early craniotomy, ultra-early treatment of bleeding,
or early neuroprotection.9 A description of the evolution of these surgical
studies follows.
In a Cochrane Review of traditional surgical evacuation utilizing standard
craniotomy compared to conservative treatment with medical care alone, 10
applicable randomized controlled trials, including 2059 participants, demonstrated
an odds ratio of 0.74 in favor of standard craniotomy for patients with ICH for
mortality and an odds ratio of 0.71 for odds of being dead or dependent at final
follow-up.81 This large meta-analysis was dominated by the STICH (Surgical Trial
in Intracerebral Haemorrhage) trial, making generalization difficult, but offered
some support for the use of craniotomy in the typical patient included in these
studies (ie, adults an average of 60 years of age with a GCS score between
5 and 15, with altered consciousness or severe neurologic deficit, and presenting
within 24 hours of onset). No evidence supports surgery after 72 hours, nor did
trials deal specifically with the deteriorating patient.54,78
In STICH I, which included 1033 patients (503 randomly assigned to surgery
and 530 randomly assigned to conservative medical therapy), it should be noted
that patients with spontaneous supratentorial ICH showed no overall benefit
from early surgery compared with initial conservative treatment.54 At 6 months,
26% of the patients who underwent surgery had a favorable outcome compared
1614 DECEMBER 2018

with 24% of those randomly assigned to conservative medical treatment (odds KEY POINTS
ratio of 0.89 [95% confidence interval, 0.66–1.19], P=.41).
● Minimally invasive
To further clarify potential subgroups that may benefit from open surgery, a endoscopic surgery for
follow-up per patient, meta-analysis of eight prospective randomized controlled intraventricular hemorrhage
trials of surgical treatment for spontaneous supratentorial ICH (2186 cases) volume reduction is
demonstrated improved outcomes with surgery for the following groups: considered experimental at
this time and awaits
(1) patients who were randomly assigned within 8 hours of hemorrhage (actual
large-scale studies to
surgery times were unavailable), (2) patients with a hematoma volume between evaluate safety and
20 mL and 50 mL, (3) patients with GCS scores of 9 to 12, and (4) patients efficacy.
between the ages of 50 and 69 years.77 Including only patients with lobar
hematomas who did not have IVH demonstrated a nonsignificant trend toward ● After several randomized
clinical trials and
a benefit of surgery. contemporary meta-
The apparent benefit of surgery for superficial (≤1 cm below the cortical analyses, no clear
surface) lobar hematoma locations without IVH54,82 and with moderate a GCS recommendation for open
score8–14,16,17 was investigated in STICH II, which tested the hypothesis that craniotomy and surgical
evacuation of supratentorial
surgery within 12 hours in this select cohort of patients with ICH could produce hypertensive hematomas
a 12% benefit of improved functional outcome.78 The concept that superficial can be made at this point.
corticotomy would limit the damage to normal brain tissue while providing
the benefit of clot volume reduction was certainly plausible, but the results of
STICH II suggested otherwise. Early surgery in conscious patients with lobar
hemorrhage did not decrease the rate of death or disability at 6 months compared
with best medical treatment. Median hematoma volume was 36 mL (23.0 mL
to 55.5 mL), and at 6 months 59% of the surgery group (n = 307) had an
unfavorable outcome versus 62% in the initial conservative group (n = 294).
Again, a high crossover rate of 21% from initial conservative to surgical treatment
was reported. Although noninferiority of surgery may be one interpretation,
the approach of standard bulk resection using open craniotomy in a pragmatic
trial did not demonstrate the hypothesized clinical improvement.
A more recent meta-analysis of 15 trials of surgery for ICH including STICH I
and II with individual patient data reported an odds ratio for death or disability of
0.74 (95% confidence interval, 0.64–0.86; P<.0001), showing a significant
advantage for surgery, although the heterogeneity score was highly significant
owing to different types of patients and surgical procedures.78 Analysis of the
subgroup with lobar ICH and no IVH demonstrated no significant benefit from
surgery, whereas inspection of point estimates from the full cohort for only
minimally invasive surgery trials suggests that minimal access techniques might
be more beneficial, especially for deeper clots.
A further meta-analysis of randomized controlled trials comparing
stereotactic puncture or endoscopic drainage versus other treatments included
12 trials with 1855 patients and reported significant reduction in end points of
both death and dependence at the end of study follow-up in favor of minimally
invasive techniques.83 Although the meta-analysis may be critiqued for
including both conservative management and craniotomy in the “other treatment”
group, the literature on minimally invasive surgery is beginning to sway the
equipoise regarding volume-reduction therapy.47,52,57 This meta-analysis also
determined that patients with smaller hemorrhages (25 mL to 40 mL) and a good
presenting GCS score (>8) were most likely to benefit from minimally
invasive surgery.
Auer and colleagues84 performed an early randomized study of endoscopic
evacuation versus medical treatment in 100 patients with ICH within 48 hours

and reported significant mortality benefit at 6 months although mostly in

patients who were noncomatose with subcortical (versus deep) ICH.
Independent but parallel work by investigators in North America and Asia has
refined a variety of procedures that now include image-guided cannulation of
the hematoma with or without use of thrombolytic drugs, mechanical clot
disruption, and endoscopic removal including with continuous focal ultrasound
delivered directly into the ICH.85–94
Data from Asia have been studied in a meta-analysis of four randomized
controlled trials including 2996 participants and compared stereotactic aspiration
to craniotomy.95 Both the odds of death or dependence at final follow-up (odds
ratio of 0.80; 95% confidence interval, 0.69–0.93; P=.004) and the risk of
intracerebral rebleeding (odds ratio of 0.44; 95% confidence interval, 0.26–0.74;
P=.002) were significantly lower in subjects treated with minimally invasive
techniques versus conventional craniotomy.
The MISTIE II (Minimally Invasive Surgery Plus Recombinant Tissue-Type
Plasminogen Activator for ICH Evacuation) study was a prospective randomized
phase 2 clinical trial designed to determine if minimally invasive surgery using
aspiration with an image-guided catheter followed by intrahematomal delivery
of alteplase over 3 days is safe and could reduce clot burden.96 At presentation,
96 subjects had a mean ICH clot size of 35 mL, the median GCS score was 11, and
the median National Institutes of Health Stroke Scale score was 24. In 54 surgical
subjects, ICH volume was significantly reduced at the end of treatment
compared with the medically treated group (average of 19.6 mL versus 40.7 mL,
P<.001) (FIGURE 2-4).97 The trial results also reported significantly decreased
cerebral edema and potential improvement in clinical outcomes in surgically
treated patients (10% improved function at the 0 to 3 mRS cut point compared to
the control arm). The finding of reduction in perihematomal edema volume in
FIGURE 2-4
Patient with right basal ganglia intracerebral hemorrhage (A) treated with minimally invasive
clot aspiration after the intraclot administration of recombinant tissue plasminogen
activator, showing near complete resolution of intracerebral hemorrhage 13 days later (B).
Reprinted with permission from Barrett RJ, et al, Neurocrit Care.97 © 2005 Humana Press.
1616 DECEMBER 2018

the surgical arm (22 ± 35%) versus an increase in edema volume of 47 ± 46% KEY POINT
in the medical arm (P<.001)98 is consistent with the reduction of toxic metabolic
● Several new devices
injury seen in animal models87,92 and inconsistent with a small number of prior placed via stereotactic
“convenience samples” and clinical reports demonstrating an increase in edema imaging enable enhanced
after exposure to recombinant tissue plasminogen activator.99 No significant drainage of parenchymal
differences were found between groups in primary safety outcomes including clots and await
demonstration of whether
mortality, symptomatic hemorrhage, and infection.
successful volumetric
Parallel efforts to the MISTIE trials have included endoscopic evacuation clot reduction is matched
without thrombolysis and minimally invasive surgery with focused ultrasound.100 with improvement in
The phase 1 multisite surgical trial ICES (Intraoperative Stereotactic CT-guided functional outcomes.
Endoscopic Surgery) demonstrated in 20 subjects that CT-guided endoscopic
evacuation without thrombolysis can achieve rapid removal of clot (71.2%
reduction in ICH volume) and similar safety results to the MISTIE phase 2 data.100
The SLEUTH (Minimally Invasive Evacuation of Spontaneous Intracerebral
Hemorrhage Using Sonothrombolysis) study combined mechanical and
thrombolytic approaches to ICH and IVH using a standard EVD catheter with a
small ultrasound-emitting probe inserted in the catheter tip fenestration.101 The
distal catheter tip emitted ultrasound (2 MHz and 0.45 W) for 24 hours in nine
subjects (six subjects with ICH and three subjects with EVD) and demonstrated
clot lysis rates of 54.1% per day in ICH, faster than was observed in the MISTIE and
CLEAR IVH studies. No rebleeding was reported.
These successes have initiated investigation and refinement of several
device-based approaches currently undergoing testing in clinical trials. The
ENRICH (Early MiNimally Invasive Removal of Intracerebral Hemorrhage) trial
is a phase 3 study investigating the procedure described as minimally invasive
subcortical parafascicular transsulcal access for clot evacuation (Mi SPACE) for
ICH.102 This approach involves a small craniotomy (2 cm to 3 cm in length)
overlying the sulcus of interest and introduction of an image-guided 13.5 mm
sheath followed by clot removal performed with standard microsurgical
techniques.103 A multicenter feasibility study including 39 subjects with an
average ICH volume of 36 mL demonstrated safety with a relatively high
rate of clot evacuation and functional independence.103 The ongoing clinical
trial has inclusion criteria similar to MISTIE III except for the exclusion of a
GCS score of <5, an upper limit of 80 mL for ICH size, and surgery within
24 hours of symptom onset. The primary end point is utility-weighted mRS at
180 days.
Another device, the Penumbra Apollo vibration/suction system (Almeda,
California), initially approved for endoscopic intraventricular neurosurgery (510[k]
device approval) utilizes a 2.6 mm aspiration wand containing a vibrational
wire, an irrigation system, and a vacuum pump used within an endoscope,
which is currently being investigated for evacuation of intracerebral hematomas.104
Use of a specific image guidance protocol is not described for the Penumbra
Apollo device and does not appear to be standardized. A retrospective review
of clinical cases (no set entry criteria or standardized technique) included
29 patients from four centers with an initial mean ICH volume of 45.4 mL.105 The
mean postoperative volume was 21.8 mL. The mortality rate was 13.8%. Further
studies of this device are ongoing. Ultimately, the goal of these and other
minimally invasive surgical approaches is to determine whether hematoma
evacuation removes the toxic effects of the blood clot and whether early and
sufficient removal translates into better clinical outcomes.

CONCLUSION
The management of ICH continues to evolve. Recent progress has been made in
the understanding of the best management practices of coagulopathy, blood
pressure control, cerebral edema, and outcome prognostication during the care
of patients with ICH. However, specific treatments for the underlying
pathophysiologic process(es) triggered by the exposure of viable brain
parenchyma to blood and its degradation products are lacking.
The surgical management of ICH is undergoing rapid evolution with the early
successes of minimally invasive surgical techniques. The path forward will
require phase 3 randomized controlled trials to determine individual device
efficacy, best approaches, and the populations of patients with ICH most likely
to benefit from surgery.
REFERENCES
1 Andrews CM, Jauch EC, Hemphill JC 3rd, et al. 9 Keep RF, Hua Y, Xi G. Intracerebral haemorrhage:
Emergency neurological life support: intracerebral mechanisms of injury and therapeutic targets.
hemorrhage. Neurocrit Care 2012;17(suppl 1): Lancet Neurol 2012;11(8):720–731. doi:10.1016/
S37–S46. doi:10.1007/s12028-012-9757-2. S1474-4422(12)70104-7.
2 Hanley DF, Awad IA, Vespa PM, et al. 10 Hemphill JC 3rd, Greenberg SM, Anderson CS,
Hemorrhagic stroke: introduction. Stroke 2013; et al. Guidelines for the management of
44(6 suppl 1):S65–S66. doi:10.1161/STROKEAHA. spontaneous intracerebral hemorrhage: a
113.000856. guideline for healthcare professionals from the
American Heart Association/American Stroke
3 Broderick J, Connolly S, Feldmann E, et al.
Guidelines for the management of spontaneous
doi:10.1161/STR.0000000000000069.
intracerebral hemorrhage in adults: 2007 update:
a guideline from the American Heart Association/ 11 Delgado Almandoz JE, Jagadeesan BD, Moran CJ,
American Stroke Association Stroke Council, et al. Independent validation of the secondary
High Blood Pressure Research Council, and the intracerebral hemorrhage score with catheter
Quality of Care and Outcomes in Research angiography and findings of emergent
Interdisciplinary Working Group. Circulation hematoma evacuation. Neurosurgery 2012;70(1):
2007;116(16):e391–e413. doi:10.1161/ 131–140; discussion 140. doi:10.1227/NEU.
CIRCULATIONAHA.107.183689. 0b013e31822fbf43.
4 Balami JS, Buchan AM. Complications of 12 Stam J. Thrombosis of the cerebral veins and
intracerebral haemorrhage. Lancet Neurol 2012; sinuses. N Engl J Med 2005;352(17):1791–1798.
11(1):101–118. doi:10.1016/S1474-4422(11)70264-2. doi:10.1056/NEJMra042354.
5 Hemphill JC 3rd, Bonovich DC, Besmertis L, et al. 13 Becker KJ, Baxter AB, Cohen WA, et al.
The ICH score: a simple, reliable grading scale for Withdrawal of support in intracerebral
intracerebral hemorrhage. Stroke 2001;32(4): hemorrhage may lead to self-fulfilling
891–897. doi:10.1161/01.STR.32.4.891. prophecies. Neurology 2001;56(6):766–772.
doi:10.1212/WNL.56.6.766.
6 Dye JA, Rees G, Yang I, et al. Neuropathologic
analysis of hematomas evacuated from patients 14 Hemphill JC 3rd, White DB. Clinical nihilism in
with spontaneous intracerebral hemorrhage. neuroemergencies. Emerg Med Clin North Am
Neuropathology 2014;34(3):253–260. doi:10.1111/ 2009;27(1):27–37, vii–viii. doi:10.1016/j.emc.
neup.12089. 2008.08.009.
7 Halpin SF, Britton JA, Byrne JV, et al. Prospective 15 Gebel JM, Sila CA, Sloan MA, et al. Comparison of
evaluation of cerebral angiography and the ABC/2 estimation technique to computer-
computed tomography in cerebral hematoma. assisted volumetric analysis of intraparenchymal
J Neurol Neurosurg Psychiatry 1994;57:1180–1186. and subdural hematomas complicating the
GUTO-1 trial. Stroke 1998;29(9):1799–1801.
8 Zhu XL, Chan MS, Poon WS. Spontaneous
intracranial hemorrhage: which patients need 16 Dowlatshahi D, Demchuk AM, Flaherty ML, et al.
diagnostic cerebral angiography? A prospective Defining hematoma expansion in intracerebral
study of 206 cases and review of the literature. hemorrhage: relationship with patient outcomes.
Stroke 1997;28(7):1406–1409. doi:10.1161/01. Neurology 2011;76(14):1238–1244. doi:10.1212/
STR.28.7.1406. WNL.0b013e3182143317.
1618 DECEMBER 2018

17 Rost NS, Smith EE, Chang Y, et al. Prediction of 29 Moussouttas M, Malhotra R, Fernandez L, et al.
functional outcome in patients with primary Role of antiplatelet agents in hematoma
intracerebral hemorrhage: FUNC score. expansion during the acute period of
Stroke 2008;39(8):2304–2309. doi:10.1161/ intracerebral hemorrhage. Neurocrit Care 2010;
STROKEAHA.107.512202. 12(1):24–29. doi:10.1007/s12028-009-9290-0.
18 Anderson CS, Heeley E, Huang Y, et al. Rapid 30 Sansing LH, Messe SR, Cucchiara BL, et al. Prior
blood-pressure lowering in patients with acute antiplatelet use does not affect hemorrhage growth
intracerebral hemorrhage. N Engl J Med 2013; or outcome after ICH. Neurology 2009;72(16):
368(25):2355–2365. doi:10.1056/NEJMoa1214609. 1397–1402. doi:10.1212/01.wnl.0000342709.31341.88.
19 Qureshi AI, Palesch YY, Barsan WG, et al. 31 Lambourne MD, Eltringham-Smith LJ, Gataiance S,
Intensive blood-pressure lowering in patients et al. Prothrombin complex concentrates reduce
with acute cerebral hemorrhage. N Engl J blood loss in murine coagulopathy induced by
Med 2016;375(11):1033–43. doi:10.1056/ warfarin, but not in that induced by dabigatran
NEJMoa1603460. etexilate. J Thromb Haemost 2012;10(9):
1830–1840. doi:10.1111/j.1538-7836.2012.04863.x.
20 Wada R, Aviv RI, Fox AJ, et al. CT angiography
“spot sign” predicts hematoma expansion in 32 Marlu R, Hodaj E, Paris A, et al. Effect of non-
acute intracerebral hemorrhage. Stroke specific reversal agents on anticoagulant activity
2007;38(4):1257–1262. doi:10.1161/01.STR. of dabigatran and rivaroxaban: a randomised
0000249633.5904.f3. crossover ex vivo study in healthy volunteers.
Thromb Haemost 2012;108(2):217–224. doi:10.1160/
21 Hallevi H, Abraham AT, Barreto AD, et al.
TH12-03-0179.
The spot sign in intracerebral hemorrhage: the
importance of looking for contrast extravasation. 33 Baharoglu MI, Cordonnier C, Al-Shahi Salman R,
Cerebrovasc Dis 2010;29(3):217–220. doi:10.1159/ et al; PATCH Investigators. Platelet transfusion
000267842. versus standard care after acute stroke due to
spontaneous cerebral haemorrhage associated
22 Demchuk AM, Dowlatshahi D, Rodriguez-Luna D,
with antiplatelet therapy (PATCH): a randomised,
et al; PREDICT/Sunnybrook ICH CTA study group.
open-label, phase 3 trial. Lancet 2016;387(10038):
Prediction of haematoma growth and outcome in
2605–2613. doi:10.1016/S0140-6736(16)30392-0.
patients with intracerebral haemorrhage using
the CT-angiography spot sign (PREDICT): a 34 Qureshi AI, Mendelow AD, Hanley DF. Intracerebral
prospective observational study. Lancet Neurol haemorrhage. Lancet 2009;373(9675):1632–1644.
2012;11(4):307–314. doi:10.1016/S1474-4422(12) doi:10.1016/S0140-6736(09)60371-8.
70038-8.
35 Steiner T, Diringer MN, Schneider D, et al.
23 Brouwers HB, Raffeld MR, van Nieuwenhuizen Dynamics of intraventricular hemorrhage in
KM, et al. CT angiography spot sign in patients with spontaneous intracerebral
intracerebral hemorrhage predicts active hemorrhage: risk factors, clinical impact, and
bleeding during surgery. Neurology 2014;83(10): effect of hemostatic therapy with recombinant
883–889. doi:10.1212/WNL.0000000000000747. activated factor VII. Neurosurgery 2006;59(4):
767–774; discussion 773–774. doi:10.1227/01.
24 Kim HT, Lee JM, Koh EJ, Choi HY. Surgery versus
NEU.0000232837.34992.32.
conservative treatment for spontaneous
supratentorial intracerebral hemorrhage in spot 36 Venkatasubramanian C, Mlynash M, Finley-
sign positive patients. J Korean Neurosurg Soc Caulfield A, et al. Natural history of
2015;58(4):309–315. doi:10.3340/jkns.2015.58.4.309. perihematomal edema after intracerebral
hemorrhage measured by serial magnetic
25 Mayer SA, Brun NC, Begtrup K, et al; FAST Trial
resonance imaging. Stroke 2011;42(1):73–80.
Investigators. Efficacy and safety of
doi:10.1161/STROKEAHA.110.590646.
recombinant activated factor VII for acute
intracerebral hemorrhage. N Engl J Med 2008; 37 Urday S, Kimberly WT, Beslow LA, et al. Targeting
358(20):2127–2137. doi:10.1056/NEJMoa0707534. secondary injury in intracerebral haemorrhage—
perihaematomal oedema. Nat Rev Neurol 2015;
26 Saloheimo P, Ahonen M, Juvela S, et al. Regular
11(2):111–122. doi:10.1038/nrneurol.2014.264.
aspirin-use preceding the onset of primary
intracerebral hemorrhage is an independent 38 Fu Y, Hao J, Zhang N, et al. Fingolimod for the
predictor for death. Stroke 2006;37(1):129–133. Treatment of Intracerebral Hemorrhage: a 2-arm
doi:10.1161/01.STR.0000196991.03618.31. proof-of-concept Study. JAMA Neurol 2014;71(9):
1092–1101. doi:10.1001/jamaneurol.2014.1065.
27 Flaherty K, Bath PM, Dineen R, et al. Statistical
analysis plan for the ‘Tranexamic acid for 39 Yeatts SD, Palesch YY, Moy CS, Selim M. High
hyperacute primary IntraCerebral Haemorrhage’ dose deferoxamine in intracerebral hemorrhage
(TICH-2) trial. Trials 2017;18(1):607. doi:10.1186/ (HI-DEF) trial: rationale, design, and methods.
s13063-017-2341-5. Neurocrit Care 2013;19(2):257–266. doi:10.1007/
s12028-013-9861-y.
28 Toyoda K, Okada Y, Minematsu K, et al.
Antiplatelet therapy contributes to acute 40 Bladin CF, Alexandrov AV, Bellavance A, et al.
deterioration of intracerebral hemorrhage. Seizures after stroke: a prospective multicenter
Neurology 2005;65(7):1000–1004. doi:10.1212/ study. Arch Neurol 2000;57(11):1617–1622.
01.wnl.0000179178.37713.69. doi:10.1001/archneur.57.11.1617.

41 De Herdt V, Dumont F, Hénon H, et al. Early 54 Mendelow AD, Gregson BA, Fernandes HM, et al.
seizures in intracerebral hemorrhage: incidence, Early surgery versus initial conservative treatment
associated factors, and outcome. Neurology in patients with spontaneous supratentorial
2011;77(20):1794–1800. doi:10.1212/WNL. intracerebral haematomas in the International
0b013e31823648a6. Surgical Trial in Intracerebral Haemorrhage
(STICH): a randomised trial. Lancet 2005;365(9457):
42 Szaflarski JP, Rackley AY, Kleindorfer DO, et al.
387–397. doi:10.1016/S0140-6736(05)17826-X.
Incidence of seizures in the acute phase of
stroke: a population-based study. Epilepsia 2008; 55 Tuhrim S, Horowitz DR, Sacher M, Godbold JH.
49(6):974–981. doi:10.1111/j.1528-1167.2007.01513.x. Volume of ventricular blood is an important
determinant of outcome in supratentorial
43 Beghi E, D’Alessandro R, Beretta S, et al.
intracerebral hemorrhage. Crit Care Med 1999;
Incidence and predictors of acute symptomatic
27(3):617–621.
seizures after stroke. Neurology 2011;77(20):
1785–1793. doi:10.1212/WNL.0b013e3182364878. 56 Mayfrank L, Kissler J, Raoofi R, et al. Ventricular
dilatation in experimental intraventricular
44 Biffi A, Rattani A, Anderson CD, et al. Delayed
hemorrhage in pigs. Characterization of
seizures after intracerebral haemorrhage. Brain
cerebrospinal fluid dynamics and the effects of
2016;139(pt 10):2694–2705. doi:10.1093/brain/
fibrinolytic treatment. Stroke 1997;28(1):141–148.
aww199.
57 Pang D, Sclabassi RJ, Horton JA. Lysis of
45 Fargen KM, Hoh BL, Neal D, et al. The burden and
intraventricular blood clot with urokinase in a
risk factors of ventriculostomy occlusion in a
canine model: part 1. Canine intraventricular
high-volume cerebrovascular practice: results of
blood cast model. Neurosurgery 1986;19(4):
an ongoing prospective database. J Neurosurg
540–546. doi:10.1227/00006123-198610000-
2016;124(6):1805–1812. doi:10.3171/2015.5.
00008.
JNS15299.
58 Mayfrank L, Kim Y, Kissler J, et al. Morphological
46 Jaffe J, Melnychuk E, Muschelli J, et al.
changes following experimental intraventricular
Ventricular catheter location and the clearance
haemorrhage and intraventricular fibrinolytic
of intraventricular hemorrhage. Neurosurgery
treatment with recombinant tissue plasminogen
2012;70(5):1258–1264; discussion 1263–1264.
activator. Acta Neuropathol 2000;100(5):561–567.
doi:10.1227/NEU.0b013e31823f6571.
59 Narayan RK, Narayan TM, Katz DA, et al. Lysis
47 Liu DZ, Waldau B, Ander BP, et al. Inhibition of Src
of intracranial hematomas with urokinase in a
family kinases improves cognitive function after
rabbit model. J Neurosurg 1985;62(4):580–586.
intraventricular hemorrhage or intraventricular
doi:10.3171/jns.1985.62.4.0580.
thrombin. J Cereb Blood Flow Metab 2017;37(7):
2359–2367. doi:10.1177/0271678X16666291. 60 Wagner KR, Xi G, Hua Y, et al. Ultra-early clot
aspiration after lysis with tissue plasminogen
48 Vespa PM, O’Phelan K, Shah M, et al. Acute
activator in a porcine model of intracerebral
seizures after intracerebral hemorrhage: a factor
hemorrhage: edema reduction and blood-brain
in progressive midline shift and outcome.
barrier protection. J Neurosurg 1999;90(3):
Neurology 2003;60(9):1441–1446.
491–498. doi:10.3171/jns.1999.90.3.0491.
49 Hanley DF. Intraventricular hemorrhage: severity
61 Gaberel T, Magheru C, Parienti JJ, et al.
factor and treatment target in spontaneous
Intraventricular fibrinolysis versus external
intracerebral hemorrhage. Stroke 2009;40(4):
ventricular drainage alone in intraventricular
1533–1538. doi:10.1161/STROKEAHA.108.535419.
hemorrhage: a meta-analysis. Stroke 2011;42(10):
50 Tuhrim S, Dambrosia JM, Price TR, et al. 2776–2781. doi:10.1161/STROKEAHA.111.615724.
Intracerebral hemorrhage: external validation
62 Khan NR, Tsivgoulis G, Lee SL, et al. Fibrinolysis
and extension of a model for prediction of
for intraventricular hemorrhage: an updated
30-day survival. Ann Neurol 1991;29(6):658–663.
meta-analysis and systematic review of the
doi:10.1002/ana.410290614.
literature. Stroke 2014;45(9):2662–2669.
51 Young WB, Lee KP, Pessin MS, et al. Prognostic doi:10.1161/STROKEAHA.114.00599.
significance of ventricular blood in supratentorial
63 LaPointe M, Haines S. Fibrinolytic therapy for
hemorrhage: a volumetric study. Neurology 1990;
intraventricular hemorrhage in adults. Cochrane
40(4):616–619. doi:10.1212/WNL.40.4.616.
Database Syst Rev 2002;(3):CD003692.
52 Herrick DB, Ullman N, Nekoovaght-Tak S, et al.
64 Nieuwkamp DJ, de Gans K, Rinkel GJ, Algra A.
Determinants of external ventricular drain
Treatment and outcome of severe
placement and associated outcomes in patients
intraventricular extension in patients with
with spontaneous intraventricular hemorrhage.
subarachnoid or intracerebral hemorrhage: a
systematic review of the literature. J Neurol
s12028-014-9959-x.
2000;247(2):117–121.
53 Lovasik BP, McCracken DJ, McCracken CE, et al.
65 Staykov D, Bardutzky J, Huttner HB, Schwab S.
The effect of external ventricular drain use in
Intraventricular fibrinolysis for intracerebral
intracerebral hemorrhage. World Neurosurg 2016;
hemorrhage with severe ventricular involvement.
94:309–318. doi:10.1016/j.wneu.2016.07.022.
s12028-010-9390-x.
1620 DECEMBER 2018

66 Hanley DF, Lane K, McBee N, et al. Thrombolytic 77 Gregson BA, Broderick JP, Auer LM, et al.
removal of intraventricular haemorrhage in Individual patient data subgroup meta-analysis
treatment of severe stroke: results of the of surgery for spontaneous supratentorial
randomised, multicentre, multiregion, placebo- intracerebral hemorrhage. Stroke 2012;43(6):
controlled CLEAR III trial. Lancet 2017;389(10069): 1496–1504. doi:10.1161/STROKEAHA.111.640284.
603–611. doi:10.1016/S0140-6736(16)32410-2.
78 Mendelow AD, Gregson BA, Rowan EN, et al.
67 Naff N, Williams MA, Keyl PM, et al. Low-dose Early surgery versus initial conservative
recombinant tissue-type plasminogen activator treatment in patients with spontaneous
enhances clot resolution in brain hemorrhage: supratentorial lobar intracerebral haematomas
the intraventricular hemorrhage thrombolysis (STICH II): a randomised trial. Lancet 2013;
trial. Stroke 2011;42(11):3009–3016. doi:10.1161/ 382(9890):397–408. doi:10.1016/S0140-
STROKEAHA.110.610949. 6736(13)60986-1.
68 Baker AD, Rivera Perla KM, Yu Z, et al. Fibrinolytic 79 Prasad KS, Gregson BA, Bhattathiri PS, et al. The
for treatment of intraventricular hemorrhage: a significance of crossovers after randomization in
meta-analysis and systematic review. Int J Stroke the STICH trial. Acta Neurochir Suppl 2006;96:
2018;13(1):11–23. doi:10.1177/1747493017730745. 61–64.
69 Wang D, Liu J, Norton C, et al. Local fibrinolytic 80 Steiner T, Kaste M, Forsting M, et al.
therapy for intraventricular hemorrhage: a meta- Recommendations for the management of
analysis of randomized controlled trials. World intracranial haemorrhage—part I: spontaneous
Neurosurg 2017;107:1016.e1–1024.e1. doi:10.1016/ intracerebral haemorrhage. The European Stroke
j.wneu.2017.07.135. Initiative Writing Committee and the Writing
Committee for the EUSI Executive Committee.
70 Chen Q, Tang J, Tan L, et al. Intracerebral
Cerebrovasc Dis 2006;22(4):294–316. doi:10.
hematoma contributes to hydrocephalus after
1159/000094831.
intraventricular hemorrhage via aggravating iron
accumulation. Stroke 2015;46(10):2902–2908. 81 Prasad K, Mendelow AD, Gregson B. Surgery
doi:10.1161/STROKEAHA.115.009713. for primary supratentorial intracerebral
haemorrhage. Cochrane Database Syst Rev
71 Ziai WC, Melnychuk E, Thompson CB, et al.
2008;(4):Cd000200. doi:10.1002/14651858.
Occurrence and impact of intracranial pressure
CD000200.pub2.
elevation during treatment of severe
intraventricular hemorrhage. Crit Care Med 82 Bhattathiri PS, Gregson B, Prasad KS, et al.
2012;40(5):1601–1608. doi:10.1097/CCM. Intraventricular hemorrhage and hydrocephalus
0b013e318241e380. after spontaneous intracerebral hemorrhage:
results from the STICH trial. Acta Neurochir Suppl
72 Chen CC, Liu CL, Tung YN, et al. Endoscopic
2006;96:65–68.
surgery for intraventricular hemorrhage (IVH)
caused by thalamic hemorrhage: comparisons 83 Zhou X, Chen J, Li Q, et al. Minimally invasive
of endoscopic surgery and external ventricular surgery for spontaneous supratentorial
drainage (EVD) surgery. World Neurosurg 2011; intracerebral hemorrhage: a meta-analysis of
75(2):264–268. doi:10.1016/j.wneu.2010.07.041. randomized controlled trials. Stroke 2012;43(11):
2923–2930. doi:10.1161/STROKEAHA.112.667535.
73 Li Y, Zhang H, Wang X, et al. Neuroendoscopic
surgery versus external ventricular drainage 84 Auer LM, Deinsberger W, Niederkorn K, et al.
alone or with intraventricular fibrinolysis for Endoscopic surgery versus medical treatment
intraventricular hemorrhage secondary to for spontaneous intracerebral hematoma: a
spontaneous supratentorial hemorrhage: a randomized study. J Neurosurg 1989;70(4):
systematic review and meta-analysis. PLoS 530–535. doi:10.3171/jns.1989.70.4.0530.
One 2013;8(11):e80599. doi:10.1371/journal.
85 Chen CC, Cho DY, Chang CS, et al. A stainless
pone.0080599.
steel sheath for endoscopic surgery and its
74 Staykov D, Kuramatsu JB, Bardutzky J, et al. application in surgical evacuation of putaminal
Efficacy and safety of combined intraventricular haemorrhage. J Clin Neurosci 2005;12(8):937–940.
fibrinolysis with lumbar drainage for prevention doi:10.1016/j.jocn.2005.04.006.
of permanent shunt dependency after
86 Cho DY, Chen CC, Chang CS, et al. Endoscopic
intracerebral hemorrhage with severe ventricular
surgery for spontaneous basal ganglia
involvement: a randomized trial and individual
hemorrhage: comparing endoscopic surgery,
patient data meta-analysis. Ann Neurol 2017;81(1):
stereotactic aspiration, and craniotomy in
93–103. doi:10.1002/ana.24834.
noncomatose patients. Surg Neurol 2006;65(6):
75 Zhang Z, Li X, Liu Y, et al. Application of 547–556; discussion 555–556. doi:10.1016/
neuroendoscopy in the treatment of j.surneu.2005.09.032.
intraventricular hemorrhage. Cerebrovasc Dis
87 Hsieh PC, Cho DY, Lee WY, Chen JT. Endoscopic
2007;24(1):91–96. doi:10.1159/000103122.
evacuation of putaminal hemorrhage: how to
76 Steiner T, Al-Shahi Salman R, Beer R, et al. improve the efficiency of hematoma evacuation.
European Stroke Organisation (ESO) guidelines Surg Neurol 2005;64(2):147–153; discussion 153.
for the management of spontaneous doi:10.1016/j.surneu.2004.11.028.
intracerebral hemorrhage. Int J Stroke 2014;9(7):
840–855. doi:10.1111/ijs.12309.

88 Kuo LT, Chen CM, Li CH, et al. Early 98 Mould WA, Carhuapoma JR, Muschelli J, et al.
endoscope-assisted hematoma evacuation in Minimally invasive surgery plus recombinant
patients with supratentorial intracerebral tissue-type plasminogen activator for
hemorrhage: case selection, surgical technique, intracerebral hemorrhage evacuation decreases
and long-term results. Neurosurg Focus 2011; perihematomal edema. Stroke 2013;44(3):
30(4):E9. doi:10.3171/2011.2.FOCUS10313. 627–634. doi:10.1161/STROKEAHA.111.000411.
89 Nagasaka T, Tsugeno M, Ikeda H, et al. Early 99 Ducruet AF, Hickman ZL, Zacharia BE, et al.
recovery and better evacuation rate in Exacerbation of perihematomal edema and
neuroendoscopic surgery for spontaneous sterile meningitis with intraventricular
intracerebral hemorrhage using a multifunctional administration of tissue plasminogen activator
cannula: preliminary study in comparison with in patients with intracerebral hemorrhage.
craniotomy. J Stroke Cerebrovasc Dis 2011;20(3): Neurosurgery 2010;66(4):648–655. doi:10.1227/
208–213. doi:10.1016/j.jstrokecerebrovasdis. 01.NEU.0000360374.59435.60.
2009.11.021.
100 Vespa P, Hanley D, Betz J, et al. ICES
90 Nakano T, Ohkuma H, Ebina K, Suzuki S. (intraoperative stereotactic computed
Neuroendoscopic surgery for intracerebral tomography-guided endoscopic surgery) for
haemorrhage—comparison with traditional brain hemorrhage: a multicenter randomized
therapies. Minim Invasive Neurosurg 2003;46(5): controlled trial. Stroke 2016;47(11):2749–2755.
278–283. doi:10.1055/s-2003-44451. doi:10.1161/STROKEAHA.116.013837.
91 Nishihara T, Nagata K, Tanaka S, et al. Newly 101 Newell DW, Shah MM, Wilcox R, et al.
developed endoscopic instruments for the Minimally invasive evacuation of spontaneous
removal of intracerebral hematoma. Neurocrit intracerebral hemorrhage using sonothrombolysis.
Care 2005;2(1):67–74. doi:10.1385/NCC:2:1:067. J Neurosurg 2011;115(3):592–601. doi:10.3171/
2011.5.JNS10505.
92 Ochalski P, Chivukula S, Shin S, et al. Outcomes
After Endoscopic Port Surgery for Spontaneous 102 ClinicalTrials.gov. ENRICH: Early MiNimally-
Intracerebral Hematomas. J Neurol Surg A Cent invasive removal of intracerebral hemorrhage
Eur Neurosurg 2014;75(3):195–205; discussion (ICH) (ENRICH), NCT02880878. clinicaltrials.gov/
206. doi:10.1055/s-0033-1348348. ct2/show/NCT02880878. Updated June 4, 2018.
Accessed October 4, 2018.
93 Yamamoto T, Nakao Y, Mori K, Maeda M.
Endoscopic hematoma evacuation for 103 Labib MA, Shah M, Kassam AB, et al. The
hypertensive cerebellar hemorrhage. Minim safety and feasibility of image-guided
Invasive Neurosurg 2006;49(3):173–178. BrainPath-mediated transsulcul hematoma
doi:10.1055/s-2006-944242. evacuation: a multicenter study. Neurosurgery
2017;80(4):515–524. doi:10.1227/NEU.
94 Zhu H, Wang Z, Shi W. Keyhole endoscopic
0000000000001316.
hematoma evacuation in patients. Turk
Neurosurg 2012;22(3):294–299. doi:10.5137/ 104 Turner RD, Vargas J, Turk AS, et al. Novel
1019-5149.JTN.5136-11.1. device and technique for minimally invasive
intracerebral hematoma evacuation in the same
95 Wang JW, Li JP, Song YL, et al. Stereotactic
setting of a ruptured intracranial aneurysm:
aspiration versus craniotomy for primary
combined treatment in the neurointerventional
intracerebral hemorrhage: a meta-analysis of
angiography suite. Neurosurgery 2015;11(suppl 2):
randomized controlled trials. PLoS One 2014;9(9):
43–50; discussion 50–51. doi:10.1227/NEU.
e107614. doi:10.1371/journal.pone.0107614.
0000000000000650.
96 Hanley DF, Thompson RE, Muschelli J, et al.
105 Spiotta AM, Fiorella D, Vargas J, et al. Initial
Safety and efficacy of minimally invasive surgery
multicenter technical experience with the
plus alteplase in intracerebral haemorrhage
Apollo device for minimally invasive intracerebral
evacuation (MISTIE): a randomized, controlled,
hematoma evacuation. Neurosurgery 2015;
open-label, phase 2 trial. Lancet Neurol 2016;15(12):
11(suppl 2):243–251; discussion 251. doi:10.1227/
1228–1237. doi:10.1016/S1474-4422(16)30234-4.
NEU.0000000000000698.
97 Barrett RJ, Hussain R, Coplin WM, et al. Frameless
stereotactic aspiration and thrombolysis of
spontaneous intracerebral hemorrhage.
NCC:3:3:237.
1622 DECEMBER 2018

Subarachnoid REVIEW ARTICLE

Hemorrhage C O N T I N U UM AUDIO
INTERVIEW AVAILABLE
ONLINE
By Susanne Muehlschlegel, MD, MPH, FNCS, FCCM
CITE AS:
ABSTRACT 1623–1657.
PURPOSE OF REVIEW: This article reviews the epidemiology, clinical
presentation, diagnosis, and management of patients with aneurysmal Dr Susanne Muehlschlegel,
subarachnoid hemorrhage (SAH). SAH is a type of hemorrhagic stroke and Departments of Neurology,
is a neurologic emergency with substantial morbidity and mortality. This Anesthesia/Critical Care &
Surgery, University of
article reviews the most common and potentially life-threatening Massachusetts Medical School,
neurologic and medical complications to promote their early recognition University Campus, S-5, 55 Lake
and prevent secondary brain injury. Ave N, Worcester, MA 01655,
susanne.muehlschlegel@
umassmemorial.org.
RECENT FINDINGS: Over the past 30 years, the incidence of SAH has remained
RELATIONSHIP DISCLOSURE:
stable; yet, likely because of improved care in specialized neurocritical
Dr Muehlschlegel has received
care units, discharge mortality has considerably decreased. Two research/grant support from the
consensus guidelines by the American Heart Association/American Stroke National Institutes of Health/
National Institute of Child Health
Association and the Neurocritical Care Society have outlined best and Human Development and
practices for the management of patients with SAH. The most important the Prize for Academic
recommendations include admission of patients to high-volume centers Collaboration and Excellence
(PACE) from the University of
(defined as more than 35 SAH admissions per year) under the management Massachusetts Memorial
of a multidisciplinary, specialized team; expeditious identification and Medical Group. Dr Muehlschlegel
treatment of the bleeding source with evaluation by a multidisciplinary receives partial research salary
support as the site principal
team consisting of cerebrovascular neurosurgeons, neuroendovascular investigator for the INTREPID
specialists, and neurointensivists; management of patients in a (Impact of Fever Prevention
Continued on page 1657
neurocritical care unit with enteral nimodipine, blood pressure control,
euvolemia, and close monitoring for neurologic and medical
UNLABELED USE OF
complications; and treatment of symptomatic cerebral vasospasm/ PRODUCTS/INVESTIGATIONAL
delayed cerebral ischemia with induced hypertension and endovascular USE DISCLOSURE:
therapies. This article also highlights new insights of SAH pathophysiology Dr Muehlschlegel discusses
the unlabeled/investigational
and provides updates in the management approach. short-term use of
antifibrinolytics (ε-aminocaproic
SUMMARY: SAH remains a neurologic emergency. Management of patients acid and tranexamic acid) for
the treatment of early aneurysm
with SAH includes adherence to published guidelines, but some areas of bleeding, the use of
SAH management remain understudied. Clinical trials are required to fludrocortisone for the
treatment of cerebral salt
elucidate the role of these controversial management approaches in
wasting syndrome after
improving patient outcomes. subarachnoid hemorrhage, the
use of levetiracetam for seizure
prophylaxis, and the use of
INTRODUCTION milrinone, nicardipine, and
N
verapamil as endovascular
ontraumatic subarachnoid hemorrhage (SAH) is a type of therapy using intraarterial
hemorrhagic stroke most commonly due to the rupture of saccular vasodilators for the treatment
(berry) aneurysms and comprises 3% of all stroke types.1 While a of subarachnoid hemorrhage.
17% to 50% decrease in the worldwide case fatality rate has been © 2018 American Academy
reported in the last 2 to 3 decades, thought to be due to more rapid of Neurology.

SUBARACHNOID HEMORRHAGE
recognition and improved treatment strategies, the prehospital and 30-day

mortality rates remain high (15% and 35%, respectively).2,3 The annual incidence
of aneurysmal SAH has not declined, affecting 9/100,000 people in the
United States and with approximately 600,000 cases worldwide.4
Despite a decline in the mortality, SAH remains a highly morbid disease.
Survivors are commonly left with permanent disability, cognitive deficits
(particularly executive functioning and short-term memory), and mental health
symptoms (eg, depression, anxiety), resulting in a significant reduction in
health-related quality of life, which has been reported to occur in 35% of patients
1 year after SAH.5–7 The average age at aneurysm rupture is 53 years; SAH onset
at this young age results in a high societal cost and a number of years of
lost productivity.8
The most common cause for SAH is a ruptured cerebral aneurysm (85%);
however, despite modern neuroimaging techniques, 10% of SAHs may not
reveal a bleeding source, while the minority of cases (5%) may be due to
other vascular causes (eg, arteriovenous malformation, arteriovenous fistula,
reversible cerebral vasoconstriction syndrome [RCVS]).9 Particularly in RCVS,
the presence of high-convexity SAH, rather than SAH in the basal cisterns, in
addition to the typical “sausage shape” areas of constriction/vasodilation on
vessel imaging has been described.10
TABLE 3-1 Risk Factors for Subarachnoid Hemorrhagea
Modifiable Risk Factors

◆ Hypertension
◆ Smoking
◆ Heavy alcohol use
◆ Sympathomimetic drug use (eg, cocaine)
Nonmodifiable Risk Factors
◆ Increasing age (peak in fifth and sixth decade)
◆ Female sex
◆ African American ethnicity
◆ Hispanic ethnicity
◆ Japanese or Finnish ethnicity
◆ Prior history of subarachnoid hemorrhage
◆ Family history of subarachnoid hemorrhage
◆ History of aneurysm in two or more first-degree relatives
◆ Autosomal dominant polycystic kidney disease
◆ Type IV Ehlers-Danlos syndrome
◆ Cerebral aneurysms of more than 7 mm in diameter
a
Modified with permission from Suarez JI, Continuum (Minneap Minn).12
© 2015 American Academy of Neurology.
1624 DECEMBER 2018

Several epidemiologic and genetic risk factors for SAH have been identified. KEY POINTS
Notably, SAH is predominant in women (female to male ratio of 1.6:1), African
● Despite a decline in the
Americans, and Hispanics.9 Hypertension, smoking, and excess alcohol intake mortality, subarachnoid
are modifiable risk factors that individually double the risk of SAH.9,11 Several hemorrhage remains a highly
other nonmodifiable and genetic risk factors have emerged (TABLE 3-1).1,9,12–14 morbid disease.
Patient counseling on the modifiable risk factors is recommended to reduce the
risk of SAH.
subarachnoid hemorrhage,
Current guidelines recommend screening for aneurysms if the patient has two aneurysm rupture occurs at
or more first-degree relatives with aneurysms or SAH.15–18 This is based on an average age of 53 years.
several long-term cohort screening studies from the Familial Intracranial This young age at onset
Aneurysm study and the International Study of Unruptured Intracranial results in a high societal cost
and number of years of
Aneurysms.15–18 Notably, siblings are more likely than children of patients productivity lost.
with SAH to have an unruptured intracranial aneurysm detected.
A plethora of genetic studies have emerged for unruptured and ruptured ● The most common cause
intracranial aneurysms using linkage and genome-wide association approaches. for subarachnoid
hemorrhage is a ruptured
The American Heart Association’s “Guidelines for the Management of Patients cerebral aneurysm (85%);
With Unruptured Intracranial Aneurysms” provides a brief overview of the however, 10% of
current state of genetics in intracranial aneurysms and SAH.19 While a subarachnoid hemorrhages
meta-analysis of both unruptured and ruptured intracranial aneurysms may not reveal a bleeding
source, while the minority of
identified the interleukin-6 (IL6) gene polymorphism G572C (chromosome 7)
cases (5%) may be due to
to have an elevated risk for aneurysm formation, no predominant genetic risk other vascular causes.
factor has been identified. Several other single-nucleotide polymorphisms
have been associated with aneurysm formation, with the strongest associations ● Subarachnoid
on chromosome 9 (near CDKN2B antisense inhibitor gene), chromosome 8 hemorrhage is predominant
in women, African
(near the SOX17 transcription regulator gene), and chromosome 4 (near the Americans, and Hispanics.
EDNRA gene).19 Hypertension, smoking, and
Controversy surrounds the heritability of aneurysms and SAH, and several excess alcohol intake are
studies, including a twin study, have suggested that environmental risk factors modifiable risk factors that
individually double the
(many of which are modifiable) are far more important than genetic or familial risk of subarachnoid
inheritance.20–22 A recent review suggests that familial does not equal genetic hemorrhage.
because of the familial aggregation of risk factors (such as smoking and
hypertension).20 Currently, routine genetic screening is not performed. ● Current guidelines
recommend screening for
SAH remains one of the top neurologic emergencies treated in a neurocritical
aneurysms if the patient has
care unit. Neurologists should familiarize themselves with this highly morbid two or more first-degree
disease, particularly in light of the emerging science on acute brain injury after relatives with aneurysms or
SAH and changes in the classic teaching of the etiology of cerebral vasospasm and subarachnoid hemorrhage.
delayed cerebral ischemia.
● Subarachnoid
The acute phase of SAH can be divided into two disease phases (TABLE 3-2): hemorrhage typically
(1) prompt evaluation, recognition, and diagnosis23; immediate transfer to presents with a sudden and
appropriate SAH centers13,24; and rapid treatment of the bleeding source25,26 and severe headache (“worst
(2) close monitoring in a neurocritical care unit with expertise in SAH and overall headache of life”), which is
distinctly different from
good neurocritical care that adheres to existing management guidelines to usual headaches.
prevent or ameliorate secondary neurologic and medical complications.13,24
CLINICAL PRESENTATION
SAH typically presents with a sudden and severe headache (often described as
the “worst headache of life”), which is distinctly different from usual headaches
and is often accompanied by loss of consciousness, nausea, vomiting,
photophobia, and neck pain (CASE 3-1). A small proportion of patients may
experience a headache without many or any of the associated symptoms

(sentinel headache) and may either not seek medical attention or are
misdiagnosed, thereby remaining unrecognized, with a high risk for major
life-threatening rebleeding within a short period of time (hours to days).23,27,28
Other less typical presenting signs may be seizures, acute encephalopathy, and
concomitant subdural hematoma with or without associated head trauma (due to
the SAH-related syncope), which may make a diagnosis of aneurysmal SAH
more difficult.23
The physical examination should include determination of the level of
consciousness and the patient’s score on the Glasgow Coma Scale, evaluation for
meningeal signs, and presence of focal neurologic deficits. In cases with unusual
presentation or uncertainty, funduscopic evaluation may be helpful. Intraocular
hemorrhage associated with SAH (Terson syndrome) is associated with
increased mortality and may be seen in 40% of patients with SAH.29
Transient elevation in intracranial pressure (ICP) is the cause of nausea,
vomiting, and syncope and may be associated with additional cardiac and
pulmonary complications after SAH. The intraocular hemorrhages in Terson
syndrome are thought to be due to the sudden elevation in the ICP. When ICP
elevations are severe and sustained, coma and rapid deterioration to brain death
can result.
DIAGNOSIS
Several diagnostic modalities may be used for the diagnosis of SAH.
Head Computed Tomography

The most rapidly available and appropriate initial diagnostic test for patients with
suspected SAH is a noncontrast head CT (CASE 3-1). It is important to correlate
head CT findings to the time of headache onset, as the sensitivity of head CT
changes over the first 7 days from 93% (first 6 hours), to close to 100% (first
TABLE 3-2 Two Phases of Caring for a Patient With Subarachnoid Hemorrhage
Phase 1: Diagnosis and Aneurysm Treatment (Minutes to Hours)

◆ Recognize that subarachnoid hemorrhage may be present
◆ Aggressive blood pressure control (systolic blood pressure of less than 160 mm Hg)
◆ Rapid diagnostics
◆ Rapid initiation of aneurysm treatment (if not at a high-volume center, transfer to one)
◆ Secure aneurysm (clip or coil)
Phase 2: Vasospasm Period and Prevention of Delayed Cerebral Ischemia (Days to Weeks)
◆ Admit to neurocritical care unit specialized in subarachnoid hemorrhage care
◆ Hemodynamic and oxygenation monitoring
◆ Guideline-driven, protocolized neurocritical care
◆ Clinical examination and monitoring for vasospasm
◇ If available, daily transcranial Doppler trend evaluation
◇ If available, multimodality brain monitoring
◇ If symptomatic vasospasm, medical and interventional therapy
1626 DECEMBER 2018

12 hours), to 93% (first day), to less than 60% (at 7 days).30 The characteristic KEY POINTS
appearance of hyperdense blood in the basal cisterns or sylvian, interhemispheric,
● Physical examination of a
and interpeduncular fissures should immediately lead to the suspicion of an patient with subarachnoid
aneurysmal etiology. In fact, any SAH on head CT, especially in the absence of a hemorrhage should include
trauma history, should prompt further vessel imaging. In addition to the determination of the level of
presence of SAH, clinicians should also evaluate the head CT for presence of consciousness and the
patient’s score on the
hydrocephalus, intraventricular hemorrhage, and intracerebral hemorrhage.
Glasgow Coma Scale,
evaluation for meningeal
Lumbar Puncture signs, and the presence of
In cases of negative or equivocal head CT findings in which a high suspicion for focal neurologic deficits.
SAH still exists, a lumbar puncture is the immediate next recommended step
● Transient elevation in the
(FIGURE 3-2).31 Opening pressure should be measured routinely. To differentiate intracranial pressure is the
a traumatic tap from true SAH, CSF should be collected in four consecutive cause of nausea, vomiting,
tubes, with red blood cell count measured in tubes one and four. CSF should be and syncope and may be
spun down and evaluated for xanthochromia by visual inspection and, if associated with additional
cardiac and pulmonary
available, spectrophotometry, which is superior in diagnostic accuracy for complications after
xanthochromia than visual inspection alone.32 subarachnoid hemorrhage.
Xanthochromia takes approximately 12 hours to develop and may not be
present if a lumbar puncture is performed earlier after headache onset. Most ● The most rapidly available
and appropriate initial
hospitals do not offer spectrophotometry, and it is unknown what the false-
diagnostic test for patients
negative rate for xanthochromia is at various time intervals after SAH onset. with suspected
subarachnoid hemorrhage is
Magnetic Resonance Imaging a noncontrast head CT.
Head CT and MRI are considered to be equally sensitive in detecting SAH in the
● In cases of negative or
first 2 days, except in the hyperacute first 6 hours after SAH, during which head equivocal head CT findings
CT may miss a small proportion of SAHs and MRI may be slightly superior.30 in which a high suspicion still
Because of its rapid image acquisition, its widespread availability in the exists for subarachnoid
emergency department, and its very high sensitivity in the first 2 days after SAH, hemorrhage, a lumbar
puncture is the immediate
head CT remains the diagnostic modality of choice for early SAH. However, next recommended step.
hemosiderin-sensitive MRI sequences (gradient recalled echo [GRE] and
susceptibility-weighted imaging [SWI]) or fluid-attenuated inversion recovery ● CSF should be spun down
(FLAIR) sequences have superior sensitivity to detect subacute or chronic SAH and evaluated for
xanthochromia by visual
compared to head CT.33 Additionally, MRI may be helpful in differentiating
inspection and, if available,
alternative pathologies, such as arteriovenous malformations and inflammatory, spectrophotometry.
infectious, and neoplastic etiologies.30 Xanthochromia takes
approximately 12 hours to
develop and may not be
Identifying the Bleeding Source
present if a lumbar puncture
Vessel imaging should be the next step in all patients with a diagnostic head CT, is performed earlier after
lumbar puncture, or MRI. The gold standard of vessel imaging remains cerebral headache onset.
digital subtraction angiography (DSA) (CASE 3-1). CT angiography (CTA) has
become widely available and is now commonly performed as the first-line ● Head CT and MRI are
considered equally sensitive
vascular imaging in many institutions. Depending on the technique, slice in detecting subarachnoid
thickness, and postimaging data processing, the sensitivity and specificity of hemorrhage in the first
CTA can range from 90% to 97% and 93% to 100%, respectively, when compared 2 days, except in the
to DSA.34,35 However, CTA may miss aneurysms as small as 4 mm or less.34 hyperacute first 6 hours
after subarachnoid
In specialized stroke and aneurysm centers, DSA is readily available for hemorrhage, during which
diagnostic as well as treatment purposes. At the author’s institution, two- head CT may miss a small
dimensional and three-dimensional DSA is pursued as standard diagnostics for proportion of subarachnoid
aneurysm detection as soon as the diagnosis of SAH has been established; CTA is hemorrhages and MRI may
be slightly superior.
usually omitted at the author’s institution to prevent the additional exposure to

CASE 3-1 A 43-year-old woman with a past medical history of smoking and
depression presented to a community hospital with sudden onset of
severe headache, brief loss of consciousness, nausea, and vomiting while
using the bathroom. She reported a moderate, persistent, sudden-onset
headache that had continued for 36 hours.
In the emergency department, her blood pressure was 185/100 mm Hg,
pulse was 105 beats/min, arterial oxygen saturation was 95% on room air,
and temperature was 36.8°C (98.2°F).
On examination, she reported neck pain, was disoriented (Glasgow
Coma Scale score of 13), but had no focal deficits. Her World Federation
of Neurological Surgeons Scale (WFNSS) score was 2, Hunt and Hess
Scale score was 3, and modified Fisher Scale score was 4. Bolus
injections of 10 mg IV labetalol and 4 mg IV morphine resulted in a partial
blood pressure reduction to 170/90 mm Hg. She was started on a
nicardipine infusion to achieve and maintain a systolic blood pressure of
less than 160 mm Hg.
A noncontrast head CT revealed subarachnoid hemorrhage (SAH) in
multiple cisterns, intraventricular hemorrhage, and mild hydrocephalus
(FIGURE 3-1A). She received a loading dose of IV levetiracetam.
The patient was immediately transferred to a comprehensive stroke
center and was admitted to the neurocritical care unit. Since her Glasgow
Coma Scale had worsened to a score of 10 because of hydrocephalus, she
had an external ventricular drain (EVD) placed, which was kept clamped
prior to aneurysm coiling with intermittent opening to drain 5 mL to 10 mL
of CSF hourly.
After discussion among the interventional neuroradiologist,
cerebrovascular neurosurgeon, and neurointensivist, the patient
underwent digital subtraction angiography (DSA) and coiling of her
unsecured aneurysm (FIGURE 3-1B–D). Following the coiling, the patient was
transferred back to the neurocritical care unit, where she received
enteral nimodipine, pain control, IV normal saline to maintain euvolemia,
intermittent compression devices, and subcutaneous enoxaparin for
chemoprophylaxis for deep vein thrombosis, and IV dexamethasone
(for 2 days only) for refractory headaches. The levetiracetam was
discontinued based on guideline recommendations, as further discussed
in the section on seizures and seizure prophylaxis in this article.
Nicardipine was discontinued, and she maintained a systolic blood
pressure between 140 mm Hg and 165 mm Hg spontaneously, with a goal
systolic blood pressure of 100 mm Hg to 200 mm Hg with the secured
aneurysm. Her EVD was leveled at 10 mm Hg and open, and 5 mL to 12 mL
of CSF was drained hourly. Her neurologic examination improved to a
Glasgow Coma Scale score of 15 with no focal deficits, and she was
mobilized out of bed.
1628 DECEMBER 2018

FIGURE 3-1
Neuroimaging of the patient in CASE 3-1. Noncontrast head CT (A) showing acute
subarachnoid hemorrhage, and two-dimensional (B) and three-dimensional (C) four-vessel
angiogram showing a large top of the basilar artery aneurysm with an irregular shape
(B, C, arrows). Two-dimensional angiogram after coiling shows obliteration of the aneurysm
(D, arrow).
This case demonstrates the first phase of SAH management: rapid COMMENT
diagnosis of SAH and hydrocephalus and immediate emergency treatment
with blood pressure lowering, EVD placement, and obliteration of the
ruptured cerebral aneurysm.

FIGURE 3-2
Diagnostic algorithm for subarachnoid hemorrhage.
CT = computed tomography; CTA = computed tomography angiography; DSA = digital subtraction angiography.
Reprinted with permission from Suarez JI, et al, N Engl J Med.31 © 2006 Massachusetts Medical Society.
radiation, iodine contrast, and its potential for anaphylactic reaction and
nephrotoxicity. Patients with a negative initial DSA should have a repeat study
7 to 14 days after the initial one. In addition, in those with negative initial DSA,
MRI of the brain and, depending on the location of the SAH, MRI of the cervical
spine should be performed to search for a possible arteriovenous malformation
of the brain, brainstem, or spinal cord.9,30
Perimesencephalic Subarachnoid Hemorrhage

Approximately 15% of patients with SAH will have negative imaging studies for a
source of bleeding, of which approximately 38% have nonaneurysmal
perimesencephalic SAH,36 a special form of nontraumatic SAH with blood
isolated to the perimesencephalic cisterns (CASE 3-2).37 The clinical course has
been reported to be more benign,36 although case reports have been published
1630 DECEMBER 2018

demonstrating rare cases of small aneurysms in the posterior circulation,
fenestration of the vertebral or basilar arteries, or anterior spinal artery
abnormalities. Therefore, DSA should still be performed. At the author’s
institution, a brain and cervical spine MRI, as well as a repeat DSA approximately
7 days after the initial tests, are performed in all cases of perimesencephalic SAH.
Patients with perimesencephalic SAH are monitored in a step-down unit and,
if no bleeding source is discovered, discharged after 8 to 10 days at the author’s
institution, as long as their hospital course remains uncomplicated.
INITIAL EVALUATION
This section focuses on the emergency department evaluation and management
of a patient with SAH.
A 23-year-old man presented to the emergency department for CASE 3-2

evaluation after developing a sudden-onset headache posteriorly while
lifting weights in the gym. He described the headache as the “worst
headache of his life.”
A noncontrast head CT revealed blood around the perimesencephalic
and prepontine cisterns (FIGURE 3-3). Six-vessel angiography did not reveal
a bleeding cause. He underwent MRI of the brain and cervical spine,
which also did not reveal a cause for the bleeding. He was observed in the
step-down unit until 8 days posthemorrhage, when he received a repeat
six-vessel angiogram, which was again negative. The patient was
discharged home on day 10 post–subarachnoid hemorrhage in normal
condition.
FIGURE 3-3
Imaging of the patient in CASE 3-2. Noncontrast
head CT at two brainstem levels showing blood in
the perimesencephalic cisterns (arrows) consistent
with a perimesencephalic subarachnoid hemorrhage.
This case demonstrates a typical presentation and hospital course for a COMMENT
perimesencephalic subarachnoid hemorrhage, with the adequate workup
for a bleeding source. The patient’s course was benign, as commonly seen
in patients with perimesencephalic subarachnoid hemorrhage.

Airway, Breathing, and Circulation

The emergency evaluation and management of patients with SAH should focus
on the airway, breathing, and circulation (the ABCs).13,24,38 Those patients
unable to protect their airway should be intubated immediately, which includes
patients in coma, in stupor from hydrocephalus, with seizures, or patients in
need of sedation for agitation.
Rebleeding
The focus in the first few minutes to hours after SAH, until the patient can
undergo treatment of the ruptured aneurysm, should be directed toward the
prevention of rebleeding. This life-threatening complication, with a mortality rate
of 20% to 60%, has its highest rate (8% to 23%) within the first 72 hours after SAH,
with the majority of rebleeding (50% to 90%) occurring within the first 6 hours,
not including patients who die before hospital arrival.9 After the first month,
rebleeding rates are low, at 3% per year. Risk factors for rebleeding include
poor-grade SAH, hypertension, a large aneurysm, and, potentially, the use of
antiplatelet drugs.27 Particularly, blood pressure fluctuations and extreme blood
pressure peaks should be avoided because of the presumed propensity to
cause rebleeding.39
Current guideline recommendations for blood pressure goals are to keep
systolic blood pressure below 160 mm Hg.13,24 Continuous blood pressure
TABLE 3-3 Clinical and Radiologic Grading Scales for Subarachnoid Hemorrhagea
World Federation of Neurological

Surgeons Scale43 Hunt and Hess Scale44 Modified Fisher Scale45
Glasgow Neurologic Subarachnoid Intraventricular

Grade Coma Scale Examination Grade Neurologic Examination Scale Hemorrhage Hemorrhage
1 15 No motor 1 Awake, alert, no cranial 0 Absent Absent

deficit nerve or motor deficits,
mild headache, minimal or
no nuchal rigidity
2 13–14 No motor 2 Awake, alert, moderate to 1 Thin Absent

deficit severe headache, nuchal
rigidity, no motor deficits,
may have cranial nerve
deficit
3 13–14 Motor deficit 3 Confusion or lethargy, with 2 Thin Present

or without mild focal
neurologic deficits
4 7–12 With or without 4 Stuporous, more severe 3 Thickb Absent

motor deficit focal neurologic deficit
5 3–6 With or without 5 Comatose, motor 4 Thickb Present

motor deficit posturing or no motor
response
a
Modified with permission from Suarez JI, et al, N Engl J Med.31 © 2006 Massachusetts Medical Society.
b
Thick is defined as a subarachnoid hemorrhage filling one or more cisterns or fissures out of a total of 10 cisterns/fissures: interhemispheric
fissure, the quadrigeminal cistern, both suprasellar cisterns, both ambient cisterns, both basal sylvian fissures, and both lateral sylvian fissures.
1632 DECEMBER 2018

monitoring with an arterial line is highly recommended. IV medications KEY POINTS
to control blood pressure should preferably be continuous infusions of
● Hemosiderin-sensitive
antihypertensives (nicardipine 5 mg/h to 15 mg/h or labetalol 5 mg/h to 20 mg/h) MRI sequences (gradient
over bolus infusions (labetalol 5 mg bolus to 20 mg bolus, captopril) to prevent recalled echo and
wide fluctuations of blood pressure that may be as detrimental to aneurysm susceptibility-weighted
rebleeding as high blood pressure itself. imaging) or fluid-attenuated
inversion recovery
Therefore, at the author’s institution, hydralazine is avoided as it can cause
sequences have superior
rebound hypertension. Pain control is best achieved with short-acting opiates. sensitivity to detect
Meningeal chemical irritation from the SAH often responds to one or several subacute or chronic
single doses of dexamethasone (2 mg to 10 mg). subarachnoid hemorrhage
compared to head CT.
Antifibrinolytics ● The “gold standard”

While prolonged infusion of antifibrinolytics can result in deep vein thrombosis, vessel imaging remains
venous thromboembolism, stroke, and myocardial infarction, and should cerebral digital subtraction
angiography.
therefore not be applied, the short-term use (up to a maximum of 72 hours until
aneurysm securement) of antifibrinolytics (tranexamic acid or ε-aminocaproic ● Approximately 15% of
acid) is recommended by guidelines based on a randomized controlled trial patients with subarachnoid
and several small observational studies.24,40,41 Recently, however, a large hemorrhage will have
retrospective study including 341 patients over 12 years, of whom 146 patients negative imaging studies for
a source of bleeding, of
received ε-aminocaproic acid before their endovascular coiling, showed that which approximately 38%
short-term antifibrinolytic therapy was safe but did not reduce preprocedural have nonaneurysmal
rebleeding.42 Therefore, institutional variation may occur in the use of short- perimesencephalic
term antifibrinolytics to prevent rebleeding until a randomized controlled trial subarachnoid hemorrhage.
confirms or refutes the guideline recommendations.
● The focus in the first few
minutes to hours after
Disease Severity Scoring subarachnoid hemorrhage,
The importance of severity scoring lies in the observation that outcome and until the patient can undergo
treatment of the ruptured
delayed cerebral ischemia are associated with clinical and radiologic scales,
aneurysm, should be
respectively (TABLE 3-3).43–45 Disease severity scoring provides a common directed toward the
language between all providers caring for a patient with SAH.31 The two most prevention of rebleeding.
commonly used clinical scales, the World Federation of Neurological Surgeons
Scale (WFNSS) and the Hunt and Hess Scale, are strong predictors of outcome.43,44 ● Risk factors for
rebleeding include
Higher scores are associated with worse clinical outcome. The most reliable and poor-grade subarachnoid
validated radiologic scale is the modified Fisher Scale (TABLE 3-3), which is hemorrhage, hypertension,
nearly linearly associated with worse cerebral vasospasm and delayed cerebral a large aneurysm, and,
ischemia.46 potentially, the use of
antiplatelet drugs.
The classic Fisher Scale score holds several disadvantages and has largely been
replaced by the modified Fisher Scale.45 For example, it was developed on 1980s ● The short-term use (up to
head CTs and does not reflect modern multislice head CT imaging. Also, it is not a maximum of 72 hours until
linear (a Fisher Scale score of 3 has a higher vasospasm risk than a score of 4), aneurysm securement) of
antifibrinolytics (tranexamic
making it more difficult to apply in regression models in research and simply
acid or ε-aminocaproic
making it less intuitive to apply at the bedside. acid) is recommended by
guidelines, although there is
Admission to High-volume Centers institutional variation in
If the patient is not already at a high-volume SAH-specialized center (defined their use.
as more than 35 SAH cases per year with experienced cerebrovascular
surgeons, endovascular specialists, and neurocritical care services),12 transfer
to such a center should be initiated immediately (TABLE 3-4). Likely owing to
lack of protocolized care and expertise, admission of patients with SAH to
low-volume centers is associated with a higher 30-day mortality.47 Admission

TABLE 3-4 Summary of Key Recommendations for the Management of Patients With
Subarachnoid Hemorrhagea
American Heart Association/American

Treatment Decision Stroke Association13,b Neurocritical Care Society24,c
Hospital/system Low-volume hospitals (eg, fewer than 10 Patients with SAH should be treated at high-
characteristics subarachnoid hemorrhage [SAH] cases per year) volume centers (moderate quality of evidence,
should consider early transfer of patients with SAH strong recommendation).
to high-volume centers (eg, more than 35 SAH
High-volume centers should have appropriate
cases per year) with experienced cerebrovascular
specialty neurointensive care units,
surgeons, endovascular specialists, and
neurointensivists, vascular neurosurgeons, and
multidisciplinary neurointensive care services
interventional neuroradiologists to provide the
(Class I, Level B).
essential elements of care (moderate quality of
After discharge, it is reasonable to refer patients evidence, strong recommendation).
with SAH for a comprehensive evaluation, including
cognitive, behavioral, and psychosocial
assessments (Class IIa, Level B).
Aneurysm treatment Surgical clipping or endovascular coiling of the Early aneurysm repair should be undertaken,
ruptured aneurysm should be performed as early as when possible and reasonable to prevent
feasible in the majority of patients to reduce the rebleeding (high quality of evidence, strong
rate of rebleeding after SAH (Class I, Level B). recommendation).
For patients with ruptured aneurysms judged to be An early, short course of antifibrinolytic therapy
technically amenable to either endovascular coiling prior to early aneurysm repair (begun at
or neurosurgical clipping, endovascular coiling diagnosis and continued up to the point at which
should be considered (Class I, Level B). the aneurysm is secured or at 72 hours postictus,
whichever is shorter) should be considered (low
Complete obliteration of the aneurysm is
quality of evidence, weak recommendation).
recommended whenever possible (Class I, Level B).
Delayed (more than 48 hours after the ictus) or
Stenting of a ruptured aneurysm is associated with
prolonged (more than 3 days) antifibrinolytic
increased morbidity and mortality (Class III, Level C).
therapy exposes patients to side effects of
For patients with an unavoidable delay in therapy when the risk of rebleeding is sharply
obliteration of aneurysm, a significant risk of reduced and should be avoided (high quality of
rebleeding, and no compelling medical evidence, strong recommendation).
contraindications, short-term (less than 72 hours)
therapy with tranexamic acid or aminocaproic acid
is reasonable to reduce the risk of early aneurysm
rebleeding (Class IIa, Level B).
Blood pressure Between the time of SAH symptom onset and Treat extreme hypertension in patients with an
control aneurysm obliteration, blood pressure should be unsecured, recently ruptured aneurysm. Modest
controlled with a titratable agent to balance the elevations in blood pressure (mean blood
risk of stroke, hypertension-related rebleeding, pressure of less than 110 mm Hg) do not require
and maintenance of cerebral perfusion pressure therapy. Premorbid baseline blood pressures
(Class I, Level B). should be used to refine targets and
hypotension should be avoided (low quality of
The magnitude of blood pressure control to reduce
evidence, strong recommendation).
the risk of rebleeding has not been established, but
a decrease in systolic blood pressure to less than
160 mm Hg is reasonable (Class IIa, Level C).
Intravascular Maintenance of euvolemia and normal circulating Intravascular volume management should target
volume status blood volume is recommended to prevent delayed euvolemia and avoid prophylactic hypervolemic
cerebral ischemia (Class I, Level B). therapy. In contrast, there is evidence for harm
from aggressive administration of fluid aimed at
achieving hypervolemia (moderate quality of
CONTINUED ON PAGE 1635
1634 DECEMBER 2018

CONTINUED FROM PAGE 1634

Cardiopulmonary No recommendations given Baseline cardiac assessment with serial

complications enzymes, ECG, and echocardiography is
recommended, especially in patients with
evidence of myocardial dysfunction (low quality
of evidence, strong recommendation).
Monitoring of cardiac output may be useful in
patients with evidence of hemodynamic
instability or myocardial dysfunction (low quality
Seizures The use of prophylactic anticonvulsants may be Routine use of anticonvulsant prophylaxis with
considered in the immediate posthemorrhagic phenytoin is not recommended after SAH (low
period (Class IIb, Level B). quality of evidence, strong recommendation).
The routine long-term use of anticonvulsants is not If anticonvulsant prophylaxis is used, a short
recommended (Class III, Level B). course (3–7 days) is recommended (low quality
of evidence, weak recommendation).
Continuous EEG monitoring should be
considered in patients with poor-grade SAH
who fail to improve or who have neurologic
deterioration of undetermined etiology (low
quality of evidence, strong recommendation).
Fever treatment Aggressive control of fever to a target of During the period of risk for delayed cerebral
normothermia by use of standard or advanced ischemia, control of fever is desirable; intensity
temperature modulating systems is reasonable in should reflect the individual patient’s relative
the acute phase of SAH (Class IIa, Level B). risk of ischemia (low quality of evidence, strong
recommendation).
Surface cooling or intravascular devices are
more effective and should be employed when
antipyretics fail in cases where fever control is
highly desirable (high quality of evidence, strong
recommendation).
Glucose control Careful glucose management with strict avoidance Hypoglycemia (serum glucose of less than
of hypoglycemia may be considered as part of the 80 mg/dL) should be avoided (high quality of
general critical care management of patients with evidence, strong recommendation).
SAH (Class IIb, Level B)
Serum glucose should be maintained below
200 mg/dL (moderate quality of evidence,
strong recommendation).
Deep vein Heparin-induced thrombocytopenia and deep vein Measures to prevent deep vein thrombosis
thrombosis thrombosis are relatively frequent complications should be employed in all patients with SAH
prophylaxis after SAH. Early identification and targeted (high quality of evidence, strong
treatment are recommended, but further research recommendation).
is needed to identify the ideal screening paradigms
The use of unfractionated heparin for
(Class I, Level B)
prophylaxis could be started 24 hours after
undergoing aneurysm obliteration (moderate
quality of evidence, strong recommendation).


Delayed cerebral Oral nimodipine should be administered to all Oral nimodipine (60 mg every 4 hours) should be
ischemia patients with SAH (Class I, Level A). administered after SAH for a period of 21 days
(high quality of evidence, strong
Maintenance of euvolemia and normal circulating
recommendation).
blood volume is recommended to prevent delayed
cerebral ischemia (Class I, Level B). The goal should be maintaining euvolemia, rather
than attempting hypervolemia (moderate quality
Prophylactic hypervolemia or balloon angioplasty
before the development of angiographic spasm is
not recommended (Class III, Level B). Transcranial Doppler may be used for monitoring
and detection of large artery vasospasm with
Transcranial Doppler is reasonable to monitor for
variable sensitivity (moderate quality of
the development of arterial vasospasm (Class IIa,
Level B).
Digital subtraction angiography is the gold
Perfusion imaging with CT or MRI can be useful to
standard for detection of large artery
identify regions of potential brain ischemia (Class
vasospasm (high quality of evidence, strong
IIa, Level B).
recommendation).
Induction of hypertension is recommended for
Patients clinically suspected of delayed cerebral
patients with delayed cerebral ischemia unless
ischemia should undergo a trial of induced
blood pressure is elevated at baseline or cardiac
hypertension (moderate quality of evidence,
status precludes it (Class I, Level B).
strong recommendation).
Cerebral angioplasty and/or selective intraarterial
Endovascular treatment using intraarterial
vasodilator therapy is reasonable in patients with
vasodilators and/or angioplasty may be
symptomatic vasospasm, particularly those who
considered for vasospasm-related delayed
are not responding to hypertensive therapy (Class
cerebral ischemia (moderate quality of
IIa, Level B).
Anemia and The use of packed red blood cell transfusion to Patients should receive packed red blood cell
transfusion treat anemia might be reasonable in patients with transfusions to maintain hemoglobin
SAH who are at risk of cerebral ischemia. The concentration above 8–10 g/dL (moderate
optimal hemoglobin goal is still to be determined quality of evidence, strong recommendation).
(Class IIb, Level B).
Hyponatremia The use of fludrocortisone acetate and hypertonic Fluid restriction should not be used to treat
saline solution is reasonable for preventing and hyponatremia (weak quality of evidence, strong
correcting hyponatremia (Class IIa, Level B). recommendation).
Early treatment with hydrocortisone or
fludrocortisone may be used to limit natriuresis
and hyponatremia (moderate quality of
evidence, weak recommendation).
Mild hypertonic saline solutions can be used to
correct hyponatremia (very low quality of
CT = computed tomography; ECG = electrocardiogram; EEG = electroencephalogram; MRI = magnetic resonance imaging.
a
Reprinted with permission from Suarez JI, Continuum (Minneap Minn).12 © 2015 American Academy of Neurology.
b
American Heart Association/American Stroke Association recommendations follow the American Heart Association Stroke Council’s methods of
classifying the level of certainty of the treatment effect and the class of evidence.
c
For the Neurocritical Care Society’s guidelines, the quality of the data was assessed and recommendations developed using the Grading of
Recommendations, Assessment, Development, and Evaluation (GRADE) system.
1636 DECEMBER 2018

to a neurocritical care unit staffed by dedicated neurointensivists has been
associated with lower in-hospital mortality in patients with stroke,
including SAH.48
Aneurysm Treatment
With the publication of the ISAT (International Subarachnoid Aneurysm Trial),25,26
which compared endovascular coiling to surgical clipping after SAH, the treatment
of an unsecured aneurysm has shifted from surgical clipping to mostly
endovascular coiling.49 ISAT showed that patients in the endovascular coiling
group had significantly higher odds of survival free of disability 1 year after SAH
and a lower risk of epilepsy when compared to the surgical clipping group.
Even 10 years after SAH, patients who underwent endovascular coiling had
better outcomes.50 In contrast, the risk of rebleeding and incomplete occlusion of
the aneurysm was lower with surgical clipping. With the introduction of newer
techniques such as stent-assisted or balloon-assisted coiling, even broad-neck
aneurysms can now be treated with endovascular coiling.
Currently, endovascular coiling is preferred over surgical clipping whenever
possible. However, follow-up angiograms are necessary, as the recurrence rate of
aneurysms is higher when they are treated with endovascular coiling.50
At the author’s institution, approximately 95% of aneurysms are treated with
endovascular coiling (CASE 3-1). Many aneurysms are not equally suited for
endovascular coiling or surgical clipping (TABLE 3-5 and CASE 3-3A). The choice
of treatment depends on the patient’s age as well as the aneurysm location,
morphology, and relationship to adjacent vessels. A multidisciplinary approach
to the swift treatment decision with consensus between cerebrovascular
neurosurgeons, neuroendovascular specialists, and neurointensivists is
recommended given the complexity of the decision. Regardless of the treatment
modality, rebleeding must be prevented, and the unsecured aneurysm must be
treated as soon as possible (TABLE 3-4).
Critical Care Management of Subarachnoid Hemorrhage

SAH is a systemic disease and is not isolated to the brain. It is commonly
associated with systemic inflammatory response syndrome (SIRS) (75%), which
is related to elevated levels of inflammatory cytokines. SIRS has been associated
with long-term cognitive dysfunction and has been linked to nonconvulsive
seizures in SAH.51 SIRS has been found to precede nonconvulsive seizures, and
patients with in-hospital nonconvulsive seizures are almost twice as likely to
have SIRS as those without nonconvulsive seizures. Therefore, it has been
Preference for Treatment of Unsecured Aneurysmsa TABLE 3-5
Treatment Type Clinical or Aneurysm Factors Supporting Treatment Type

Endovascular coiling Older age, poor clinical grade, multiple comorbidities, top of the basilar aneurysm, high surgical risk,
aneurysm suitable for coiling or clipping
Surgical clipping Aneurysm with wide neck-to-body ratio, crucial arteries arising from aneurysm dome, middle cerebral
artery aneurysm, aneurysm with large parenchymal hematoma
a
Modified with permission from Suarez JI, Continuum (Minneap Minn).12 © 2015 American Academy of Neurology.

CASE 3-3A A 51-year-old man presented to the emergency department of a

comprehensive stroke center for evaluation of sudden-onset severe
headache with rapidly worsening left-sided hemiplegia that began while
shoveling snow. Upon arrival, his blood pressure was 205/110 mm Hg, his
heart rate was 98 beats/min, his oxygen saturation was 98% on room air,
and his temperature was 36.5°C (97.7°F). His Glasgow Coma Scale score
was 14, and he had left facial weakness and a dense left hemiplegia. His
World Federation of Neurological Surgeons Scale (WFNSS) score was 3,
his Hunt and Hess Scale score was 3, and his modified Fisher Scale score
was 4.
A noncontrast head CT revealed a subarachnoid hemorrhage with a
large right frontotemporal intraparenchymal clot and intraventricular
hemorrhage (FIGURE 3-4A). He received immediate blood pressure–
lowering agents. A CT angiogram confirmed a suspected right middle
cerebral artery aneurysm (FIGURE 3-4B). A three-dimensional digital
subtraction angiogram revealed the complicated anatomy of this middle
cerebral artery aneurysm with multiple vessels coming off the aneurysm
and areas of irregular outpouching (FIGURE 3-4C). The patient was taken to
the operating room immediately for craniotomy, clipping of the
aneurysm, and clot evacuation. He was admitted to the neurocritical care
unit for routine postclipping subarachnoid hemorrhage care.
COMMENT This case shows a patient with a clear indication for surgical clipping due to
the more superficial location of the aneurysm at the distal middle cerebral
artery and the aneurysm’s anatomy with multiple vessels coming off the
aneurysm. It also shows how patients with large temporal hematomas (with
or without subarachnoid hemorrhage) should always undergo vessel
imaging, as a middle cerebral artery aneurysm may be the culprit.
1638 DECEMBER 2018

FIGURE 3-4
Ruptured right middle cerebral artery aneurysm requiring clipping, and subsequent cerebral
vasospasm and delayed cerebral ischemia, in the patient in CASE 3-3A. A, Noncontrast head CT
shows a large intraparenchymal component (thick arrows), small subarachnoid hemorrhage
(arrowheads), and intraventricular hemorrhage with hydrocephalus (thin arrows). B, Cerebral
angiogram reveals a right middle cerebral artery aneurysm (yellow arrow) as the culprit of
the hemorrhage. C, Three-dimensional cerebral angiogram shows the complicated anatomy
of this middle cerebral artery aneurysm with multiple vessels coming off the aneurysm and
areas of irregular outpouching (blue arrows).

postulated that the negative impact of SIRS on functional outcome is mediated in

part by nonconvulsive seizures.51
Additionally, patients with SAH are at risk for several additional neurologic
complications, including hydrocephalus, brain edema, delayed cerebral ischemia,
rebleeding, seizures, and neuroendocrine disorders, the latter of which can lead
to impaired regulation of sodium, volume, and glucose. Furthermore, mediated
through the hypothalamus, sympathetic release can result in cardiac and pulmonary
complications, including neurogenic ECG changes, arrhythmias, diminished
cardiac contractility (stress Takotsubo cardiomyopathy), troponin leaks, and
myocardial contraction band necrosis. The early recognition and treatment of these
complications is key to achieve the best possible outcome of the patient with SAH.
NEUROLOGIC COMPLICATIONS
Several serious neurologic complications may occur after SAH.
Rebleeding
Rebleeding is the most immediately life-threatening neurologic complication
after SAH. The best measure to reduce the risk of rebleeding is the early and rapid
treatment of the unsecured, ruptured aneurysm. The prevention of rebleeding
via aggressive blood pressure control should begin during the prehospital
transport and in the emergency department.
Hydrocephalus
Acute symptomatic hydrocephalus occurs in 20% of patients with SAH, usually
within minutes to days after SAH onset (FIGURE 3-1A and FIGURE 3-4A). Clinical
signs of hydrocephalus are decreased levels of consciousness, impaired upgaze,
hypertension, and delirium. The diagnosis is made by repeat head CT and
clinical symptoms.
Hydrocephalus can resolve spontaneously in 30% of patients but can also
rapidly worsen. Insertion of an external ventricular drain (EVD) can be
lifesaving. Some centers insert a lumbar drain instead of an EVD in cases of
communicating hydrocephalus, while some centers insert both. Reluctance to
place an EVD includes the risks of infection, bleeding (intracerebral or
intraventricular), and changes in the transmural pressure precipitating the
rebleeding of an unsecured aneurysm. The bleeding and infection risk for EVD
insertions are close to 8% for each.9
A rapid weaning of the EVD is recommended after aneurysm obliteration or
within 48 hours of insertion if the patient is neurologically stable. In those for
whom weaning is unsuccessful (approximately 40%), placement of a chronic
ventriculoperitoneal shunt may be required. A small retrospective study from
Germany has suggested that dexamethasone dosed 12 mg/d for at least 5 days
may lower the risk of hydrocephalus after SAH.52 Given the lack of randomized
controlled studies, the routine use of corticosteroids outside its application for
headache control from meningeal chemical irritation after SAH cannot
be recommended.
Seizures and Seizure Prophylaxis

Determining the true incidence of seizures in patients with SAH is difficult as
many patients (up to 26%) present with seizurelike episodes, but these episodes
are not easy to characterize as they occur at the onset of symptoms.13,24 If seizures
1640 DECEMBER 2018

occur prior to aneurysm securement, they are usually a sign of early rebleeding. KEY POINTS
Long-term epilepsy develops in 2% of patients with SAH and is correlated to
● The two most commonly
a higher severity of SAH.9 The occurrence of nonconvulsive seizures (7% to used clinical scales, the
18%) and nonconvulsive status epilepticus (3% to 13%) is more common in World Federation of
patients with SAH who are comatose and has been associated with delayed Neurological Surgeons Scale
cerebral ischemia and worse outcomes.51,53,54 It is not completely understood and the Hunt and Hess
Scale, are strong predictors
whether nonconvulsive seizures are the cause of delayed cerebral ischemia
of outcome in patients with
and worse outcomes or are an epiphenomenon of poor-grade SAH with subarachnoid hemorrhage.
outcomes due to the severity of SAH. Because nonconvulsive seizures are The most reliable and
treatable, continuous EEG monitoring should be considered in patients with validated radiologic scale is
the modified Fisher Scale.
high-grade SAH.54 A 2015 review summarized nonconvulsive seizures and
status epilepticus in SAH.54 However, at some institutions, the limitations lie ● For the treatment of
in the capacity of performing and interpreting prolonged continuous EEG aneurysm, endovascular
monitoring. Furthermore, recent discoveries show that surface EEG may detect coiling is preferred over
nonconvulsive seizures in only 8% of patients with SAH, while they were seen in surgical clipping whenever
possible, but the choice
38% of patients when intracortical depth electrodes were placed via a burr hole.53 of treatment depends on
Such depth electrode recording, however, is more invasive, is limited to very the patient’s age as well as
few centers, and is currently not considered standard of care. The suspicion the aneurysm’s location,
of nonconvulsive status epilepticus should be high in patients with SAH who morphology, and
relationship to adjacent
are comatose. vessels.
Currently, in the absence of randomized controlled trials of antiepileptic drug
treatment in patients with SAH, treatment with antiepileptic drugs should be ● Patients with
limited to the preaneurysm treatment time frame only, considering the known subarachnoid hemorrhage
are at risk for several
negative effects of anticonvulsants, particularly phenytoin, on neurocognitive
additional neurologic
recovery after SAH.55,56 complications, including
Guidelines and experts recommend stopping antiepileptics in patients in hydrocephalus, brain
whom a clinical examination can be followed reliably as soon as the aneurysm has edema, delayed cerebral
been secured and not to extend prophylaxis beyond 3 to 7 days unless the patient ischemia, rebleeding,
seizures, and
presented with a seizure at the onset of SAH.9,24 At the author’s institution, only neuroendocrine disorders,
patients who are comatose and patients with poor-grade SAH are continued on the latter of which can lead
antiepileptics after the aneurysm has been secured given the high risk for to impaired regulation
nonconvulsive seizures in these patients. of sodium, volume,
and glucose.
Levetiracetam has become a popular antiepileptic because of its high
bioavailability, favorable side effect profile, and lack of drug-drug interactions. ● Acute symptomatic
However, it should be noted that no studies have shown an advantage of hydrocephalus occurs in
levetiracetam over other antiepileptic drugs. In addition, levetiracetam has 20% of patients with
subarachnoid hemorrhage,
not been approved by the US Food and Drug Administration (FDA) for
usually within minutes to
monotherapy in epilepsy; therefore, no specific antiepileptic drugs can be days after subarachnoid
recommended for seizure prophylaxis in patients with SAH. hemorrhage onset. In cases
of hydrocephalus, insertion
of an external ventricular
Delayed Cerebral Ischemia
drain can be lifesaving.
Delayed cerebral ischemia is one of the most feared neurologic complications
after SAH, as cerebral infarction from delayed cerebral ischemia is the leading ● If seizures occur prior to
cause for morbidity in patients who survive the initial SAH. Monitoring for aneurysm securement, they
delayed cerebral ischemia is the main reason for the recommended prolonged are usually a sign of early
rebleeding.
ICU stay for patients with SAH. Delayed cerebral ischemia is defined as any
neurologic deterioration that persists for more than 1 hour and cannot be
explained by any other neurologic or systemic condition, such as fever, seizures,
hydrocephalus, sepsis, hypoxemia, sedation, and other metabolic causes
(CASE 3-3B).57 Delayed cerebral ischemia is diagnosed when other causes of

CASE 3-3B The patient in CASE 3-3A was admitted to the neurocritical care unit. He
was monitored with daily transcranial Doppler (TCD). On day 5 post–
subarachnoid hemorrhage (SAH), he had an uptrend in the daily TCD mean
velocities from 50 cm/s to 153 cm/s in the right middle cerebral artery
(FIGURE 3-5A–B).
On day 6 post-SAH, he developed worse headaches, disorientation,
neglect, dysarthria, and worse left-sided weakness. He was afebrile, had
a normal glucose level, and was euvolemic. He was given a fluid bolus,
and his systolic blood pressure goal was elevated to greater than
180 mm Hg with phenylephrine infusion.
An emergent EEG did not reveal nonconvulsive seizures. The blood
pressure augmentation improved the disorientation and neglect but not
the weakness and dysarthria. Four-vessel angiography was performed,
which revealed cerebral vasospasm in the right middle cerebral artery
and right anterior cerebral artery (FIGURE 3-5C–D). He was treated with
intraarterial nicardipine. Postprocedure he had resolution of symptoms
to his immediate post-SAH baseline, but they returned by the next
morning. Repeat four-vessel angiography was performed, and he was
treated with intraarterial nicardipine and angioplasty (FIGURE 3-5E).
He experienced complete resolution of his symptoms. He was
maintained on hypertensive and mild hypervolemic therapy until day 15
post-SAH. His TCD velocities were downtrending, and he was slowly
weaned off induced hypertension. He was eventually discharged to
rehabilitation.
COMMENT This case demonstrates how asymptomatic cerebral vasospasm on TCD

was initially appropriately treated with just euvolemia without induced
hypertension. However, given the elevated alert level of the providers,
when the patient worsened and developed symptomatic cerebral
vasospasm, treatment with induced hypertension and endovascular
treatment was swiftly initiated.
1642 DECEMBER 2018

FIGURE 3-5
Ruptured right middle cerebral artery aneurysm requiring clipping, and subsequent cerebral
vasospasm and delayed cerebral ischemia, in the patient in CASE 3-3B. Transcranial Doppler
reveals the baseline (on day 1 post–subarachnoid hemorrhage) waveform of the right middle
cerebral artery with a normal mean velocity of 50 cm/s (A). Transcranial Doppler reveals
the change in the waveform and increase in the mean velocity to 153 cm/s on day 5
post–subarachnoid hemorrhage (B). Note the change in the y axis between the two images
required to record the severe velocity elevation. Together with an elevated Lindegaard ratio of
5 (not shown), this indicates cerebral vasospasm. Cerebral angiogram performed on day 6
post–subarachnoid hemorrhage shows severe vasospasm of the middle cerebral artery and
anterior cerebral arteries (C, arrowheads), as compared to FIGURE 3-4B, and also reveals the delay
in distal perfusion (D, arrows) in the middle cerebral artery territory as a result of the vasospasm.
Cerebral angiogram after intraarterial vasodilator infusion and angioplasty shows the
posttreatment improvement of the middle cerebral artery vasospasm (E, arrowheads).

neurologic deterioration have been excluded or deemed insufficient to cause the

neurologic deterioration and is therefore a diagnosis of exclusion.
Historically, delayed cerebral ischemia was thought to be caused by cerebral
vasospasm. However, evidence now indicates that the pathophysiology of
delayed cerebral ischemia includes an interaction of early brain injury,
microthrombosis, cortical spreading depolarizations, related ischemia, and
cerebral vasospasm (FIGURE 3-6).58,59 Increasingly, some experts believe that
cerebral vasospasm is only an epiphenomenon and that the underlying biochemical
and biophysical changes that lead to delayed cerebral ischemia occur as early as at
SAH onset.58,59 This fundamental change in the approach to delayed cerebral
ischemia is supported by the negative endothelin 1 antagonist trials in patients
with SAH undergoing clipping or coiling.60,61
Endothelin 1 has been implicated to be the strongest vasoconstriction mediator
in SAH. However, the administration of clazosentan, a potent inhibitor of the
endothelin 1 receptor, resulted in less angiographic cerebral vasospasm yet did
not ameliorate delayed cerebral ischemia and did not lead to improvement in
outcomes 3 months after SAH.
Delayed cerebral ischemia occurs on average 3 to 14 days after SAH. The risk
for delayed cerebral ischemia increases with SAH thickness and intraventricular
hemorrhage, as demonstrated by the modified Fisher Scale. Additional risk
factors include poor clinical grade, loss of consciousness at ictus, cigarette
FIGURE 3-6
The pathophysiology of delayed cerebral ischemia.
Reprinted with permission from Macdonald RL, Nat Rev Neurol.59 © 2013 Springer Nature Limited.
1644 DECEMBER 2018

smoking, cocaine use, SIRS, hyperglycemia, hydrocephalus, and nonconvulsive KEY POINTS
seizures.59 Predicting who will develop delayed cerebral ischemia has proven
● The occurrence of
very difficult but is of great importance. Not only does such prediction have an nonconvulsive seizures (7%
impact on ICU monitoring, early recognition, and treatment, but also on resource to 18%) and nonconvulsive
allocation and early ICU discharge for low-grade, lower-risk patients with SAH. status epilepticus (3% to 13%)
The best predictors for patients requiring less frequent monitoring include older is more common in patients
with subarachnoid
age (older than 65 years of age), a low WFNSS score of 1 to 3, and a modified
hemorrhage who are
Fisher Scale score of less than 3 (TABLE 3-4).59 comatose and has been
associated with delayed
cerebral ischemia and
Delayed Cerebral Ischemia Prophylaxis worse outcomes.
Calcium channel blockers (nimodipine) and maintenance of normal intravascular
volume status have the strongest evidence of prophylactic interventions for the ● Continuous EEG
monitoring should be
prevention of delayed cerebral ischemia. Nimodipine (60 mg every 4 hours for
considered in patients with
21 days) is neuroprotective and has Class I evidence for decreasing the risk of high-grade subarachnoid
poor functional outcome.13,24,59 Notably, however, it does not decrease the hemorrhage.
frequency of angiographic vasospasm. A common side effect of nimodipine
is hypotension, which may lead to hypoperfusion and decreased cerebral ● In the absence of
randomized controlled trials
perfusion pressure. Therefore, to prevent hypotension, a dose reduction with of antiepileptic drug
an increase in frequency to 30 mg every 2 hours may be necessary. treatment in subarachnoid
In all cases, adequate maintenance of intravascular euvolemia is hemorrhage, but with known
recommended.13,24 Decreased intravascular volume and a negative fluid balance negative effects of
anticonvulsants, particularly
have been associated with a higher incidence of delayed cerebral ischemia and phenytoin, on
poor neurologic outcomes.59 neurocognitive recovery
How to monitor for euvolemia has not been defined. Trending the central after subarachnoid
venous pressure has fallen out of favor as it has been shown to be a poor predictor hemorrhage, treatment
with antiepileptic drugs
of fluid responsiveness and intravascular volume.62 Measurements of pulse should be limited to the
pressure variation or respiratory variability of the inferior vena cava diameter preaneurysm treatment
using point-of-care bedside ultrasound are easy to perform and are much more time frame only.
reliable monitoring techniques for fluid responsiveness of patients who are
● Delayed cerebral
critically ill, including those with SAH.62 Prophylactic hypervolemia must be
ischemia after subarachnoid
avoided, as this strategy has not been shown to improve cerebral blood flow or hemorrhage is defined as
decrease the frequency of cerebral vasospasm or delayed cerebral ischemia but any neurologic deterioration
increases adverse cardiopulmonary complications. that persists for more than
Maintenance of a euvolemic state may be difficult in the presence of cerebral 1 hour and cannot be
explained by any other
salt wasting, a common neuroendocrine disorder in SAH (see the following neurologic or systemic
section on hyponatremia). In patients with SAH and significant diuresis and condition.
natriuresis, additional administration of fludrocortisone can be helpful in
maintaining intravascular volume and normal sodium values (fludrocortisone ● Delayed cerebral ischemia
occurs on average 3 to
0.2 mg to 0.4 mg enterally every 12 hours) (TABLE 3-4). 14 days after subarachnoid
hemorrhage. The risk for
delayed cerebral ischemia
Delayed Cerebral Ischemia Diagnosis and Monitoring increases with subarachnoid
Diagnosing delayed cerebral ischemia is not easy. The combination of neurologic hemorrhage thickness and
examination and imaging studies can enhance the early detection and proper intraventricular hemorrhage,
management. Admission to the neurocritical care unit with frequent neurologic as demonstrated by the
modified Fisher Scale.
examination by experienced nurses and providers every 1 to 2 hours is necessary.
Delayed cerebral ischemia should be suspected if patients with SAH develop a
focal or global neurologic deficit or have a decrease of 2 or more points on the
Glasgow Coma Scale that lasts for at least 1 hour and cannot be explained by
another cause.

Experts have recommended that all patients with SAH undergo a head
CT at 24 to 48 hours after aneurysm treatment to establish any treatment-
related infarctions.57 Any subsequent new hypodensities not attributable to
EVD insertion or intraparenchymal hematoma should be regarded as cerebral
infarctions from delayed cerebral ischemia regardless of the clinical signs.
Patients with SAH should undergo physiologic or imaging monitoring
routinely during the risk period for delayed cerebral ischemia.13,24 Such
monitoring is usually multimodal and includes ICP, cerebral perfusion pressure,
continuous EEG, and transcranial Doppler (TCD) monitoring; DSA, CTA, and
CT perfusion (CTP) imaging are also used when indicated as well as brain tissue
oxygenation and microdialysis monitoring, when available.
TCD has been the longest used and best studied of all the monitoring
modalities. In the large vessels of the circle of Willis, TCD has adequate
sensitivity and specificity to detect increased cerebral blood flow velocities
secondary to cerebral vasospasm but is highly dependent on the operator and
cranial bone window (CASE 3-3B).63
Practitioners need to be aware that the sensitivity/specificity of TCD is good
for the middle cerebral and internal carotid arteries but is much lower for the
anterior cerebral arteries and posterior circulation arteries. Thresholds for the
diagnosis of cerebral vasospasm have been summarized.63
In addition, cerebral blood flow velocities can be elevated for other reasons
(hyperemia due to fever, induced hypertension, anemia), and therefore a
diagnosis of cerebral vasospasm should only be made when the ratio of mean
cerebral blood flow velocity of the intracranial vessel to mean cerebral blood flow
velocity of the extracranial internal carotid artery is elevated. Therefore, for the
diagnosis of middle cerebral artery vasospasm, routine measurement of the
Lindegaard ratio (mean velocity in the middle cerebral artery/mean velocity in
ipsilateral extracranial internal carotid artery) is prudent. A Lindegaard ratio
of >3 indicates cerebral vasospasm. Similar ratios exist for the other main
intracranial vessels.
DSA remains the gold standard for the detection of large- and middle-sized
artery vasospasm. CTA is now widely available and is often applied for
vasospasm screening before DSA given its high degree of specificity and lack of
invasiveness. CTA, however, can overestimate cerebral vasospasm. CTP imaging
with elevated mean transit time may be of additional value to CTA to assess for
decreased cerebral perfusion, but further investigations on the application of
CTP in SAH are needed.
Brain tissue oxygenation, cerebral blood flow, and microdialysis monitoring
can provide additional information when used in the context of multimodality
monitoring and may be able to detect early cerebral vasospasm before it becomes
symptomatic and before delayed cerebral ischemia occurs. Clinicians must
bear in mind the limitations of such monitoring, including the restriction to
monitoring local rather than global brain areas.
Continuous scalp EEG offers the advantage of monitoring broader regions of
the brain. Quantitative continuous EEG, if available, may offer easier
interpretation of bedside data even by providers not trained or certified to
interpret EEG. The cost, and thereby lack of widespread availability, is currently
limiting quantitative continuous EEG from becoming standard of care.
It is very important to differentiate between angiographic/TCD vasospasm
and clinical symptomatic vasospasm (CASE 3-3B).64 The former occurs in the
1646 DECEMBER 2018

majority of patients with SAH (70%) but has not been associated with outcome KEY POINTS
after SAH. Only symptomatic vasospasm, occurring in 30% of patients with
● Delayed cerebral
SAH, has been associated with delayed cerebral ischemia and poor outcome after ischemia should be
SAH.64 Given the risks of endovascular cerebral vasospasm treatment, experts suspected if patients with
recommend such treatment only for patients with symptomatic vasospasm, subarachnoid hemorrhage
while angiographic/TCD vasospasm should be managed with a careful watch develop a focal or global
neurologic deficit or have a
and wait approach with a very low threshold to trigger DSA and endovascular
decrease of 2 or more points
treatment.13,24 on the Glasgow Coma Scale
Some variability exists regarding the timing and frequency of the application that lasts for at least 1 hour
of the various monitoring modalities. At a minimum, the care for patients with and cannot be explained by
any other cause.
SAH should be protocolized using a written protocol and an algorithm. Patients
with SAH should be admitted to a neurocritical care unit and have their ● Patients with
aneurysm secured as quickly as possible, preferably within the first 6 to 12 hours subarachnoid hemorrhage
after presentation. should undergo physiologic
Monitoring in the neurocritical care unit includes daily TCD monitoring, or imaging monitoring
routinely during the risk
although in low-risk patients it may be sufficient to monitor every other day as period for delayed cerebral
long as neurologic monitoring can be performed every 1 to 2 hours. Some centers ischemia.
perform CTA/CTP or DSA routinely for all patients 5 to 7 days after admission to
screen for cerebral vasospasm. Most centers, however, perform these tests only if ● Transcranial Doppler has
adequate sensitivity and
concern arises for symptomatic vasospasm.
specificity to detect
In patients who are comatose or obtunded, it may be difficult to obtain a increased cerebral blood
reliable examination, and therefore TCD examination and, in addition, flow velocities secondary to
CTA/CTP or DSA for vasospasm screening may be required. Particularly in cerebral vasospasm in the
middle cerebral and basilar
patients with high-grade SAH with poor examination findings, diagnosis and
arteries but is highly
treatment initiation of delayed cerebral ischemia may be difficult, somewhat dependent on the operator
subjective, and mostly based on neuromonitoring findings. and cranial bone window.
At the author’s institution, the SAH treatment protocol dictates induced
hypertension, CT/CTA, and DSA when these patients have elevated TCD mean ● Digital subtraction
angiography remains the
velocities and elevated Lindegaard (or other) ratios (CASE 3-3B). Other centers gold standard for detection
use information from multimodality monitoring (continuous EEG or even of large- and middle-sized
intracortical depth EEG, brain tissue oxygenation, microdialysis, cerebral blood flow artery vasospasm.
monitoring) to decide if hypoperfusion or vasospasm may be occurring and when
● Only symptomatic
to send the patient for DSA. While randomized controlled data are lacking on the
vasospasm, occurring in
value of multimodality monitoring on short-term and long-term outcomes in the 30% of patients with
treatment of patients with SAH, resources at the various institutions may dictate, subarachnoid hemorrhage,
at least in part, how much advanced multimodality monitoring can be employed. has been associated with
delayed cerebral ischemia
and poor outcome after
Management subarachnoid hemorrhage.
At the author’s institution, all patients are treated with nimodipine and euvolemia.
Low-risk patients whose neurologic examination, TCD, and, if performed, CTA ● Hypervolemic,
remain unchanged are transferred to the neuro step-down unit under the care of hypertensive, and
hemodilutional (Triple H)
neurointensivists between days 8 and 10 and receive neurologic examinations less therapy is no longer
frequently (every 2 hours instead of hourly) than in the ICU. Patients are then supported by guidelines
discharged to the floor or home by day 14. High-risk patients whose neurologic because of the existing
examination, TCD, and, if performed, CTA remain unchanged will transfer to the evidence of adverse
associations with outcomes
floor or other lower level of care 14 days after SAH. If at any time the patient after the use of
develops elevated TCD mean cerebral blood flow velocities and abnormal CTA hemodilution. Standard
findings, the intensity of neurologic monitoring is escalated (CASE 3-3B). treatment is now
If patients experience a neurologic deterioration suggestive of delayed cerebral hypertensive and mild
hypervolemic therapy (HHT).
ischemia, certain rescue therapies are initiated. In the case of symptomatic

cerebral vasospasm and delayed cerebral ischemia, induced hypertension is

indicated per current guidelines (TABLE 3-4 and FIGURE 3-6). Hypervolemic,
hypertensive, and hemodilutional (Triple H) therapy is no longer supported
by guidelines because of the existing evidence of adverse associations with
outcomes after the use of hemodilution,13,24,65 and the standard treatment is now
hypertensive and mild hypervolemic therapy (HHT). Many institutions initiate
an IV fluid bolus (1 L to 2 L of 0.9% saline) and maintain fluids for euvolemia
or mild hypervolemia. Hypertension is preferably induced using α1 receptor
agonists by a continuous infusion (norepinephrine or phenylephrine). This
group of drugs are the vasopressors of choice in SAH, as the brain vessels lack α1
receptors, and therefore only systemic but not brain vasoconstriction is achieved.
Blood pressure augmentation should progress in a stepwise fashion with frequent
neurologic assessments at each step.
At the author’s institution, a mean arterial pressure (MAP) about 20 mm Hg
above the baseline MAP is set as the first goal (often MAP >90 mm Hg). Some
institutions use systolic blood pressure goals instead of MAP goals, and no evidence
exists to guide clinicians regarding whether one parameter is better than the other.
In institutions where systolic blood pressure goals are set for induced hypertension,
the initial goal for induced hypertension should be approximately 20 mm Hg to
40 mm Hg above the baseline systolic blood pressure. This commonly results in a
systolic blood pressure goal of >180 mm Hg or >200 mm Hg.
If the clinical examination has returned to the prior baseline examination, no
further blood pressure elevations are necessary, unless the clinical examination
deteriorates further at this blood pressure goal. In the latter case, further
increases in the blood pressure goal should be attempted. While no optimal or
maximum blood pressure goal is known, adverse effects on the cardiac and
pulmonary systems and the brain (for example, autoregulatory side effects with
elevated ICP with increasing MAP or posterior reversible encephalopathy
syndrome [PRES]) should be considered for each patient.
At the author’s institution, inotropic agents (milrinone, dobutamine) are
reserved for those patients with known poor cardiac output from acute or
chronic cardiomyopathy. If neurologic deficits persist despite induced
hypertension, the patient is sent for CT/CTA followed by DSA with endovascular
therapy if cerebral vasospasm is confirmed. A noncontrast head CT prior to DSA
is useful in ruling out hydrocephalus and determining preexisting stroke prior to
endovascular treatment. If other causes for the neurologic deterioration have
already been ruled out, and TCDs suggest cerebral vasospasm, the CTA may be
skipped to spare the patient radiation, iodine contrast, and to save time and allow
the patient to go to DSA immediately for treatment. Endovascular therapy using
intraarterial vasodilators (nicardipine, milrinone, verapamil) or angioplasty is
supported by prospective and retrospective observational data and is
recommended by guidelines (TABLE 3-4 and CASE 3-3B).13,24 At the author’s
institution, prophylactic angioplasty is not performed if cerebral vasospasm is
detected during TCD or CTA without neurologic deterioration because this
practice is associated with higher complication rates.13,24
MEDICAL COMPLICATIONS
SAH is a systemic disease, and patients commonly experience additional medical
complications. Anticipation of these complications leads to rapid recognition
and treatment.
1648 DECEMBER 2018

Cardiopulmonary KEY POINTS
Cardiopulmonary dysfunction is a well-known complication of SAH and can
● In the management of
range from minor ECG changes to severe stress cardiomyopathy and neurogenic patients with symptomatic
pulmonary edema. The incidence of left ventricular dysfunction in the first week vasospasm, hypertension is
after SAH ranges from 9% to 30%66,67 and can include regional wall motion preferably induced using α1
abnormalities not correlating with coronary artery territories or severe systolic receptor agonists by a
continuous infusion
left ventricular dysfunction with an ejection fraction of less than 30%. ECG
(norepinephrine or
changes can include T-wave inversions and prolonged QTc intervals and may be phenylephrine).
the culprit for arrhythmias such as bradycardia, atrial fibrillation, ventricular
tachycardia, and ventricular fibrillation. The severity of SAH is an independent ● Cardiopulmonary
predictor of cardiopulmonary injury, suggesting that the cardiopulmonary injury dysfunction is a well-
known complication of
is neurally mediated.68 Troponin elevation can be seen in up to 30% of patients, subarachnoid hemorrhage
but its significance is unclear. Certain ECG changes, such as prolonged QTc and can range from minor
intervals, have been reported to predict death.69 ECG changes to severe
Similarities have been observed between pheochromocytoma crisis and SAH, stress cardiomyopathy
and neurogenic
linking the observed cardiac changes to a catecholamine surge. Patients with SAH pulmonary edema.
can have a threefold increase in norepinephrine levels for 10 days or longer after
ictus, which normalize after this time period.70 Myocardial cell necrosis, also known ● The severity of
as contraction band necrosis, is the hallmark of a catecholamine surge and can be subarachnoid hemorrhage is
an independent predictor of
found in patients with pheochromocytoma and SAH.71 The central catecholamine
cardiopulmonary injury,
release has been localized to the posterior hypothalamus based on postmortem suggesting that the
pathologic studies,72 which found microscopic hypothalamic lesions including cardiopulmonary injury is
small hemorrhages and infarctions in patients with contraction band necrosis. neurally mediated.
Clinically, the catecholamine surge during aneurysm rupture results in direct
● Takotsubo
myocardial injury, leading to decreased inotropy and an increase in cardiac cardiomyopathy in
preload due to venous constriction, as well as increased cardiac afterload due to subarachnoid hemorrhage is
peripheral arterial constriction (FIGURE 3-7).73 Consequently, stroke volume associated with higher
diminishes, which cannot be compensated by reflex tachycardia, resulting in mortality and worse
long-term outcomes.
decreased cardiac output and neurocardiogenic shock. Because of loss of
myocardial compliance (“stunning” of the heart), the cardiac silhouette on a
ventriculogram and on chest radiograph has the characteristic shape of a
Japanese octopus fishing pot (tako-tsubo), which is why this disease has also been
named Takotsubo cardiomyopathy. Neurogenic cardiomyopathy in SAH is
associated with higher mortality and worse long-term outcomes.74
Pulmonary edema leading to hypoxia is also frequently encountered and
may occur either as a result from the acute left ventricular dysfunction or
independently as neurogenic pulmonary edema from substantial increases in
pulmonary capillary pressures from the sympathetic surge. FIGURE 3-7
summarizes the pathophysiology of cardiopulmonary complications mediated by
alpha and beta receptors. Cardiopulmonary complications after SAH are usually
transient and resolve within several days to 2 weeks. However, during this
period, the patient must be maximally supported to prevent secondary brain
injury from hypoxia and decreased cerebral perfusion. In extreme cases, the
insertion of an intraaortic balloon pump may be required to support the patient
until resolution of the transient symptoms.
At the author’s institution, every patient with SAH receives a baseline ECG,
echocardiogram, and chest x-ray on admission. Excessive fluid intake is avoided
with a goal of euvolemia and not hypervolemia. Lung-protective mechanical
ventilation with tidal volumes of less than 7 cc/kg of ideal body weight without
permissive hypercarbia is performed routinely. If delayed cerebral ischemia is

FIGURE 3-7
The pathophysiology of cardiopulmonary complications in subarachnoid hemorrhage (SAH).
SAH leads to a sudden catecholamine surge, which activates alpha, alpha + beta, and beta
receptors, leading to pulmonary and myocardial dysfunction as well as platelet aggregation.
Consequently, patients can develop neurogenic pulmonary edema, left ventricular
dysfunction (Takotsubo cardiomyopathy), and shock.
LVEF = left ventricular ejection fraction.
Modified with permission from Bassil R, et al.73 © 2017 Wolters Kluwer.
present and induced hypertension is initiated, inotropic agents may be used in

addition to vasopressors to increase cardiac contractility. Hypervolemia is
restricted to avoid or prevent worsening of pulmonary edema. A repeat
follow-up echocardiogram is performed 10 to 14 days after ictus to evaluate for
resolution of Takotsubo cardiomyopathy.
Fever
Fever is the most common medical complication after SAH, occurring in up to
70% of patients.13,24 Fever is more likely to occur in patients with high-grade
SAH and poor neurologic status. Fever has been associated with delayed cerebral
ischemia and worse clinical outcomes and is likely related to SIRS and chemical
meningitis rather than an infectious process. While fever is commonly treated
with therapeutic temperature modulation and induced normothermia, no clear
evidence currently indicates that such treatment is beneficial. Current
recommendations are to monitor body temperature and to rule out or treat
infectious etiologies. If fever is suppressed with induced normothermia,
shivering should be strictly avoided and aggressively treated if it occurs.
1650 DECEMBER 2018

Thromboembolism and Prophylaxis KEY POINTS
Deep vein thrombosis after SAH is common, with rates between 2% and 20%
● Fever is the most common
depending on the screening method. Patients with high-grade SAH are at medical complication after
greatest risk, presumably because of limited mobility. To prevent the potential subarachnoid hemorrhage,
life-threatening consequences of pulmonary embolism, mechanical venous occurring in up to 70%
thromboembolism prophylaxis should be initiated immediately on admission of patients.
with the use of pneumatic compression devices. At the author’s institution,
● Fever has been
chemoprophylaxis with subcutaneous fractionated or unfractionated heparin is associated with delayed
usually initiated immediately after endovascular aneurysm repair and within cerebral ischemia and worse
24 hours after craniotomy for clipping.13,24 Heparin-induced thrombocytopenia clinical outcomes and is
type II affects 6% of patients with SAH.75 The exact mechanism for this likely related to systemic
inflammatory response
observation is not clear. However, a clinical suspicion of this syndrome should syndrome and chemical
immediately be raised and diagnostic measures and treatment should be initiated meningitis, rather than an
when the platelet counts are decreasing rapidly and the patient has been exposed infectious process.
to fractionated or unfractionated heparin.
● Deep vein thrombosis
after subarachnoid
Glycemic Dysfunction hemorrhage is common,
with rates between 2%
Glycemic dysfunction is very common after SAH because of stress and has and 20%.
been associated with delayed cerebral ischemia and poor clinical outcome.
However, it remains unclear whether this is merely an association or causative. ● Mechanical venous
Hypoglycemia can lead to brain metabolic crisis and must be vigilantly avoided. thromboembolism
prophylaxis should be
In the absence of clinical trials of glucose control in patients with SAH, current initiated immediately on
recommendations are to maintain a blood glucose level between 80 mg/dL and admission with the use of
200 mg/dL (TABLE 3-4).13,24 pneumatic compression
devices. At the
author’s institution,
Hyponatremia chemoprophylaxis with
Hyponatremia is the most common electrolyte disorder in SAH and can occur in subcutaneous fractionated
or unfractionated heparin is
up to 30% of patients. Its cause is presumed to be hypothalamic dysfunction, usually initiated immediately
most commonly from cerebral salt wasting due to an increase in circulating brain after endovascular
natriuretic peptide levels. The syndrome of inappropriate secretion of aneurysm repair and within
antidiuretic hormone (SIADH) should always be considered but is generally 24 hours after craniotomy
for clipping.
uncommon in patients with SAH. Knowledge about how to differentiate cerebral
salt wasting and SIADH is a basic and very important skill of any clinician caring ● In the absence of clinical
for patients with SAH. Reviews of the diagnoses of both syndromes have been trials of glucose control in
published (FIGURE 3-8).76,77 patients with subarachnoid
hemorrhage, current
It is important to note that in both cerebral salt wasting and SIADH, the
recommendations are to
laboratory findings are similar: low serum sodium (<134 mEq/mL), low serum maintain a blood glucose
osmolality (<274 mmol/L), high urine sodium (>20 mmol/L), and high urine level between 80 mg/dL and
osmolality (>100 mmol/L). The only differentiating finding is the patient’s 200 mg/dL.
intravascular volume status; cerebral salt wasting is a hypovolemic state, while
● Hyponatremia is the most
patients with SIADH are euvolemic or even hypervolemic. It is of utmost common electrolyte
importance to correctly differentiate these two syndromes because treatment is disorder in patients with
opposite and an incorrect diagnosis with improper treatment can lead to subarachnoid hemorrhage
detrimental effects in patients with SAH. and can occur in up to 30%
of patients.
Cerebral salt wasting is treated with fluid administration and sometimes a
continuous infusion of hypertonic saline (1.5% to 3%) and fludrocortisone if
diuresis and natriuresis impede maintenance of adequate volume status. However,
patients with SIADH are treated with fluid restriction and sometimes diuresis with
loop diuretics. Serum sodium as well as the volume status must be followed closely.

FIGURE 3-8
Differentiating different types of hyponatremia. The optimal way to differentiate different
etiologies for hyponatremia is to follow this flow chart as follows: serum sodium followed by
serum osmolality followed by urine sodium followed by urine osmolality (bottom to top).
There is no difference in the laboratory findings between cerebral salt wasting syndrome
(CSWS) and syndrome of inappropriate secretion of antidiuretic hormone (SIADH), and the
only difference between both common hyponatremia syndromes in subarachnoid
hemorrhage is volume status. Cerebral salt wasting is a hypovolemic state, and SIADH is a
euvolemic or hypervolemic state. Hence, treatment strategies are opposite: Cerebral salt
wasting is treated with fluids with or without fludrocortisone, and SIADH is treated with fluid
restriction with or without diuresis.
Modified with permission from Stiefel D, et al, Neurocrit Care.76 © 2007 Humana Press.
As shown in FIGURE 3-8, it is also important to test thyroid and adrenal functions
as well as serum glucose (>500 mg/dL can result in pseudohyponatremia) and
triglyceride levels to adequately address other causes of hyponatremia.
Pseudohyponatremia (not due to true hyponatremia but from laboratory test
interference from extreme triglyceridemia) should be considered in patients
on propofol.
Anemia
Most patients with SAH will experience anemia during their hospitalization, which
is presumably due to excessive blood draws, blood loss from other reasons, or
systemic inflammation.23 Anemia and hemoglobin concentrations of less than
9 g/dL have been associated with delayed cerebral ischemia and poor clinical
1652 DECEMBER 2018

outcomes; however, optimal hemoglobin goal levels and transfusion thresholds are KEY POINTS
not known.78,79 A recent small, randomized controlled trial comparing the safety
● The laboratory findings
and efficacy of transfusion to higher hemoglobin levels (goal hemoglobin are similar in both cerebral
concentration of at least 10 g/dL or 11.5 g/dL) in patients with SAH demonstrated salt wasting and syndrome
safety of transfusing to these high hemoglobin levels, but no differences were seen of inappropriate secretion
in delayed cerebral ischemia and short-term (14 days) functional outcomes.79 of antidiuretic hormone. The
only differentiating finding is
the patient’s intravascular
PROGNOSTICATION AFTER SUBARACHNOID HEMORRHAGE volume status; cerebral salt
While established clinical scales such as the WFNSS and the Hunt and Hess Scale wasting is a hypovolemic
may be helpful in discriminating high-risk from low-risk patients, just like any state, while patients with
syndrome of inappropriate
prognostic scale, they were established for populations and must not be applied
secretion of antidiuretic
to individual patients. hormone are euvolemic or
Outcome is also influenced by many other factors, which are generally not even hypervolemic. It is of
included in prognostic scales, such as patient values and preferences, utmost importance to
comorbidities, social networks, resilience, and time for recovery. Unless the correctly differentiate these
two syndromes because
patient presents with bilaterally dilated pupils and a head CT or DSA inconsistent treatment is opposite.
with brain perfusion and life, all efforts should be made initially to salvage even
patients with very high-grade SAH with treatment of high ICP and ● Cerebral salt wasting is
hydrocephalus, followed by securing the aneurysm. Adequate protocolized treated with fluid
administration and
neurocritical care should be provided for the first 2 weeks or longer, at which sometimes a continuous
time, after establishing preexisting patient wishes and preferences with the infusion of hypertonic saline
family, further goals of care may need to be discussed. and fludrocortisone if
The Functional Recovery Expected After Subarachnoid Hemorrhage (FRESH) diuresis and natriuresis
impede maintenance of
scale is a recently published prognostic scale for expected 12-month cognitive
adequate volume status.
outcome and quality of life after SAH, and the same caution should be used with Patients with syndrome of
this scale as should be used with any prognostic scale.80 This score has excellent inappropriate secretion of
discrimination and calibration, has been externally validated, and may be antidiuretic hormone are
treated with fluid restriction
calculated at the bedside using a free smartphone app.81
and sometimes diuresis with
loop diuretics.
CONCLUSION ● Anemia and hemoglobin

SAH is a life-threatening type of hemorrhagic stroke and a neurologic emergency concentrations of less than
9 g/dL have been associated
that carries a high risk of morbidity and mortality. This female-predominant with delayed cerebral
disease should be approached in two phases: (1) expeditious and accurate ischemia and poor clinical
recognition of SAH, transfer to an appropriate high-volume center, and outcomes; however, optimal
identification and securement of the aneurysm and (2) observation, prevention, hemoglobin goal levels and
transfusion thresholds are
or swift treatment of medical and neurologic complications in a neurocritical care not known.
unit with state of the art neurocritical care adhering to established guidelines.
The main neurologic complications include hydrocephalus, seizures, brain
edema, and delayed cerebral ischemia. Common medical complications
encompass cardiopulmonary complications, neuroendocrine disorders, and
fever, which require expert care.
REFERENCES
1 Go AS, Mozaffarian D, Roger VL, et al. Heart 2 Lovelock CE, Rinkel GJ, Rothwell PM. Time trends
disease and stroke statistics–2014 update: a in outcome of subarachnoid hemorrhage:
report from the American Heart Association. population-based study and systematic review.
Circulation 2014;129(3):e28–e292. doi:10.1161/ Neurology 2010;74(19):1494–1501. doi:10.1212/
01.cir.0000441139.02102.80. WNL.0b013e3181dd42b3.

3 Nieuwkamp DJ, Setz LE, Algra A, et al. Changes in 15 Brown RD Jr, Broderick JP. Unruptured
case fatality of aneurysmal subarachnoid intracranial aneurysms: epidemiology, natural
haemorrhage over time, according to age, sex, history, management options, and familial
and region: a meta-analysis. Lancet Neurol 2009; screening. Lancet Neurol 2014;13(4):393–404.
8(7):635–642. doi:10.1016/S1474-4422(09)70126-7. doi:10.1016/S1474-4422(14)70015-8.
4 Feigin VL, Lawes CM, Bennett DA, et al. 16 Magnetic Resonance Angiography in Relatives of
Worldwide stroke incidence and early case Patients with Subarachnoid Hemorrhage Study
fatality reported in 56 population-based studies: Group. Risks and benefits of screening for
a systematic review. Lancet Neurol 2009;8(4): intracranial aneurysms in first-degree relatives
355–569. doi:10.1016/S1474-4422(09)70025-0. of patients with sporadic subarachnoid
hemorrhage. N Engl J Med 1999;341(18):1344–1350.
5 Al-Khindi T, Macdonald RL, Schweizer TA.
doi:10.1056/NEJM199910283411803.
Cognitive and functional outcome after
aneurysmal subarachnoid hemorrhage. 17 Ronkainen A, Hernesniemi J, Puranen M, et al.
Stroke 2010;41(8):e519–e536. doi:10.1161/ Familial intracranial aneurysms. Lancet 1997;
STROKEAHA.110.581975. 349(9049):380–384. doi:10.1016/S0140-
6736(97)80009-8.
6 Tjahjadi M, Heinen C, König R, et al. Health-
related quality of life after spontaneous 18 Bor AS, Rinkel GJ, van Norden J, Wermer MJ.
subarachnoid hemorrhage measured in a recent Long-term, serial screening for intracranial
patient population. World Neurosurg 2013;79(2): aneurysms in individuals with a family history
296–307. doi:10.1016/j.wneu.2012.10.009. of aneurysmal subarachnoid haemorrhage: a
cohort study. Lancet Neurol 2014;13(4):385–392.
7 Taufique Z, May T, Meyers E, et al. Predictors of
doi:10.1016/S1474-4422(14)70021-3.
poor quality of life 1 year after subarachnoid
hemorrhage. Neurosurgery 2016;78(2):256–264. 19 Thompson BG, Brown RD Jr, Amin-Hanjani S, et al.
doi:10.1227/NEU.0000000000001042. Guidelines for the management of patients with
unruptured intracranial aneurysms: a guideline
8 Taylor TN, Davis PH, Torner JC, et al. Lifetime cost
for healthcare professionals from the American
of stroke in the United States. Stroke 1996;27(9):
Heart Association/American Stroke Association.
1459–1466. doi:10.1161/01.STR.27.9.1459.
Stroke 2015;46(8):2368–2400. doi:10.1161/
9 Macdonald RL, Schweizer TA. Spontaneous STR.0000000000000070.
subarachnoid haemorrhage. Lancet 2017;
20 Korja M, Kaprio J. Controversies in epidemiology
389(10069):655–666. doi:10.1016/S0140-
of intracranial aneurysms and SAH. Nat Rev
6736(16)30668-7.
Neurol 2016;12(1):50–55. doi:10.1038/nrneurol.
10 Muehlschlegel S, Kursun O, Topcuoglu MA, 2015.228.
et al. Differentiating reversible cerebral
21 Korja M, Silventoinen K, McCarron P, et al.
vasoconstriction syndrome with subarachnoid
Genetic epidemiology of spontaneous
hemorrhage from other causes of subarachnoid
subarachnoid hemorrhage: Nordic Twin Study.
hemorrhage. JAMA Neurol 2013;70(10):1254–1260.
Stroke 2010;41(11):2458–2462. doi:10.1161/
doi:10.1001/jamaneurol.2013.3484.
STROKEAHA.110.586420.
11 Andreasen TH, Bartek J Jr, Andresen M, et al.
22 Yasuno K, Bilguvar K, Bijlenga P, et al. Genome-wide
Modifiable risk factors for aneurysmal
association study of intracranial aneurysm identifies
subarachnoid hemorrhage. Stroke 2013;44(12):
three new risk loci. Nat Genet 2010;42(5):420–425.
3607–3612. doi:10.1161/STROKEAHA.113.001575.
doi:10.1038/ng.563.
12 Suarez JI. Diagnosis and management of
23 Edlow JA, Malek AM, Ogilvy CS. Aneurysmal
subarachnoid hemorrhage. Continuum (Minneap
subarachnoid hemorrhage: update for emergency
Minn) 2015;21(5 Neurocritical Care):1263–1287.
physicians. J Emerg Med 2008;34(3):237–251.
doi:10.1212/CON.0000000000000217.
doi:10.1016/j.jemermed.2007.10.003.
13 Connolly ES Jr, Rabinstein AA, Carhuapoma JR,
24 Diringer MN, Bleck TP, Claude Hemphill J 3rd,
et al. Guidelines for the management of
et al. Critical care management of patients
aneurysmal subarachnoid hemorrhage: a
following aneurysmal subarachnoid hemorrhage:
guideline for healthcare professionals from
recommendations from the Neurocritical Care
the American Heart Association/American
Society's Multidisciplinary Consensus
Stroke Association. Stroke 2012;43(6):1711–1737.
Conference. Neurocrit Care 2011;15(2):211–240.
doi:10.1161/STR.0b013e3182587839.
doi:10.1007/s12028-011-9605-9.
14 Greving JP, Wermer MJ, Brown RD Jr, et al.
25 Molyneux A, Kerr R, International Subarachnoid
Development of the PHASES score for
Aneurysm Trial (ISAT) Collaborative Group, et al.
prediction of risk of rupture of intracranial
International Subarachnoid Aneurysm Trial (ISAT)
aneurysms: a pooled analysis of six prospective
of neurosurgical clipping versus endovascular
cohort studies. Lancet Neurol 2014;13(1):59–66.
coiling in 2143 patients with ruptured intracranial
doi:10.1016/S1474-4422(13)70263-1.
aneurysms: a randomized trial. J Stroke
Cerebrovasc Dis 2002;11(6):304–314. doi:10.1053/
jscd.2002.130390.
1654 DECEMBER 2018

26 Molyneux AJ, Kerr RS, Yu LM, et al. International 37 van Gijn J, van Dongen KJ, Vermeulen M, Hijdra A.
subarachnoid aneurysm trial (ISAT) of Perimesencephalic hemorrhage: a nonaneurysmal
neurosurgical clipping versus endovascular and benign form of subarachnoid hemorrhage.
coiling in 2143 patients with ruptured intracranial Neurology 1985;35(4):493–497. doi:10.1212/
aneurysms: a randomised comparison of effects WNL.35.4.493.
on survival, dependency, seizures, rebleeding,
38 Edlow JA, Figaji A, Samuels O. Emergency
subgroups, and aneurysm occlusion. Lancet
neurological life support: subarachnoid
2005;366(9488):809–817. doi:10.1016/S0140-6736
hemorrhage. Neurocrit Care 2015;23(suppl 2):
(05)67214-5.
S103–S109. doi:10.1007/s12028-015-0183-0.
27 Lawton MT, Vates GE. Subarachnoid
39 Tang C, Zhang TS, Zhou LF. Risk factors for
hemorrhage. N Engl J Med 2017;377(3):257–266.
rebleeding of aneurysmal subarachnoid
doi:10.1056/NEJMcp1605827.
hemorrhage: a meta-analysis. PLoS One 2014;
28 Linn FH, Wijdicks EF, van der Graaf Y, et al. 9(6):e99536. doi:10.1371/journal.pone.0099536.
Prospective study of sentinel headache in
40 Starke RM, Kim GH, Fernandez A, et al. Impact
aneurysmal subarachnoid haemorrhage.
of a protocol for acute antifibrinolytic therapy
Lancet 1994;344(8922):590–593. doi:10.1016/
on aneurysm rebleeding after subarachnoid
S0140-6736(94)91970-4.
hemorrhage. Stroke 2008;39(9):2617–2621.
29 Hassan A, Lanzino G, Wijdicks EF, et al. Terson's doi:10.1161/STROKEAHA.107.506097.
syndrome. Neurocrit Care 2011;15(3):554–558.
41 Hillman J, Fridriksson S, Nilsson O, et al.
doi:10.1007/s12028-011-9555-2.
Immediate administration of tranexamic acid and
30 de Oliveira Manoel AL, Mansur A, Murphy A, et al. reduced incidence of early rebleeding after
Aneurysmal subarachnoid haemorrhage from a aneurysmal subarachnoid hemorrhage: a
neuroimaging perspective. Crit Care 2014;18(6): prospective randomized study. J Neurosurg 2002;
557. doi:10.1186/s13054-014-0557-2. 97(4):771–778. doi:10.3171/jns.2002.97.4.0771.
31 Suarez JI, Tarr RW, Selman WR. Aneurysmal 42 Malekpour M, Kulwin C, Bohnstedt BN, et al.
subarachnoid hemorrhage. N Engl J Med 2006; Effect of short-term ε-aminocaproic acid
354(4):387–396. doi:10.1056/NEJMra052732. treatment on patients undergoing endovascular
coil embolization following aneurysmal
32 Nagy K, Skagervik I, Tumani H, et al. Cerebrospinal
subarachnoid hemorrhage. J Neurosurg 2017;
fluid analyses for the diagnosis of subarachnoid
126(5):1606–1613. doi:10.3171/2016.4.JNS152951.
haemorrhage and experience from a Swedish
study. What method is preferable when 43 Report of World Federation of Neurological
diagnosing a subarachnoid haemorrhage? Clin Surgeons Committee on a universal
Chem Lab Med 2013;51(11):2073–2086. doi:10.1515/ subarachnoid hemorrhage grading scale.
cclm-2012-0783. J Neurosurg 1988;68(6):985–986.
33 da Rocha AJ, da Silva CJ, Gama HP, et al. 44 Hunt WE, Hess RM. Surgical risk as related to time
Comparison of magnetic resonance imaging of intervention in the repair of intracranial
sequences with computed tomography to detect aneurysms. J Neurosurg 1968;28(1):14–20.
low-grade subarachnoid hemorrhage: role of doi:10.3171/jns.1968.28.1.0014.
fluid-attenuated inversion recovery sequence.
45 Fisher CM, Kistler JP, Davis JM. Relation of
J Comput Assist Tomogr 2006;30(2):295–303.
cerebral vasospasm to subarachnoid hemorrhage
34 McKinney AM, Palmer CS, Truwit CL, et al. visualized by computerized tomographic
Detection of aneurysms by 64-section scanning. Neurosurgery 1980;6(1):1–9.
multidetector CT angiography in patients acutely doi:10.1227/00006123-198001000-00001.
suspected of having an intracranial aneurysm
46 Frontera JA, Claassen J, Schmidt JM, et al.
and comparison with digital subtraction and 3D
Prediction of symptomatic vasospasm after
rotational angiography. AJNR Am J Neuroradiol
subarachnoid hemorrhage: the modified fisher
2008;29(3):594–602. doi:10.3174/ajnr.A0848.
scale. Neurosurgery 2006;59(1):21–27; discussion
35 Donmez H, Serifov E, Kahriman G, et al. 21–27. doi:10.1227/01.NEU.0000218821.34014.1B.
Comparison of 16-row multislice CT angiography
47 Rush B, Romano K, Ashkanani M, et al. Impact of
with conventional angiography for detection
hospital case-volume on subarachnoid
and evaluation of intracranial aneurysms. Eur J
hemorrhage outcomes: a nationwide analysis
Radiol 2011;80(2):455–461. doi:10.1016/j.ejrad.
adjusting for hemorrhage severity. J Crit Care
2010.07.012.
2017;37:240–243. doi:10.1016/j.jcrc.2016.09.009.
36 Kapadia A, Schweizer TA, Spears J, et al.
48 Varelas PN, Schultz L, Conti M, et al. The impact
Nonaneurysmal perimesencephalic subarachnoid
of a neuro-intensivist on patients with stroke
hemorrhage: diagnosis, pathophysiology, clinical
admitted to a neurosciences intensive care unit.
characteristics, and long-term outcome. World
Neurosurg 2014;82(6):1131–1143. doi:10.1016/
s12028-008-9050-6.
j.wneu.2014.07.006.

49 Qureshi AI, Vazquez G, Tariq N, et al. Impact of 60 Macdonald RL, Higashida RT, Keller E, et al.
International Subarachnoid Aneurysm Trial Randomized trial of clazosentan in patients
results on treatment of ruptured intracranial with aneurysmal subarachnoid hemorrhage
aneurysms in the United States. Clinical article. undergoing endovascular coiling. Stroke 2012;43(6):
J Neurosurg 2011;114(3):834–841. doi:10.3171/ 1463–1469. doi:10.1161/STROKEAHA.111.648980.
2010.6.JNS091486.
61 Macdonald RL, Higashida RT, Keller E, et al.
50 Molyneux AJ, Birks J, Clarke A, et al. The durability Randomised trial of clazosentan, an endothelin
of endovascular coiling versus neurosurgical receptor antagonist, in patients with aneurysmal
clipping of ruptured cerebral aneurysms: 18 year subarachnoid hemorrhage undergoing surgical
follow-up of the UK cohort of the International clipping (CONSCIOUS-2). Acta Neurochir Suppl
Subarachnoid Aneurysm Trial (ISAT). Lancet 2013;115:27–31. doi:10.1007/978-3-7091-1192-5_7.
2015;385(9969):691–697. doi:10.1016/S0140-
62 Marik PE. Techniques for assessment of
6736(14)60975-2.
intravascular volume in critically ill patients.
51 Claassen J, Albers D, Schmidt JM, et al. J Intensive Care Med 2009;24(5):329–337.
Nonconvulsive seizures in subarachnoid doi:10.1177/0885066609340640.
hemorrhage link inflammation and outcome.
63 Suarez JI, Qureshi AI, Yahia AB, et al. Symptomatic
Ann Neurol 2014;75(5):771–781. doi:10.1002/
vasospasm diagnosis after subarachnoid
ana.24166.
hemorrhage: evaluation of transcranial Doppler
52 Schürkämper M, Medele R, Zausinger S, et al. ultrasound and cerebral angiography as related
Dexamethasone in the treatment of subarachnoid to compromised vascular distribution. Crit Care
hemorrhage revisited: a comparative analysis of Med 2002;30(6):1348–1355.
the effect of the total dose on complications and
64 Frontera JA, Fernandez A, Schmidt JM, et al.
outcome. J Clin Neurosci 2004;11(1):20–24.
Defining vasospasm after subarachnoid
doi:10.1016/S0967-5868(03)00155-3.
hemorrhage: what is the most clinically relevant
53 Claassen J, Perotte A, Albers D, et al. definition? Stroke 2009;40(6):1963–1968.
Nonconvulsive seizures after subarachnoid doi:10.1161/STROKEAHA.108.544700.
hemorrhage: Multimodal detection and
65 Naidech AM, Jovanovic B, Wartenberg KE, et al.
outcomes. Ann Neurol 2013;74(1):53–64.
Higher hemoglobin is associated with improved
doi:10.1002/ana.23859.
outcome after subarachnoid hemorrhage. Crit
54 Kondziella D, Friberg CK, Wellwood I, et al. Care Med 2007;35(10):2383–2389. doi:10.1097/
Continuous EEG monitoring in aneurysmal 01.CCM.0000284516.17580.2C.
subarachnoid hemorrhage: a systematic review.
66 Banki N, Kopelnik A, Tung P, et al. Prospective
analysis of prevalence, distribution, and rate of
s12028-014-0068-7.
recovery of left ventricular systolic dysfunction
55 Rosengart AJ, Huo JD, Tolentino J, et al. Outcome in patients with subarachnoid hemorrhage.
in patients with subarachnoid hemorrhage J Neurosurg 2006;105(1):15–20. doi:10.3171/
treated with antiepileptic drugs. J Neurosurg jns.2006.105.1.15.
2007;107(2):253–260. doi:10.3171/JNS-07/
67 Pollick C, Cujec B, Parker S, Tator C. Left
08/0253.
ventricular wall motion abnormalities in
56 Naidech AM, Kreiter KT, Janjua N, et al. Phenytoin subarachnoid hemorrhage: an echocardiographic
exposure is associated with functional and study. J Am Coll Cardiol 1988;12(3):600–605.
cognitive disability after subarachnoid doi:10.1016/S0735-1097(88)80044-5.
68 Tung P, Kopelnik A, Banki N, et al. Predictors of
doi:10.1161/01.STR.0000141936.36596.1e.
neurocardiogenic injury after subarachnoid
57 Vergouwen MD, Vermeulen M, van Gijn J, et al. hemorrhage. Stroke 2004;35(2):548–551.
Definition of delayed cerebral ischemia after doi:10.1161/01.STR.0000114874.96688.54.
aneurysmal subarachnoid hemorrhage as an
69 Hjalmarsson C, Bergfeldt L, Bokemark L, et al.
outcome event in clinical trials and observational
Electrocardiographic abnormalities and elevated
studies: proposal of a multidisciplinary research
cTNT at admission for intracerebral hemorrhage:
group. Stroke 2010;41(10):2391–2395. doi:10.1161/
predictors for survival? Ann Noninvasive
Electrocardiol 2013;18(5):441–449. doi:10.1111/
58 de Oliveira Manoel AL, Goffi A, Marotta TR, et al. anec.12056.
The critical care management of poor-grade
70 Naredi S, Lambert G, Edén E, et al. Increased
subarachnoid haemorrhage. Crit Care 2016;20:21.
sympathetic nervous activity in patients with
doi:10.1186/s13054-016-1193-9.
nontraumatic subarachnoid hemorrhage. Stroke
59 Macdonald RL. Delayed neurological deterioration 2000;31(4):901–906. doi:10.1161/01.STR.31.4.901.
after subarachnoid haemorrhage. Nat Rev Neurol
71 Hammermeister KE, Reichenbach DD. QRS
2014;10(1):44–58. doi:10.1038/nrneurol.2013.246.
changes, pulmonary edema, and myocardial
necrosis associated with subarachnoid
hemorrhage. Am Heart J 1969;78(1):94–100.
doi:10.1016/0002-8703(69)90264-6.
1656 DECEMBER 2018

72 Doshi R, Neil-Dwyer G. Hypothalamic and 77 Fraser JF, Stieg PE. Hyponatremia in the
myocardial lesions after subarachnoid neurosurgical patient: epidemiology,
haemorrhage. J Neurol Neurosurg Psychiatry pathophysiology, diagnosis, and management.
1977;40(8):821–826. Neurosurgery 2006;59(2):222–229; discussion
222–229. doi:10.1227/01.NEU.0000223440.
73 Bassil R, Kaddouh F, Greer DM, Muehlschlegel S.
35642.6E.
Neurogenic shock. In: Layon AJ, Gabrielli A, Yu M,
Wood KE, editors. Civetta, Taylor, and Kirby's 78 Oddo M, Milby A, Chen I, et al. Hemoglobin
critical care medicine. 5th ed. Philadelphia, PA: concentration and cerebral metabolism in
Wolters Kluwer, 2017. patients with aneurysmal subarachnoid
74 Pastrana EA, Saavedra FM, Murray G, et al. Acute
doi:10.1161/STROKEAHA.108.527911.
adrenal insufficiency in cervical spinal cord
injury. World Neurosurg 2012;77(3–4):561–563. 79 Naidech AM, Shaibani A, Garg RK, et al.
doi:10.1016/j.wneu.2011.06.041. Prospective, randomized trial of higher goal
hemoglobin after subarachnoid hemorrhage.
75 Alaraj A, Wallace A, Mander N, et al. Risk factors
for heparin-induced thrombocytopenia type II
s12028-010-9424-4.
in aneurysmal subarachnoid hemorrhage.
Neurosurgery 2011;69(5):1030–1036. doi:10.1227/ 80 Witsch J, Frey HP, Patel S, et al. Prognostication
NEU.0b013e3182284a81. of long-term outcomes after subarachnoid
hemorrhage: the FRESH score. Ann Neurol 2016;
76 Stiefel D, Petzold A. H2O coma. Neurocrit Care
80(1):46–58. doi:10.1002/ana.24675.
2007;6(1):67–71. doi:10.1385/NCC:6:1:67.
81 Patel S. FRESH Score. iTunes web site. itunes.
apple.com/us/app/fresh-score/id1015675236?
mt=8. Accessed October 8, 2018.
DISCLOSURE
Continued from page 1623
in Brain Injured Patients) trial sponsored by C. R.
Bard Inc. Dr Muehlschlegel has received
compensation for serving as a course director for
the American Academy of Neurology.

REVIEW ARTICLE
Management of Stroke
in the Neurocritical

Care Unit
ONLINE

S U P P L E M E N T AL D I G I T A L
CONTENT (SDC) By Chethan P. Venkatasubba Rao, MD, FNCS; Jose I. Suarez, MD, FNCS, FANA
A VA I L A B L E O N L I N E
ABSTRACT
PURPOSE OF REVIEW: This article provides updated information regarding the
diagnosis and treatment (specifically critical care management) of acute
ischemic stroke. This article also discusses the increased use of
thrombolysis and thrombectomy in clinical practice.
RECENT FINDINGS: Stroke is the leading cause of disability in the United States.
CITE AS:
CONTINUUM (MINNEAP MINN) A significant proportion of patients with acute ischemic stroke require
2018;24(6, NEUROCRITICAL CARE): critical care management. Much has changed in the early evaluation and
1658–1682.
treatment of patients presenting with acute ischemic stroke. The
Address correspondence to introduction of embolectomy in large vessel occlusions for up to 24 hours
Dr Chethan P. Venkatasubba Rao, post–symptom onset has resulted in one in every three eligible patients
Baylor College of Medicine, MS
NB 124, One Baylor Plaza,
with acute ischemic stroke with the potential to lead an independent
Houston, TX 77030, lifestyle. These patients increasingly require recognition of complications
cprao@bcm.edu. and initiation of appropriate interventions as well as earlier admission to
dedicated neurocritical care units to ensure better outcomes.
Dr Venkatasubba Rao has
received personal compensation SUMMARY: This article emphasizes issues related to the management of
as an editorial board member
of Brain Disorders & Therapy. patients with acute ischemic stroke undergoing mechanical thrombectomy
Dr Suarez has received research/ and thrombolysis and addresses the complex physiologic changes
grant support from the National
affecting neurologic and other organ systems.
Institute of Neurological
Disorders and Stroke and
as co-investigator in the
SETPOINT2 (Stroke-related Early
Tracheostomy Versus Prolonged INTRODUCTION
A
Orotracheal Intubation in cute ischemic stroke is a neurologic emergency. A recent report from
Neurocritical Care Trial) study
from the Patient-Centered
the American Heart Association has shown that acute ischemic
Outcomes Research Institute. stroke affects an average of 800,000 people annually in the
Dr Suarez is the current president United States, the majority of whom experience their first event.1
and a member of the board of
directors of the Neurocritical This translates into one person having a stroke every 40 seconds.
Care Society. About 7.2 million Americans older than 20 years of age report having had a stroke,
and the prevalence is estimated to be 2.7%.2 Recent studies have noted that acute
UNLABELED USE OF
PRODUCTS/INVESTIGATIONAL ischemic stroke affects men and women equally overall but has variable influence
USE DISCLOSURE: in different ages and ethnic groups. Acute ischemic stroke has a higher incidence
Drs Venkatasubba Rao and
Suarez report no disclosures.
in American Indians/Alaskan Natives (5.4%), non-Hispanic blacks (4.5%), and
other races and multiracial people (4.7%) compared to non-Hispanic whites (2.5%).3
Acute ischemic stroke remains the number one cause of morbidity and
© 2018 American Academy disability in the United States, costing an unprecedented $33.9 billion, which is
of Neurology. 14% of annual health care expenditure. Several disturbing trends are
1658 DECEMBER 2018

noted in stroke epidemiology. Certain models have predicted that by 2030, KEY POINTS
3.4 million more adult Americans will have had a stroke, a staggering 20.5%
● One in every four patients
increase as compared to the prevalence in 2010. This increase is predicted with acute ischemic stroke
to affect Hispanic men by more than 29%. Stroke remains the fifth most common will need critical care
cause of mortality, claiming 133,000 lives annually in the United States. In other intervention.
words, one person dies every 4 minutes from stroke. Therefore, it is imperative
● Factors predicting critical
that preventive measures and timely administration of thrombolytics,
care admission in patients
thrombectomy, and critical care management are instituted in patients with with acute ischemic stroke
acute ischemic stroke to prevent the resultant morbidity and mortality. include severity of stroke,
age, elevated systolic blood
EPIDEMIOLOGY AND RISK FACTORS FOR INTENSIVE CARE pressure, and
hyperglycemia.
UNIT ADMISSION
It is estimated that only 5% to 7% of all patients with acute ischemic stroke are
treated with IV recombinant tissue plasminogen activator (rtPA) in the
United States. Admission of patients with acute ischemic stroke treated with IV
rtPA or thrombectomy primarily to an intensive care unit (ICU) varies and
depends on local health care practices and availability of resources. Specialized
clinical care units designated to manage patients with acute ischemic stroke have
been shown to improve outcomes and reduce the length of hospitalizations.4 The
impact of a neurocritical care unit team in the improvement of outcomes of
patients with traumatic brain injury, spinal cord injury, and neurosurgical
patients is very well established.5,6 However, its impact on patients with acute
ischemic stroke has yet to be systematically studied, but some preliminary
evidence suggests that admission to a neurocritical care unit is associated with
positive effects on this population.5
In this section, the definition of a neurocritical care unit will be limited to
specialized ICUs, which consist of a treatment team led by a neurointensivist
or an intensivist well versed in the treatment of neurologic and neurosurgical
patients who are critically ill; nursing staff trained for neurologic management;
respiratory therapists; pharmacists trained in critical care; and physical,
occupational, and speech therapists in an institution that is committed to
assuring quality of care as specified by their certifying bodies. As mentioned
above, a small proportion of patients with acute ischemic stroke are currently
receiving thrombolytic therapy,7 and because of the lack of screening data, it is
unclear what percentage of them would actually be eligible for thrombectomy
for up to both 8 and 24 hours. However, it is estimated that 24% of all patients
admitted with acute ischemic stroke may need critical care intervention.8
Potential factors that predict the need for critical care intervention include the
following: being African American (odds ratio of 3.8), being male (odds ratio of
3.8), systolic blood pressure values (odds ratio of 1.45 per 10 mm Hg increase),
and National Institutes of Health Stroke Scale (NIHSS) score (odds ratio of
1.09 per 1 point increase in NIHSS) in one study.8 In another study, age,
hyperlipidemia, the presence of extracranial vascular disease, and NIHSS were
found to be independent predictors of critical care needs.9 Separate scores such
as the Intensive Care After Thrombolysis (ICAT) score (TABLE 4-1) and the
Simplified Acute Physiology Score II (SAPS II) (a score using physiological
variables to evaluate the risk of mortality in any given disease state) have been
developed to assess the need for critical care intervention in patients presenting
with acute ischemic stroke. For every 1 point increase in the ICAT score, the
odds of requiring critical care increases 2.22 times. The odds of requiring critical

MANAGEMENT OF STROKE IN THE NEUROCRITICAL CARE UNIT
care for a patient with an ICAT score of more than 2 was 13 times higher than
for a patient with a score of less than 2. A score of 5 or more predicts critical
care needs with a 94% specificity and 45.8% sensitivity. Factors that can
independently influence the need for critical care in patients with stroke are
summarized in TABLE 4-2.
INITIAL MANAGEMENT OF ACUTE ISCHEMIC STROKE

Similar to other neurologic emergencies, patients with acute ischemic stroke
should be evaluated and triaged promptly so that they can receive adequate
management. Time is of the essence. The general principles of initial
management include the following:
u Assess circulation and, if needed, perform cardiopulmonary resuscitation per the basic
life support method
u Assess and secure airway and breathing (ventilation)
u Initiate a call for rapid response of a critical care team (obtain help)
u Point of care testing for glucose, coagulopathy
u Perform a standardized and complete neurologic assessment (NIHSS, see below).
TABLE 4-1 Determination of the Intensive Care After Thrombolysis Scorea
Intensive Care After

Thrombolysis
Score Component (ICAT) Score Points
Male sex
Yes 0
No 1
Black race
No 0
Yes 1
Systolic blood pressure in mm Hg
<160 0
160–200 2
>200 4
National Institutes of Health Stroke Scale Score
≤6 0
7–12 1
≥13 2
Total Score 0–8
a
Reprinted with permission from Faigle R, et al, Crit Care.8 © 2016 Faigle et al.
1660 DECEMBER 2018

u Review vital signs prior to the current neuroemergency
u Review history and list of medications and recent treatments received
u Neuroimaging
u Appropriate laboratory tests
u Neurosurgical consultation
u Update patient/surrogate regarding the diagnosis and plan of care and clarify agreement
for treatment.
NIHSS is a standardized neurologic assessment of patients and is helpful in

determining the severity of stroke but is heavily weighted toward anterior
circulation and left hemispheric symptoms. Left hemispheric strokes score at
least 4 points more than the opposite hemisphere.10 NIHSS has a limitation as a
bedside assessment tool, and special caution has to be exercised, especially
when performed by untrained personnel.
Thrombolytic and Endovascular Therapies

Acute ischemic stroke treatment has experienced a paradigm shift over the last
3 years. Foundations for this progress were laid in 1995, wherein a National
Institute of Neurological Disorders and Stroke–rtPA trial reported benefit in
3-month outcomes with the use of IV rtPA within 3 hours of acute ischemic
stroke symptom onset (SDC 4-1, links.lww.com/CONT/A259).11–22 Using that
protocol for rtPA infusion, one in every 12 patients would be functionally
independent in 3 months post–acute ischemic stroke as compared to patients
without receiving rtPA. Extending that window of thrombolysis between 3 and
4.5 hours in the ECASS III (European Cooperative Acute Stroke Study III),
functional independence was demonstrated in one in 14 patients treated with IV
rtPA as compared to patients not receiving thrombolytics.12
The IST-3 (Third International Stroke Trial) was yet another landmark trial for
enrolling patients older than 80 years, who comprised more than 25% of the
3035 patients enrolled in the study, and tested an extended window of IV
thrombolysis up to 6 hours. The results of this trial must be interpreted
Factors Increasing the Risk of Requiring Critical Care Interventions in TABLE 4-2
Patients With Acute Ischemic Strokea
◆ Age (odds ratio of 1.3 per 10 years)

◆ African American ethnicity (odds ratio of 3.8)
◆ Male sex (odds ratio of 3.8)
◆ Systolic blood pressure (odds ratio of 1.45 for every 10 mm Hg greater than 140 mm Hg)
◆ Hyperglycemia (odds ratio of 1.25 per 50 mg/dL increase in glucose)
◆ National Institutes of Health Stroke Scale (odds ratio of 1.09 per 1 point increase)
◆ Simplified Acute Physiology Score II of greater than 22
◆ Vascular disease affecting non-neurologic locations (ie, peripheral vascular disease)
a
Data from Faigle R, et al, Crit Care.8

cautiously. While outcome improvement was better when comparing complete

recovery or recovery with minimal symptoms (defined as an Oxford Handicap
Scale score of 0 or 1) in the group treated with rtPA as compared to controls, no
such improvement was seen when the data were analyzed to compare traditional
dependency versus independency (Oxford Handicap Scale score of <3 versus >3)
in 3-month outcomes.13 Furthermore, the rates of symptomatic intracranial
hemorrhage and other cerebral-related mortality rates were higher in patients
receiving rtPA, and thus the use of IV rtPA is currently limited to up to 4.5 hours
after symptom onset. A Cochrane meta-analysis of 12 trials using rtPA for
acute ischemic stroke demonstrated an odds risk of 0.85 for reducing the
proportion of patients with dependency or death at 3 months.23
It is only recently that rtPA use has been extended beyond the traditional
4.5 hours. In a study led by the WAKE-UP (MRI-Guided Thrombolysis
for Stroke With Unknown Time of Onset) investigators, patients were
selected for thrombolysis based on brain tissue viability as determined by a
diffusion-weighted imaging (DWI) sequence demonstrating ischemia and no
visible change in T2 fluid-attenuated inversion recovery (FLAIR) sequence.14
Patients were randomly assigned to either placebo or IV rtPA. Outcomes as
defined by a modified Rankin Scale (mRS) score of 0 or 1 were better in patients
who received rtPA (53.3%) as compared to the group treated with placebo
(41.8%). A higher incidence of hemorrhage was noted in patients receiving rtPA
(4% versus 0.4% in the placebo group), but the overall death or dependency
was still higher in patients receiving placebo (18.3% versus 13.5% in the rtPA-
treated group).14
Large vessel occlusions seem resistant to the effects of IV rtPA and portend a
devastating impact on clinical outcomes. Therefore, interventional strategies for
chemical thrombolysis and mechanical thrombectomy have been developed in a
stepwise evolution to its current stage, the roots of which can be traced to the
PROACT II (Prolyse in Acute Cerebral Thromboembolism II) study.24 Initially,
as seen in the PROACT II study, local thrombolytic delivery alone failed to
provide the anticipated response in perfusion and clinical recovery, which led to
the development of interventional techniques for clot disruption. Of the many
studies investigating this concept, the IMS III (Interventional Management of
Stroke III), MR RESCUE (Mechanical Retrieval and Recanalization of Stroke
Clots Using Embolectomy), and SYNTHESIS Expansion trials evaluated the use
of mechanical embolectomy in patients with acute ischemic stroke but failed to
demonstrate clinical improvement with intervention.25–27
Despite these trials having different methodologies, three universally pervasive
themes likely responsible for the failure of these studies could be identified, namely
poor selection of cases (ie, a significant portion of patients enrolled did not have
intracranial occlusion), unclear definition of parenchymal viability, and, finally,
ineffective embolectomy devices. It comes as no surprise that addressing these
specific issues led to clinically fruitful studies. It was the pragmatically designed
MR CLEAN (Multicenter Randomized Clinical Trial of Endovascular Treatment
for Acute Ischemic Stroke) study that paved the path for future successes of
interventional trials. Patients with middle cerebral artery– or internal carotid
artery–documented occlusions on a CT angiogram were randomly assigned to
conventional treatment or recanalization using mechanical stent retrievers with or
without thrombolysis within 6 hours of symptom onset. Independent outcomes
were seen in one of every seven patients treated with embolectomy.16
1662 DECEMBER 2018

Similarly, the REVASCAT (Randomized Trial of Revascularization With KEY POINT
Solitaire FR Device Versus Best Medical Therapy in the Treatment of Acute
● Circulation, airway, and
Stroke Due to Anterior Circulation Large Vessel Occlusion Presenting Within breathing stabilization
8 Hours of Symptom Onset)20 and ESCAPE (Endovascular Treatment for Small followed by rapid
Core and Anterior Circulation Proximal Occlusion With Emphasis on Minimizing neurologic assessment,
CT to Recanalization Times)17 trials evaluated the effect of thrombectomy by neuroimaging, and point of
care testing should be the
using the Alberta Stroke Program Early Computed Tomography Score (ASPECTS)
initial response in the
as a marker of tissue viability and CT angiography/magnetic resonance evaluation of patients
angiography (MRA) confirmation of intracranial occlusion, demonstrating presenting with acute
independent outcomes in one in seven or one in four patients, respectively. The ischemic stroke.
SWIFT-PRIME (Solitaire With the Intention for Thrombectomy as Primary
Endovascular Treatment)19 and EXTEND-IA (Extending the Time for
Thrombolysis in Emergency Neurological Deficits—Intra-arterial)18 trials used
CT perfusion studies to identify infarct core using RAPID software
(iSchemaView, Menlo Park, California), and when recanalized, one in four and
one in three patients, respectively, experienced functional independence.
In the immediate past, two trials have extended the traditional windows of
endovascular treatment for up to 24 hours. The DAWN (DWI or CT Perfusion
Assessment With Clinical Mismatch in the Triage of Wake-up and Late
Presenting Strokes Undergoing Neurointervention With Trevo) investigator–led
study evaluated patients who were last seen normal within 24 hours and
evaluated imaging for large vessel occlusion and used the RAPID software to
select patients with specific ischemic core volumes (SDC 4-1, links.lww.com/
CONT/A259).21 One out of every three patients treated experienced functional
independence. The very recently released DEFUSE 3 (Diffusion and pErFUsion
Imaging Evaluation for Understanding Stroke Evolution 3) study used perfusion
criteria to define an infarct volume (ischemic core) of less than 70 mL, a ratio
of volume of ischemic tissue to initial infarct volume of 1.8 or more, and an
absolute volume of potentially reversible ischemia (penumbra) of 15 mL or more.
One in every four patients has the potential to be independent when treated
using this protocol.22
It is only recently that tenecteplase has been used as a thrombolytic agent in
acute ischemic stroke. The EXTEND-IA TNK investigator–led study,
Tenecteplase Versus Alteplase Before Thrombectomy for Ischemic Stroke,
randomly assigned patients with intracranial occlusions who were within
4.5 hours of symptom onset to receive either rtPA or tenecteplase in addition to
thrombectomy.15 They observed that 22% of patients in the tenecteplase group
had more than 50% of occluded vessels reperfused or completely resolved
thrombus as compared to 11% in the group treated with rtPA. Outcomes
indicating independence in clinical function, defined by an mRS score of 2 or less,
was insignificant in both the groups. Both groups had similarly low rates of
symptomatic intracranial hemorrhage. Therefore, although tenecteplase seemed
to provide better brain reperfusion and thrombolysis, tenecteplase did not
seem to result in better clinical outcomes in patients with acute ischemic stroke
and an intracranial occlusion.15
INDICATIONS FOR INTENSIVE CARE UNIT ADMISSION

Patients who undergo thrombolysis and thrombectomy are usually at risk of
developing complications and are typically better served in an ICU and preferably
in the neurocritical care unit. It is reported that about one in four patients with

acute ischemic stroke post-thrombolysis will need critical care interventions

possible only in an ICU.8 Systemic complications in the setting of acute ischemic
stroke can pose significant secondary brain injury requiring care in a neurocritical
care unit. The following section will address complications in individual organ
systems that indicate admission to a neurocritical care unit (TABLE 4-3).
Neurologic Indications
Perhaps the most common reasons for admission to the neurocritical care unit are
neurologic indications, which are detailed below.
POST-THROMBOLYSIS/POST-THROMBECTOMY CARE AND HEMODYNAMIC

MONITORING. Because of variable practices across health care systems, patients
with acute ischemic stroke treated with thrombolysis are admitted to either a
dedicated stroke unit or to a neurocritical care unit. Both neurologic and
hemodynamic monitoring remains the mainstay of such admissions.
Management of hemodynamics (systolic blood pressure and diastolic blood
pressure parameters prethrombolysis and post-thrombolysis) should be followed
as directed by the American Heart Association guidelines and Brain Attack
Coalition.28,29 Post-thrombolysis, blood pressure should be maintained at less
than 180/105 mm Hg during the first 24 hours. Fluctuating neurologic symptoms
that correlate with hemodynamic changes will need either fluid management or
augmentation using vasopressors. More data are emerging supporting cardiac
output augmentation as more effective in enhancing cerebral perfusion.30 An
infrequent but life-threatening angioedema associated with the use of rtPA needs
aggressive airway management in an ICU setting.
CEREBRAL EDEMA. Cerebral edema is a common complication affecting patients

with acute ischemic stroke. Even in small infarcts, cerebral edema is associated
with a significant secondary brain injury.31 The subgroup of patients who are
at risk of severe life-threatening edema includes patients with posterior
circulation strokes with the involvement of more than 35% of the cerebellar
hemisphere and middle cerebral artery territory infarcts, which have a footprint
of more than 50% involvement of its territory on a CT scan. Clinically, that
usually is representative of any neurologic deficit greater than an NIHSS score of
15, with altered mentation.
SYMPTOMATIC HEMORRHAGIC TRANSFORMATION AND COAGULOPATHY.

Intracranial hemorrhagic transformation following acute ischemic stroke with or
without thrombolysis/thrombectomy has been found to range from 4% to 6%.32
Radiographic changes are classified as types 1 and 2 hemorrhagic infarctions
and types 1 and 2 parenchymal hematomas and extraischemic hematomas.
Hemorrhagic infarctions are petechial hemorrhages that immediately follow an
ischemic infarct and are seen in the first 1 to 2 weeks after an ischemic stroke. A
hemorrhagic infarct is type 1 if the radiographic appearance shows a multifocal
petechial pattern and is type 2 when an appearance of confluent hyperdensity
occurs throughout the infarcted area without a resultant mass effect.
Parenchymal hematomas show a homogenous appearance of hyperdensity and
are type 1 if the hematoma occupies less than one-third of the infarct zone
leading to some mass effect. Parenchymal hematomas are type 2 if the hematoma
involves more than one-third of the infarct zone with significant mass effect or if
it extends beyond the borders of the infarcted brain parenchyma. Extraischemic
1664 DECEMBER 2018

Indications for Admission to Neurocritical Care Unit in Patients With Acute TABLE 4-3
Ischemic Stroke
Neurologic Indications
◆ Hemodynamic management
◇ Post-thrombectomy
◇ Post IV thrombolysis
◇ Need for continuous hemodynamic support
◆ Cerebral edema
◇ Cerebellar stroke involving more than 25% to 33% of hemisphere
◇ Involvement of more than 50% of middle cerebral artery territory
◇ Signs of herniation
◆ Hydrocephalus
◆ Symptomatic hemorrhagic transformation and coagulopathy
◆ Coma with Glasgow Coma Scale score of <9
◆ Seizures requiring continuous IV medications
Cardiac Indications
◆ Hemodynamic augmentation
◆ Acute myocardial infarction requiring monitoring and hemodynamic support
◆ Cardiac rhythm abnormalities (such as atrial flutter/fibrillation)
◆ Congestive cardiac failure requiring continuous IV infusions
◆ Cardiac mechanical hardware on anticoagulation with moderate to large strokes
Respiratory Indications
◆ Respiratory failure requiring endotracheal intubation and mechanical ventilation
◆ Hypoxic respiratory failure
◆ Hypercarbic respiratory failure
◆ High suspicion of aspiration pneumonia
◆ Central respiratory failure
Infectious Indications
◆ Signs of sepsis or septic shock
Renal Indications
◆ Renal failure requiring renal replacement therapy
IV = intravenous.

hematomas, however, are areas of confluent hyperdensity that do not overlap

with the infarcted brain parenchyma.
Symptomatic intracranial hemorrhage, defined as any hemorrhage resulting
in a change in clinical status manifested by an NIHSS score increase of 4 points or
more, seems to affect patients with larger ischemic strokes, high systolic blood
pressure, diabetes mellitus, and, in certain cases, low plasminogen inhibitor
levels in serum, and is seen in less than 2% to 7% of patients. The radiographic
appearance of parenchymal hematoma type 2 is associated with consistently
poorer clinical outcomes and hence needs aggressive management in the
neurocritical care unit.
SEIZURES. Clinical ictal events following acute ischemic stroke are relatively rare
and may occur in about 1.3% of cases.33 Males with an NIHSS score of greater
than 10 are at risk, and the seizures themselves portend an independent poor
outcome (twofold to threefold increase in odds).
Cardiac Indications
Patients with acute ischemic stroke are prone to cardiac complications, which
need neurocritical care unit management, as detailed below.
MYOCARDIAL INFARCTION. Acute myocardial infarction is seen in 2% of all

patients with acute ischemic stroke with a cumulative risk over 10 years.34
Myocardial infarction in the setting of acute ischemic stroke is associated with
poor outcomes and increased mortality. Thus, close monitoring of the patients
in the ICU is warranted.
CARDIAC ARRHYTHMIAS. The most common arrhythmias following acute

ischemic stroke include atrial fibrillation and atrial flutter, reported in as many as
10% of patients.35 Rapid ventricular rate in atrial fibrillation requires immediate
management because of the resultant hemodynamic compromise.
CONGESTIVE HEART FAILURE. Left ventricular systolic dysfunction is reported

in as many as 28% of the population with acute ischemic stroke and diastolic
dysfunction in 18%. Congestive heart failure has been recognized as an independent
marker of poor outcomes in patients with ischemic stroke and is associated with
cardiac and pulmonary complications requiring monitoring in the ICU.
Respiratory Indications
Respiratory complications that indicate admission to the neurocritical care unit
are detailed below.
AIRWAY MANAGEMENT. The inability to maintain a patent airway because of

altered mentation seems to be the most common respiratory indication (47%)
for admitting patients with acute ischemic stroke to the neurocritical care unit.
PNEUMONIA. Stroke-associated pneumonia has been variably reported in the

literature with a wide range of 4% to 56% of all patients in the stroke units.36
Dysphagia and altered level of consciousness are partly contributory to pneumonia.
However, an increased incidence of pneumonia compared to similar neurologic
illnesses can be potentially explained by stroke-induced immunodepression.
Pneumonia confers a poor functional outcome and mortality as high as 49% to 60%
in patients with acute ischemic stroke as seen in multiple studies.
1666 DECEMBER 2018

MECHANICAL VENTILATION. Approximately 1.3% of all patients with stroke KEY POINTS
undergo mechanical ventilation due to various causes of respiratory failure and
● The main neurologic
experience 40% to 80% mortality. Patients requiring mechanical ventilation reasons for intensive care
have associated poorer outcomes and a longer hospital stay, which could be unit admission in patients
due to the administration of sedatives and analgesics, ventilator-associated with acute ischemic stroke
pneumonia, or both. Various other miscellaneous complications of the include blood pressure
management
respiratory system occur that result in hypoxic or hypercarbic respiratory failure
post-thrombolysis or
that may be managed with mechanical ventilation or noninvasive modes of post-thrombectomy,
positive breathing, as will be mentioned later, but that require monitoring in the cerebral edema,
neurocritical care unit. symptomatic hemorrhagic
transformation, and
seizures.
Neurocritical Care Unit Triaging
Several predictive models have been created to assess appropriate triaging ● The main cardiac and
of patients with acute ischemic stroke to the neurocritical care unit. Of these, respiratory indications for
the ICAT score is a novel tool that scores based on the ethnicity, age, gender, admission of patients with
acute ischemic stroke to the
the degree of hypertension, and clinical severity of stroke measured by the
neurocritical care unit
NIHSS. Current literature suggests that for every point increase in the ICAT include myocardial
score, there is a 2.2-fold higher chance of requiring intervention. An ICAT infarction, cardiac
score of more than 2 has a 13-fold risk of requiring critical care intervention arrhythmias, heart failure,
and hence can potentially be used for triaging patients to the neurocritical inability to maintain the
airway, and the need for
care unit.8 mechanical ventilation.
INTENSIVE CARE UNIT MANAGEMENT

The critical care management of patients with acute ischemic stroke is complex.
An evidence-based approach is presented below, and when no or very little
evidence exists, the authors recommend what they and most other
neurointensivists do in their daily practice. The recommendations for ICU
management of patients with acute ischemic stroke are summarized in TABLE 4-4.
Neurologic Management
Neurologic complications that follow acute ischemic stroke need diligent
management, as is detailed below.
POST-THROMBOLYTIC AND POST-THROMBECTOMY CARE. The care of patients

post-thrombolysis and post-thrombectomy include neuromonitoring as
recommended by the American Heart Association and Brain Attack Coalition
guidelines, which include blood pressure measurement and neurologic
examination every 15 minutes for the first 2 hours after IV rtPA infusion, then
every 30 minutes for the next 6 hours, and subsequently every hour for the next
16 hours.28,29 As part of the neuromonitoring of patients, special attention should
be provided to the airway during or immediately after the IV administration of
rtPA because of the rare but potentially fatal angioedema association with its
administration. Concurrent use of lisinopril has been noted in the majority of
such patients. Changes in neurologic assessment should be emergently evaluated
by a noncontrast head CT scan. In addition, patients who have undergone
embolectomy will need monitoring of angiography access site(s) to assess for the
presence of hematoma and evaluation of peripheral pulses for determination
of distal perfusion. In the event of any hematoma collection, local mechanical
pressure application should be continued followed by correction of
underlying coagulopathy.

TABLE 4-4 Principles of Critical Care Management in Patients With Acute

Ischemic Stroke
Neurologic Critical Care

Issues Treatment Principles Treatment Specifics
Post-thrombolytic/ Neuromonitoring
post-thrombectomy
care Hemodynamic management See cardiac, cerebral perfusion control, and
hemorrhagic transformation for specific
management
Dysphagia assessment Nothing by mouth orders in anyone with suspected

dysphagia
Angioedema Diphenhydramine and other histamine 1 receptor

antagonists, corticosteroids
Access site hematoma Local pressure and coagulopathy reversal (see

hemorrhagic transformation for specific
management)
Cerebral perfusion Hypertension (for goals <185/110 mm Hg Short-acting injectable agents such as metoprolol,
control prethrombolysis and <180/105 mm Hg labetalol, enalaprilat, and hydralazine, or a
post-thrombolysis) continuous agent such as nicardipine
Vasopressor and inotropic support Epinephrine, norepinephrine, dobutamine, milrinone,

and dopamine
Hemorrhagic Hemodynamic management See hypertension treatment above; blood pressure

transformation parameters should be individualized, but systolic
blood pressure <160 mm Hg per intracerebral
hemorrhage management guidelines should be
followed when possible
Reversal of recombinant tissue Cryoprecipitate 10 U, repeated for goal fibrinogen

plasminogen activator administration level of >150 mg/dL; platelet transfusion 8–10 U, may
also consider recombinant factor VIIa 20–160 mcg/kg;
antifibrinolytic agents, aminocaproic acid 4 g IV first
hour followed by 1 g/h for 8 hours or tranexamic acid
10 mg/kg 3 to 4 times a day
Reversal of coagulopathy Platelet transfusion 8–10 U (for thrombocytopenia of

<100,000/mm3); fresh frozen plasma 12 mL/kg (for
patients on warfarin and especially with implanted
mechanical devices); prothrombin complex
concentrate 25–50 U/kg (based on international
normalized ratio level); vitamin K 10 mg IV (adjunctive
for reversing warfarin); recombinant factor VIIa
20–160 mcg/kg; antifibrinolytic agents, aminocaproic
acid 4 g IV first hour followed by 1 g/h for 8 hours or
tranexamic acid 10 mg/kg 3 to 4 times a day,
idarucizumab and andexanet alfa as needed for
reversal of newer anticoagulants.
IV = intravenous.
1668 DECEMBER 2018

HEMODYNAMIC MANAGEMENT. Hemodynamic management of patients with
acute ischemic stroke is perhaps the most common indication for neurocritical
care unit admission. The American Heart Association and the Brain Attack
Coalition recommend that patients post-thrombolysis should be treated with
short-acting agents or continuous drip of medications for a blood pressure
goal of less than 185/110 mm Hg before thrombolysis and less than 180/
105 mm Hg for at least 24 hours post-thrombolysis. If the patients are not
being thrombolysed, the blood pressure goals should be changed to less than
220/120 mm Hg.28,29
An intact cerebral autoregulation ensures unimpeded cerebral perfusion when
blood pressures fluctuate. Loss of such autoregulatory mechanisms can lead to
either hyperperfusion or worsening of cerebral perfusion. In ischemic strokes,
generally the intracranial pressure remains unchanged in the hyperacute period,
and hence one can surmise that the mean arterial pressure and systemic
pressures are the sole determinants of cerebral perfusion pressures. Therefore,
hemodynamic monitoring becomes a clear priority.
Various techniques such as positron emission tomography (PET), transcranial
Doppler, and MRI-based studies have resulted in variable observations. Recently,
dynamic autoregulation has been widely studied with the use of MRI or transcranial
Doppler. (In these studies, a thigh pressure cuff is applied on a lower extremity
and inflated to isolate a significant volume of blood. Cerebral vascular pulse
contours can be measured using transcranial Doppler or special sequences in MRI.
Deflating the thigh pressure cuff augments the cardiac output, which provides an
opportunity to study how the intracranial vasculature can adapt to hemodynamic
changes. With measurements of intracranial vascular flows in two different
hemodynamic states, we can understand cerebrovascular autoregulation.)
In some studies of patients with acute ischemic stroke with large vessel
occlusions, autoregulation has been demonstrated to be severely affected.37
Furthermore, patients with impaired autoregulation in the hemisphere affected
by stroke seemed to have worse outcomes when compared to patients with
preserved autoregulation. Moreover, in studies using concomitant transcranial
Doppler and arterial blood pressure waveform analyses, it is noted that an
impaired autoregulatory index can be appreciated for the first week following
acute ischemic stroke and normalized within the following week.38
Conflicting observations refuting the presence of autoregulatory failure also
are abundant in the literature. Studies evaluating autoregulation using various
modalities have found that although there is a difference between autoregulation
in patients with acute ischemic stroke and normal controls, there appears to be no
clear difference between the affected and unaffected hemispheres in patients
with acute ischemic strokes.39 Several factors are responsible for producing these
variable observations. First, patient selection in individual studies varied widely;
while some focused on moderate to large ischemic infarcts, some clearly defined
large vessel ischemia as the selection criteria. Second, multiple modalities have
been used in the evaluation of autoregulation, which makes it difficult to compare
the studies in an objective manner. Third, within evaluations of an individual
modality, researchers have used different parameters to define autoregulation.
Finally, the studies always compare immediate physiologic outcomes while
standardized long-term clinical outcomes are not reported. Given these
shortcomings, it is not surprising that such a vast difference is seen in the observed
outcomes of cerebral autoregulatory changes in acute ischemic stroke.

It is important to understand that admission blood pressures have a U-shaped

response in patients with acute ischemic stroke, with poor outcomes occurring
in patients presenting with the lowest and the highest values.40 Clinical trials
have tried to address the safety of reducing blood pressures in the setting of
acute ischemic stroke. The SCAST (Angiotensin-receptor Blocker Candesartan
for Treatment of Acute Stroke Trial) study evaluated blood pressure
management in acute stroke by randomly allocating patients with acute ischemic
stroke presenting with a systolic blood pressure of greater than 140 mm Hg to
treatment with either candesartan or placebo. Although the blood pressures were
significantly lower in the candesartan group, no significant difference was found
between the groups in mRS at 6 months or in the composite end point of vascular
death, myocardial infarction, or stroke during the first 6 months.41 Similarly, the
COSSACS (Continue or Stop Poststroke Antihypertensives Collaborative Study)
trial enrolled patients with acute ischemic stroke to stop the home regimen of
antihypertensive medications to evaluate the presence of improved outcomes
and better perfusion as manifested by higher blood pressures.42 The trial was
underpowered and observed no differences between the groups.
The CATIS (China Antihypertensive Trial in Acute Ischemic Stroke) study
enrolled the intended 4071 participants with acute ischemic stroke who presented
with a systolic blood pressure of 140 mm Hg to 220 mm Hg. CATIS investigators
randomly assigned patients to either an antihypertensive arm, which consisted of
lowering admission blood pressure by 10% to 25% within the first 24 hours after
randomization and a blood pressure goal of less than 140/90 mm Hg within 7 days,
or a complete stopping of all antihypertensive treatment during hospitalization.
The authors reported no difference in the primary outcome of death or an mRS
score of ≥3 at 2 weeks or hospital discharge between treatment arms.43
Another large study, the ENOS (Efficacy of Nitric Oxide in Stroke) clinical
trial, enrolled 4011 patients with ischemic or hemorrhagic stroke with a systolic
blood pressure of 140 mm Hg to 220 mm Hg on admission. The patients were
randomly assigned to either 7 days of transdermal nitrate therapy at 5 mg/d or
placebo. Similar to the other clinical trials, the ENOS investigators reported no
difference in the 90-day mRS score between the treatment arms.44 Currently, it
is recommended to reduce the blood pressure to less than 180/105 mm Hg for
patients with acute ischemic stroke post-thrombolysis.
As mentioned above, mechanical thrombectomy trials have benefited the
landscape of stroke treatment since 2015. With the use of stent retrievers, several
trials have demonstrated outcome improvement by 14% to 36%, which translates
as the total number of patients needed to treat to achieve one person who is
independent at 3 months between 3 and 7.16–21,45 The patients assigned to the
thrombectomy arm had reperfusion (defined as Thrombolysis in Cerebral
Infarction [TICI] grades of 2B to 3) rates of 59% to 88%. This variability in the
reperfusion rates is likely related to the study design, patient selection, treatment
times, study power, and overall different set of the population being analyzed. Of
note, no difference was found in mortality between the treatment groups across
the trials. It is currently unknown whether hemodynamic parameters need to be
changed based on the degree of reperfusion. However, it is important to
individualize blood pressure parameters in patients, especially when there is a
worsening neurologic examination associated with lower blood pressures.
Vasopressors or inotropic agents should be used to improve the neurologic
examination. A trial of an individual agent is recommended initially to assess
1670 DECEMBER 2018

neurologic examinations for a set pressure parameter, and readjusting the goals KEY POINTS
or treating agents should be performed dynamically (CASE 4-1).
● Management of
hemodynamics in patients
HEMORRHAGIC TRANSFORMATION. Neurologic worsening in the setting of with fluctuating neurologic
intracranial hemorrhage following acute ischemic stroke is associated with poor symptoms should be
outcomes and mortality rates of up to 50%, especially in patients with targeted individually and
adjusted for optimum
parenchymal hematoma type 2 (confluent hematoma that involves more than
symptom control.
one-third of infarcted tissue, as detailed above), which represents the majority of
all symptomatic intracranial hemorrhages.46 Most of these hemorrhages occur ● Interpretation of the
within 12 hours following IV thrombolysis, with a median time of 8 hours.47,48 neurologic evaluation of
It is important to point out that neurologic deterioration may not be obvious patients with acute ischemic
stroke with larger deficits
because of the hematoma development in the infarcted tissue. Changes in can be obscured when only
neurologic status are usually heralded by the expansion of the hematoma using the National Institutes
into noninfarcted tissues, mass effect, or extension onto the CSF-filled spaces. of Health Stroke Scale and
Also, neurologic patients who have severe baseline stroke impairment may no other clinical parameters.
not manifest a change in the neurologic status because of the ceiling effect of
the clinical examination and NIHSS. Therefore, a low threshold for follow-up
neuroimaging studies should be maintained. Two important factors in
determining the outcomes of symptomatic intracranial hemorrhage are
rapid hemodynamic control and correction of underlying coagulopathy.49
Symptomatic intracranial hemorrhage as a manifestation of cerebral
hyperperfusion has been noted mostly in patients undergoing elective carotid
surgical and endovascular revascularization but has occasionally been reported
after mechanical embolectomy.50 In a meta-analysis involving all factors
influencing the association of hyperperfusion and intracranial hemorrhage in the
setting of carotid reperfusion, patients undergoing carotid endarterectomy had
an increased odds of developing hyperperfusion (odds ratio of 1.4 times) as
compared to patients undergoing carotid stenting.45 Since hyperperfusion
manifests itself as a hemorrhagic stroke, evaluation of the symptomatic
intracranial hemorrhage rates in the embolectomy trials may be helpful to
identify potential patients for stricter hemodynamic control. Symptomatic
intracranial hemorrhage rates in patients treated with embolectomy were not
reported in the EXTEND-IA and the SWIFT-PRIME studies, while REVASCAT
had the same rates as the control group at 1.9%. The ESCAPE study showed a
symptomatic intracranial hemorrhage rate of 3.6%, while MR CLEAN had 7.7%
in those patients treated with embolectomy. Imaging-based selection of patients
who were beyond the traditional 6-hour window for treatment with mechanical
thrombectomy resulted in symptomatic intracranial hemorrhage rates of
6%, despite a mean time of revascularization of 13 hours after the onset of
symptoms. Hemodynamic regulation has the potential to prevent hemorrhagic
transformation. However, it is unclear as to how to individualize parameters to
obtain optimum perfusion and to prevent hyperperfusion that can lead to
hemorrhage. Until further evidence is available, best judgment should be used
to personalize hemodynamic parameters to balance cerebral, systemic, and
cardiac perfusion (CASE 4-2).
Correction of coagulopathy should be individualized based on whether patients
have received IV thrombolysis or active anticoagulant use. In patients who have
received thrombolysis with rtPA, assessment of plasma fibrinogen levels and
correction with IV cryoprecipitate (10 U) for a goal of more than 150 mg/dL should
be initiated at the earliest opportunity, as these steps perhaps have the most benefit

CASE 4-1 A 57-year-old right-handed man presented to the emergency department

for evaluation of left-sided weakness, numbness, and loss of vision he
noticed upon waking after going to bed 4 hours earlier. He had a history of
atrial fibrillation and nonischemic cardiomyopathy and had been
prescribed dabigatran with unknown compliance.
On examination, he had a left hemiplegia, hemianesthesia, hemineglect,
and left hemianopsia, which yielded a National Institutes of Health Stroke
Scale (NIHSS) score of 13.
Admission head CT showed a hyperdense right middle cerebral artery
sign (FIGURE 4-1A). He underwent thrombectomy with a full Thrombolysis
in Cerebral Infarction (TICI) grade 3 recanalization (FIGURE 4-1B and
FIGURE 4-1C), after which his NIHSS score improved to 3.
During his neurocritical care unit stay, his weakness worsened as his
systolic blood pressure decreased from 170 mm Hg to 110 mm Hg. He was
started on an IV norepinephrine infusion to elevate his systolic blood
pressure, after which his neurologic symptoms improved. However, he
experienced pulmonary edema of cardiac origin, for which he received
diuretics, and the norepinephrine infusion was titrated down followed by
dobutamine and milrinone infusions. His hemodynamic status was
monitored using an arterial catheter and attached noninvasive cardiac
output monitor, which allowed for the titration of his medications to
regulate his peripheral vascular resistance and cardiac output. His left
ventricular ejection fraction evaluated by transthoracic echocardiogram
initially was 20%, down from his baseline of 40% to 45%. The patient was
diagnosed with non–ST segment elevation myocardial infarction with
troponinemia, which was treated with an IV heparin drip with a lower
partial thromboplastin time (PTT) goal initially poststroke, in coordination
with the cardiology team.
FIGURE 4-1
Imaging of the patient in CASE 4-1. A, Head CT showing a hyperdense right middle cerebral
artery sign (arrow). B, Cerebral angiogram showing a distal right internal carotid artery
occlusion. C, Cerebral angiogram obtained after thrombectomy demonstrating Thrombolysis
in Cerebral Infarction (TICI) grade 3 recanalization. D, MRI of the brain showing no
abnormalities suggestive of cerebral infarction on diffusion-weighted imaging (DWI).
1672 DECEMBER 2018

Within the next 2 days, his respiratory and cardiac symptoms
improved, his left ventricular ejection fraction improved to his baseline
values, and his NIHSS score normalized to 0 upon discharge, as
evidenced by lack of any diffusion-weighted image (DWI) changes on his
MRI (FIGURE 4-1D). He was started on rivaroxaban for compliance issues and
remained symptom free.
This case demonstrates the need for individualization of care, using COMMENT
noninvasive monitoring devices to optimize hemodynamics, and highlights
coordination of management between care teams. The patient was not
eligible for IV recombinant tissue plasminogen activator as his intake of oral
anticoagulants was uncertain and no laboratory test was available to
determine the activity of his anticoagulant. Hence, he was treated with
embolectomy. Despite complete recanalization, subsequent
compromised hemodynamics resulted in worsening of his neurologic
deficits. He was initially supported by vasopressors that likely resulted in
congestive cardiac failure due to systolic failure of the left ventricle caused
by increased afterload. He likely developed non–ST segment elevation
myocardial infarction due to Takotsubo cardiomyopathy and pump failure.
Noninvasive cardiac monitoring was used to determine optimum
vasopressor and inotrope dosage to perfuse the brain while not
exacerbating cardiac decompensation.
This case also illustrates cautious use of anticoagulants in the setting
of acute ischemic stroke. In the setting of an acute ischemic stroke,
anticoagulation is usually not recommended. In this case, the patient
developed troponinemia with non–ST elevated myocardial infarction on
ECG, which necessitated the use of anticoagulation. After ensuring the
absence of any parenchymal injury on MRI, the patient was started on a
heparin infusion with lower anticoagulation goals (PTT of 50 to 60 seconds
instead of the standard 60 to 80 seconds). Noninvasive monitoring of
cardiac and neurologic status in patients can be very helpful in optimizing
personalized care for patients who are critically ill.

CASE 4-2 A 68-year-old man presented to the emergency department with a left
hemiplegia, hemianesthesia, and right gaze preference that began
2 hours before presentation. His head CT was remarkable for a right
middle cerebral artery hyperdense sign. He received IV recombinant
tissue plasminogen activator (rtPA), and CT angiography demonstrated a
right middle cerebral artery occlusion (FIGURE 4-2A). With a National
Institutes of Health Stroke Scale (NIHSS) score of 12, he underwent
thrombectomy with a resultant Thrombolysis in Cerebral Infarction (TICI)
grade 3 recanalization (FIGURE 4-2B and FIGURE 4-2C).
FIGURE 4-2
Imaging of the patient in CASE 4-2. CT angiogram (A) and cerebral angiogram (B) demonstrating
an occlusive thrombus (arrows), and post-thrombectomy cerebral angiogram showing
Thrombolysis in Cerebral Infarction (TICI) grade 3 recanalization (C). Noncontrast head
CTs showing intraparenchymal hemorrhage (D, arrow) and right hemispheric edema, and
after right hemicraniectomy for malignant right hemispheric edema (E).
1674 DECEMBER 2018

His NIHSS score worsened to 18 within 4 hours of IV rtPA administration
despite his systolic blood pressure being under control on a nicardipine
drip. A repeat head CT demonstrated intracranial hemorrhage with
worsening edema (FIGURE 4-2D). His systolic blood pressure goal was
reduced to 160 mm Hg, and he received 10 U of cryoprecipitate. His serum
fibrinogen levels improved from 80 mg/dL to 154 mg/dL. The
cryoprecipitate was used to reverse rtPA activity and to aid in reducing
intracranial hemorrhage expansion in anticipation of possible surgical
intervention.
A neurosurgical consultation was sought, and he underwent
prophylactic hemicraniectomy for cerebral edema (FIGURE 4-2E). Platelet
infusion was initiated during surgery. His hospital stay was remarkable
for new-onset atrial fibrillation with rapid ventricular response that
required rate control and mechanical ventilation support for 1 week
following hemicraniectomy. Three weeks following the hemorrhagic
transformation, the patient was started initially on a heparin drip and
eventually transitioned to warfarin. The latter was chosen after
discussion with patient’s family regarding all therapeutic options. He was
extubated after a week of mechanical ventilator support and eventually
was discharged to a skilled nursing facility.
Cranioplasty was performed after 2 months under careful transition
of warfarin to heparin, and he was eventually restarted on oral
anticoagulation with warfarin. Follow-up at 3 months revealed improved
dysarthria and left spastic hemiparesis, but the patient was ambulatory
with the help of a walker.
Thrombolysis and thrombectomy were performed in this patient upon COMMENT

presentation. The hemorrhagic complication associated with rtPA in this
patient was treated with hemodynamic management and coagulopathy
control by administering cryoprecipitate. Malignant cerebral edema was
treated with hemicraniectomy. Atrial fibrillation was detected during the
admission, and careful anticoagulation was initiated with IV heparin and
eventually transitioned to warfarin. The patient was transitioned to
rehabilitation and recovered with moderate disability. Management of
patients with acute ischemic stroke requires a multidisciplinary approach,
and the treatment of each individual complication should be approached
with personalized goals of care.

of all the treatment options. It is estimated that for every 10 U of cryoprecipitate

administered, plasma fibrinogen levels increase by an average of 55 mg/dL to
60 mg/dL. While interpreting the levels of fibrinogen, it is important to realize
that it is one of the acute phase reactants, and hence a lower level is more
reliable, and a normal level should be considered with healthy skepticism.
Other treatments that could be considered in the right setting include platelet
transfusion, fresh frozen plasma, prothrombin complex concentrate, activated
factor VII, antifibrinolytic medications, and reversal agents for direct acting
oral anticoagulants. Platelet transfusions of 6 U to 8 U are also usually
recommended based on a theoretic concern of thrombolysis inhibiting platelet
function. Furthermore, platelet transfusions can also be used to correct
thrombocytopenia for platelet counts of less than 100,000/μL. Fresh frozen
plasma contains endogenous procoagulant and anticoagulant proteins that
enhance intrinsic and extrinsic pathways with an end result of converting
fibrinogen to fibrin. Fresh frozen plasma is used at a dose of 12 mL/kg while
treating patients with hemorrhagic complications from rtPA. Prothrombin
complex concentrate is a concentrated form of vitamin K–dependent clotting
factors (II, VII, IX, and X) along with proteins C and S. Prothrombin complex
concentrate is the first-line treatment for warfarin-associated intracranial
hemorrhage and is extremely helpful by reducing the volume and time of IV
infusion required to achieve reversal of anticoagulation, which is usually in
30 minutes. Activated factor VII and vitamin K are used as adjuncts to reverse
warfarin. Use of prothrombin complex concentrate and activated factor VII
should be undertaken cautiously in patients with mechanical cardiac supports
and valves. The thrombogenic nature of the reversal agents can result in
mechanical cardiac valve dysfunction and clotting of the lines of mechanical
heart support devices. In such a situation, fresh frozen plasma seems to be an
optimal choice for regulating anticoagulation but presents challenges with
volume expansion and furthering cardiac stresses. Antifibrinolytic agents such
as tranexamic acid and aminocaproic acid can be used sparingly in an attempt to
prevent hematoma expansion. However, these agents also have a tendency to
cause deep vein thrombosis and cardiac and systemic thrombosis and are best
avoided in patients with mechanical heart devices and valves.
Direct-acting oral anticoagulants are increasingly being used in clinical
practice and pose a unique challenge for reversing their actions. As of now, the
agent idarucizumab, which is approved by the US Food and Drug Administration
(FDA), used at 5 g IV dose, is the only medication available to reverse the
effect of dabigatran. Agents such as rivaroxaban, apixaban, and edoxaban have
limited ability to be reversed. A factor X mimic, andexanet alfa, has been studied
and reported to reverse the activity of the direct-acting oral anticoagulants and
is only recently available for clinical use.51 Surgery should be considered after
correcting coagulopathy in selected patients where hematoma may be surgically
accessible by minimally invasive, stereotactic intervention or craniotomy, or for
decompressive hemicraniectomy.
CEREBRAL EDEMA. Cerebral edema is seen in patients with moderate to large

acute ischemic strokes and can independently worsen clinical outcomes.31
Malignant cerebral edema manifesting as a hemispheric syndrome in
combination with altered mental state and impending herniation and death is
seen in 10 to 20 per 100,000 people. It is more frequent in women and younger
1676 DECEMBER 2018

patients because of low compensatory intracranial space. Most of the trials KEY POINTS
assessing the prophylactic management of such severe edema have sought
● In patients with acute
surgical treatment with hemicraniectomy, with a minimal flap diameter of 15 cm. ischemic stroke, treatment
A pooled analysis of the clinical trials DECIMAL (Decompressive Craniectomy in of hemorrhagic
Malignant Middle Cerebral Artery Infarction), DESTINY (Decompressive transformation should be
Surgery for the Treatment of Malignant Infarction of the Middle Cerebral directed toward
hemodynamic control and
Artery), and HAMLET (Hemicraniectomy After Middle Cerebral Artery
coagulopathy reversal.
Infarction With Life-threatening Edema Trial) indicated that performing
hemicraniectomy in patients with malignant cerebral edema reduced mortality ● Recombinant tissue
by 50% and improved outcomes by 16%.52 Improved outcomes are seen plasminogen
mainly in patients younger than 60 years of age who present with an NIHSS activator–related
symptomatic intracranial
score of greater than 15 and evidence of middle cerebral artery infarction hemorrhage should be
involving more than 50% of its territory, and in whom surgery can be performed treated initially with
within 48 hours of symptom onset. While the benefit of hemicraniectomy was cryoprecipitate.
studied in a younger population in these studies, the DESTINY-2 study
● While interpreting the
demonstrated this benefit even in patients older than 60 years of age.53 The levels of fibrinogen, it is
concern that a lifesaving hemicraniectomy for prolonged life would leave important to realize that it is
patients with a moderate to severe disability was addressed in a survey where one of the acute phase
health professionals were given the options of treatment with consequent reactants and, hence, a
lower level is more reliable,
outcomes. The majority of the decision makers were agreeable to a
and a normal level should be
hemicraniectomy but were unlikely to accept the resultant quality of life. considered with healthy
After the clinical efficacy was explained, the majority did not consider skepticism.
hemicraniectomy but seemed to accept the resultant dependency.54 Therefore,
discussions with the patient’s next of kin should involve a description of ● Patients with acute
ischemic stroke with a
outcomes to provide clarity to make decisions for care. There is overwhelming National Institutes of Health
practice to perform surgical decompressions of cerebellar stroke. In a study of Stroke Scale score of
patients with cerebellar infarcts involving 25% to 33% of a hemisphere without greater than 15, altered
brainstem strokes, prophylactic suboccipital decompression prevented sensorium, and infarction of
more than 50% of the middle
significant neurologic deterioration.55 cerebral artery territory
should be considered
SEIZURES. Seizures seem to affect around 2% of the acute ischemic stroke for prophylactic
population within 24 hours.56 Initial resuscitation should focus on circulation, hemicraniectomy within
48 hours.
airway, and breathing followed by benzodiazepines such as IV lorazepam or IM
midazolam. The use of loading antiepileptic drugs followed by IV continuous ● Patients with acute
sedative agents for the management of refractory status epilepticus should be ischemic stroke who present
considered along with EEG monitoring.57 For more information, refer to the with cerebellar strokes
involving more than 25%
article “Status Epilepticus, Refractory Status Epilepticus, and Super-refractory
to 33% of a hemisphere
Status Epilepticus” by Sarah E. Nelson, MD, and Panayiotis N. Varelas, MD, should be considered for
PhD, FNCS, FAAN,58 in this issue of Continuum. suboccipital decompressive
craniectomy.
Cardiopulmonary Complications
Myocardial infarction in the setting of acute ischemic stroke needs careful
management. Hemodynamic management should be personalized based on the
systemic, cardiac, and cerebral perfusion needs for that patient. Use of
anticoagulants and antiplatelet agents to treat cardiac and pulmonary thrombotic
episodes should be evaluated based on the volume of the cerebral infarct,
hemorrhagic risks, and cardiac status. Although anticoagulation in patients with
acute ischemic stroke can be associated with increased risk of hemorrhagic
stroke, studies indicate that when initiated between 4 and 14 days, complications
can be minimized.55 Coronary revascularization procedures also should be

carefully considered as these entail the use of dual antiplatelet agents and
should be carefully coordinated with cardiologists.
Tachyarrhythmias and bradyarrhythmias are frequently seen in patients with
acute ischemic stroke. Atrial fibrillation is commonly seen and is best managed
with rate control aiming for less than 110 beats/min.59 Short-acting IV
beta-blockers and calcium channel blockers, digoxin and amiodarone, should be
considered. Atropine, cardioselective beta-agonists, and electrical pacing should
be the mainstay of treating bradyarrhythmias. Patients with mechanical implanted
devices pose an immediate need for resumption of anticoagulants, which should
be balanced against the risk of hemorrhagic transformation. Coordinated
management along with an experienced cardiac team is recommended.
Pulmonary complications usually stem from altered mentation and the
inability to protect the airway, aspiration pneumonia, and underlying primary
pulmonary pathology. In the neurocritical care unit, the incidence of pneumonia
in patients with stroke is variably reported between 10% and 56%.36 With the
increasing incidence of obstructive sleep apnea, there has been an increase in
the use of noninvasive positive pressure ventilation. In a meta-analysis, no
significant difference in secondary stroke or other vascular morbidity or
mortality was seen in patients treated with noninvasive positive pressure
ventilation, but there seemed to be an overall improvement in the
clinical outcomes.60
Many centers perform routine endotracheal intubation for patients
undergoing mechanical embolectomy. However, post hoc analysis of the MR
CLEAN study demonstrated clearly worse outcomes in the routine use of general
anesthesia in patients undergoing mechanical thrombectomy compared to
patients undergoing conscious sedation.61 In the SIESTA (Sedation Versus
Intubation for Endovascular Stroke Treatment) study, the primary outcome
was defined as a neurologic improvement within 24 hours. In that respect, no
difference was found between patients undergoing general anesthesia and
conscious sedation. However, many complications occurred in the general
anesthesia group, such as delayed extubation, hypothermia, and pneumonia.
Surprisingly, 3-month outcomes as measured by mRS were significantly better
in the general anesthesia group.62
In a case series, as many as 14% of patients with acute ischemic stroke had
respiratory insufficiency requiring mechanical ventilation.63 Furthermore, when
compared to patients without respiratory insufficiency, patients who were
ventilated had 1.4 times higher risk of 1-year mortality. This risk was further
exemplified if the patients were stuporous (2.6 times) or if they had absent
corneal reflexes or ischemic heart disease (3.4 times).64 The authors of this
article discourage the routine use of endotracheal intubation and mechanical
ventilation in the management of patients with acute ischemic stroke.
In patients who are intubated, it is unclear if early tracheostomy can
facilitate early mobilization and enhance early recovery. The SETPOINT2
(Stroke-related Early Tracheostomy Versus Prolonged Orotracheal Intubation
in Neurocritical Care Trial) study is actively enrolling patients to address
this question.65
FUTURE OF ACUTE ISCHEMIC STROKE TREATMENT

This is an exciting time for stroke research as new clinical trials and
observational studies provide new evidence to enhance the treatment options
1678 DECEMBER 2018

for patients with acute ischemic stroke. DEFUSE 3 has recently announced a KEY POINTS
favorable outcome in patients selected by imaging criteria for embolectomy
● Short-acting IV agents
when presenting between 6 and 16 hours of symptoms. It is anticipated that should be used to provide
more patients with acute ischemic stroke will be eligible for embolectomy, as optimal hemodynamic
the WAKE-UP (Efficacy and Safety of MRI-Based Thrombolysis in Wake-up management in patients with
Stroke) and EXTEND (Extending the Time for Thrombolyis in Emergency acute ischemic stroke.
Neurological Deficits [International]) trials address the ability to perform
● Early mobilization in the
mechanical embolectomy in patients with wake-up stroke. Furthermore, intensive care unit should be
studies are ongoing to determine the ability to triage these patients earlier encouraged in patients with
and more accurately by using a mobile stroke unit where initiation of acute ischemic stroke.
thrombolysis can be performed at the site. Expansion of telestroke networks is
● When possible,
also anticipated and thus will certainly rearrange local and regional stroke endotracheal intubation and
systems of care to adapt to new treatment paradigms. Further research general anesthesia
addressing the metabolic demands of patients with acute ischemic stroke via the administration should be
SHINE (Stroke Hyperglycemia Insulin Network Effort) study and early avoided for patients
undergoing mechanical
tracheostomy and mobilization through the SETPOINT2 clinical trial are thrombectomy.
anticipated soon.
CONCLUSION
This article is meant to highlight the main points in the critical care management
of patients with acute ischemic stroke and supplement the information in the
Continuum series addressing the management of acute ischemic stroke.66
Significant advances have been made in the treatment of patients with acute
ischemic stroke over the last decade. In 1995, initiation of thrombolysis was once
the only mainstay of treatment. The armamentarium to tackle stroke has taken a
significant step forward with the interventional trials, which have extended the
treatment windows from 4.5 hours through 24 hours. Despite these strides, we
still are left with many unanswered questions in the management of patients
with acute ischemic stroke. Peri-interventional anesthetic and hemodynamic
management still need further clarification. A large population still exists who
are not eligible for thrombectomy or thrombolysis who will need novel treatment
strategies. We are still unsure about the optimal hemodynamic management in
patients with acute ischemic stroke with or without thrombectomy. Clinical
management is driven by the symptom-based response, but no clear
neuromonitoring strategies have been developed to predict neurologic
worsening. A noninvasive measurement such as dynamic autoregulation seems
to have promising potential to intervene prior to neurologic decompensation.
Further research should be conducted to optimize stroke outcomes based on
hemodynamic management.
It is still unclear how to identify patients who will develop symptomatic
intracerebral hemorrhage. The management of such hemorrhages currently is
reflexive use of plasma products without clear evidence for improvement in
outcomes. Especially in patients with mechanical cardiac valves and support
devices, balancing between symptomatic intracranial hemorrhage expansion and
cardiac protection is a tough clinical decision. Clarity for management is very
much desired here.
Cerebral edema has been conventionally managed with osmotherapy and
surgical decompression. Newer approaches to reduce edema formation and
prevent secondary cerebral injury are warranted, especially in patients who

would not be candidates for either treatment. Furthermore, the burden of

addressing these issues has to be shared by the health care community.
Enrolling patients in trials whenever possible should be actively pursued.
Interest has grown in a collaborative approach between specialties and societies
to address many such issues, and we anticipate that such continued efforts will
make strides in caring for our patients with stroke.
USEFUL WEBSITE
NEUROCRITICAL CARE SOCIETY: EMERGENCY
NEUROLOGICAL LIFE SUPPORT
Refer to the Emergency Neurological Life Support
website to obtain further information on initial
management of neurologic emergencies.
neurocriticalcare.org/enls
REFERENCES
1 Mozaffarian D, Benjamin EJ, Go AS, et al. Heart 10 Lyden P. Using the National Institutes of Health
disease and stroke statistics–2015 update: a Stroke Scale: a cautionary tale. Stroke 2017;48(2):
report from the American Heart Association. 513–519. doi:10.1161/STROKEAHA.116.015434.
Circulation 2015;131(4):e29–e322. doi:10.1161/
11 National Institute of Neurological Disorders and
CIR.0000000000000152.
Stroke rt-PA Stroke Study Group. Tissue
2 Behavioral Risk Factor Surveillance System 2013 plasminogen activator for acute ischemic stroke.
survey data and documentation. Centers for N Engl J Med 1995;333(24):1581–1587. doi:10.1056/
Disease Control and Prevention. cdc.gov/brfss/ NEJM199512143332401.
annual_data/annual_2013.html. Updated June 23,
12 Hacke W, Kaste M, Bluhmki E, et al. Thrombolysis
2014. Accessed October 10, 2018.
with alteplase 3 to 4.5 hours after acute ischemic
3 Writing Group Members, Mozaffarian D, stroke. N Engl J Med 2008;359(13):1317–1329.
Benjamin EJ, et al. Heart Disease and Stroke doi:10.1056/NEJMoa0804656.
Statistics–2016 update: a report from the
13 IST-3 collaborative group, Sandercock P,
American Heart Association. Circulation 2016;133(4):
Wardlaw JM, et al. The benefits and harms of
e38–e360. doi:10.1161/CIR.0000000000000350.
intravenous thrombolysis with recombinant
4 Stroke Unit Trialists’ Collaboration. Organised tissue plasminogen activator within 6 h of acute
inpatient (stroke unit) care for stroke. Cochrane ischaemic stroke (the third international stroke
Database Syst Rev 2013(9):CD000197. trial [IST-3]): a randomised controlled trial.
doi:10.1002/14651858.CD000197.pub3. Lancet 2012;379(9834):2352–2363. doi:10.1016/
S0140-6736(12)60768-5.
5 Bershad EM, Feen ES, Hernandez OH, et al.
Impact of a specialized neurointensive care team 14 Thomalla G, Simonsen CZ, Boutitie F, et al. MRI-
on outcomes of critically ill acute ischemic stroke guided thrombolysis for stroke with unknown
patients. Neurocrit Care 2008;9(3):287–292. time of onset. N Engl J Med 2018;379(7):611–622.
doi:10.1007/s12028-008-9051-5. doi:10.1056/NEJMoa1804355.
6 Kramer AH, Zygun DA. Do neurocritical care units 15 Campbell BCV, Mitchell PJ, Churilov L, et al.
save lives? Measuring the impact of specialized Tenecteplase versus alteplase before
ICUs. Neurocrit Care 2011;14(3):329–333. thrombectomy for ischemic stroke. N Engl J Med
doi:10.1007/s12028-011-9530-y. 2018;378(17):1573–1582. doi:10.1056/NEJMoa1716405.
7 Goldstein LB. IV tPA for acute ischemic stroke: 16 Berkhemer OA, Fransen PS, Beumer D, et al. A
times are changing. Neurology 2016;87(21): randomized trial of intraarterial treatment for
2178–2179. doi:10.1212/WNL.0000000000003366. acute ischemic stroke. N Engl J Med 2015;372(1):
11–20. doi:10.1056/NEJMoa1411587.
8 Faigle R, Marsh EB, Llinas RH, et al. ICAT: a simple
score predicting critical care needs after 17 Goyal M, Demchuk AM, Menon BK, et al.
thrombolysis in stroke patients. Crit Care 2016; Randomized assessment of rapid endovascular
20:26. doi:10.1186/s13054-016-1195-7. treatment of ischemic stroke. N Engl J Med 2015;
372(11):1019–1030. doi:10.1056/NEJMoa1414905.
9 Papamichalis P, Karagiannis S, Dardiotis E, et al.
Predictors of need for critical care support, 18 Campbell BC, Mitchell PJ, Kleinig TJ, et al.
adverse events, and outcome after stroke Endovascular therapy for ischemic stroke with
thrombolysis. J Stroke Cerebrovasc Dis perfusion-imaging selection. N Engl J Med 2015;
2018;27(3):591–598. doi:10.1016/j. 372(11):1009–1018. doi:10.1056/NEJMoa1414792.
jstrokecerebrovasdis.2017.09.042.
1680 DECEMBER 2018

19 Saver JL, Goyal M, Bonafe A, et al. Stent-retriever 32 Lansberg MG, Albers GW, Wijman CA.
thrombectomy after intravenous t-PA vs. t-PA Symptomatic intracerebral hemorrhage
alone in stroke. N Engl J Med 2015;372(24): following thrombolytic therapy for acute
2285–2295. doi:10.1056/NEJMoa1415061. ischemic stroke: a review of the risk factors.
Cerebrovasc Dis 2007;24(1):1–10. doi:10.
20 Jovin TG, Chamorro A, Cobo E, et al.
1159/000103110.
Thrombectomy within 8 hours after symptom
onset in ischemic stroke. N Engl J Med 2015; 33 Xu Y, Hackett ML, Chalmers J, et al. Frequency,
372(24):2296–2306. doi:10.1056/NEJMoa1503780. determinants, and effects of early seizures after
thrombolysis for acute ischemic stroke: the
21 Nogueira RG, Jadhav AP, Haussen DC, et al.
ENCHANTED trial. Neurol Clin Pract 2017;7(4):
Thrombectomy 6 to 24 hours after stroke
324–332. doi:10.1212/CPJ.0000000000000384.
with a mismatch between deficit and infarct.
N Engl J Med 2018;378(1):11–21. doi:10.1056/ 34 Prosser J, MacGregor L, Lees KR, et al. Predictors
NEJMoa1706442. of early cardiac morbidity and mortality after
ischemic stroke. Stroke 2007;38(8):2295–2302.
22 Albers GW, Mark MP, Kemp S, et al.
doi:10.1161/STROKEAHA.106.471813.
Thrombectomy for stroke at 6 to 16 hours with
selection by perfusion imaging. N Engl J Med 35 Wira CR 3rd, Rivers E, Martinez-Capolino C, et al.
2018;378(8):708–718. doi:10.1056/NEJMoa1713973. Cardiac complications in acute ischemic stroke.
West J Emerg Med 2011;12(4):414–420. doi:10.5811/
23 Wardlaw JM, Murray V, Berge E, del Zoppo GJ.
westjem.2011.2.1765.
Thrombolysis for acute ischaemic stroke.
Cochrane Database Syst Rev 2014(7):CD000213. 36 Hannawi Y, Hannawi B, Rao CP, et al. Stroke-
doi:10.1002/14651858.CD000213.pub3. associated pneumonia: major advances and
obstacles. Cerebrovasc Dis 2013;35(5):430–443.
24 Furlan A, Higashida R, Wechsler L, et al. Intra-arterial
doi:10.1159/000350199.
prourokinase for acute ischemic stroke. The
PROACT II study: a randomized controlled trial. 37 Reinhard M, Rutsch S, Lambeck J, et al. Dynamic
Prolyse in acute cerebral thromboembolism. cerebral autoregulation associates with infarct
JAMA 1999;282(21):2003–2011. doi:10.1001/ size and outcome after ischemic stroke. Acta
jama.282.21.2003. Neurol Scand 2012;125(3):156–162. doi:10.1111/
j.1600-0404.2011.01515.x.
25 Broderick JP, Palesch YY, Demchuk AM, et al.
Endovascular therapy after intravenous t-PA 38 Petersen NH, Ortega-Gutierrez S, Reccius A, et al.
versus t-PA alone for stroke. N Engl J Med 2013; Dynamic cerebral autoregulation is transiently
368(10):893–903. doi:10.1056/NEJMoa1214300. impaired for one week after large-vessel acute
ischemic stroke. Cerebrovasc Dis 2015;39(2):
26 Ciccone A, Valvassori L, Nichelatti M, et al.
144–150. doi:10.1159/000368595.
Endovascular treatment for acute ischemic
stroke. N Engl J Med 2013;368(10):904–913. 39 Panerai RB, Jara JL, Saeed NP, et al. Dynamic
doi:10.1056/NEJMoa1213701. cerebral autoregulation following acute
ischaemic stroke: comparison of transcranial
27 Kidwell CS, Jahan R, Gornbein J, et al. A trial of
Doppler and magnetic resonance imaging
imaging selection and endovascular treatment
techniques. J Cereb Blood Flow Metab 2016;
for ischemic stroke. N Engl J Med 2013;368(10):
36(12):2194–2202.
914–923. doi:10.1056/NEJMoa1212793.
40 Vemmos KN, Tsivgoulis G, Spengos K, et al.
28 Alberts MJ, Latchaw RE, Jagoda A, et al.
U-shaped relationship between mortality and
Revised and updated recommendations for the
admission blood pressure in patients with acute
establishment of primary stroke centers: a
stroke. J Intern Med 2004;255(2):257–265.
summary statement from the brain attack
doi:10.1046/j.1365-2796.2003.01291.x.
coalition. Stroke 2011;42(9):2651–2665. doi:10.1161/
STROKEAHA.111.615336. 41 Sandset EC, Bath PM, Boysen G, et al. The
angiotensin-receptor blocker candesartan
29 Jauch EC, Saver JL, Adams HP Jr, et al. Guidelines
for treatment of acute stroke (SCAST): a
for the early management of patients with acute
randomised, placebo-controlled, double-blind
ischemic stroke: a guideline for healthcare
trial. Lancet 2011;377(9767):741–750. doi:10.1016/
professionals from the American Heart Association/
S0140-6736(11)60104-9.
American Stroke Association. Stroke 2013;44(3):
870–947. doi:10.1161/STR.0b013e318284056a. 42 Robinson TG, Potter JF, Ford GA, et al. Effects
of antihypertensive treatment after acute
30 Fuhrer H, Reinhard M, Niesen WD. Paradigm
stroke in the Continue or Stop Post-Stroke
change? Cardiac output better associates with
Antihypertensives Collaborative Study
cerebral perfusion than blood pressure in
(COSSACS): a prospective, randomized, open,
ischemic stroke. Front Neurol 2017;8:706.
blinded-endpoint trial. Lancet Neurol 2010;9(8):
doi:10.3389/fneur.2017.00706.
767–775. doi:10.1016/S1474-4422(10)70163-0.
31 Battey TW, Karki M, Singhal AB, et al. Brain edema
43 He J, Zhang Y, Xu T, et al. Effects of immediate
predicts outcome after nonlacunar ischemic
blood pressure reduction on death and major
stroke. Stroke 2014;45(12):3643–3648. doi:10.1161/
disability in patients with acute ischemic stroke:
the CATIS randomized clinical trial. JAMA 2014;
311(5):479–489. doi:10.1001/jama.2013.282543.

44 ENOS Trial Investigators. Efficacy of nitric oxide, 55 Kim MJ, Park SK, Song J, et al. Preventive
with or without continuing antihypertensive suboccipital decompressive craniectomy for
treatment, for management of high blood cerebellar infarction: a retrospective-matched
pressure in acute stroke (ENOS): a partial- case-control study. Stroke 2016;47(10):
factorial randomised controlled trial. Lancet 2565–2573. doi:10.1161/STROKEAHA.116.014078.
2015;385(9968):617–628. doi:10.1016/S0140-
56 Burn J, Dennis M, Bamford J, et al. Epileptic
6736(14)61121-1.
seizures after a first stroke: the Oxfordshire
45 Galyfos G, Sianou A, Filis K. Cerebral Community Stroke Project. BMJ 1997;315(7122):
hyperperfusion syndrome and intracranial 1582–1587. doi:10.1136/bmj.315.7122.1582.
hemorrhage after carotid endarterectomy or
57 Claassen J, Goldstein JN. Emergency
carotid stenting: a meta-analysis. J Neurol Sci
neurological life support: status epilepticus.
2017;381:74–82. doi:10.1016/j.jns.2017.08.020.
Neurocrit Care 2017;27(suppl 1):152–158.
46 Yaghi S, Willey JZ, Cucchiara B, et al. Treatment doi:10.1007/s12028-017-0460-1.
and outcome of hemorrhagic transformation
58 Nelson SE, Varelas PN. Status epilepticus,
after intravenous alteplase in acute ischemic
refractory status epilepticus, and super-refractory
stroke: a scientific statement for healthcare
status epilepticus. Continuum (Minneap Minn)
professionals from the American Heart
2018;24(6 Neurocritical Care):1683–1707.
Association/American Stroke Association.
Stroke 2017;48(12):e343–e361. doi:10.1161/ 59 Skanes AC, Healey JS, Cairns JA, et al. Focused
STR.0000000000000152. 2012 update of the Canadian Cardiovascular
Society atrial fibrillation guidelines:
47 Gore JM, Sloan M, Price TR, et al. Intracerebral
recommendations for stroke prevention and
hemorrhage, cerebral infarction, and subdural
rate/rhythm control. Can J Cardiol 2012;28(2):
hematoma after acute myocardial infarction and
125–136. doi:10.1016/j.cjca.2012.01.021.
thrombolytic therapy in the Thrombolysis in
Myocardial Infarction Study. Thrombolysis in 60 Tsivgoulis G, Alexandrov AV, Katsanos AH,
Myocardial Infarction, phase II, pilot and clinical Barlinn K, Mikulik R, Lambadiari V, et al.
trial. Circulation 1991;83(2):448–459. Noninvasive ventilatory correction in patients
with acute ischemic stroke: a systematic review
48 Yaghi S, Boehme AK, Dibu J, et al. Treatment and
and meta-analysis. Stroke 2017;48(8):2285–2288.
outcome of thrombolysis-related hemorrhage:
doi:10.1161/STROKEAHA.117.017661.
a multicenter retrospective study. JAMA Neurol
2015;72(12):1451–1457. doi:10.1001/jamaneurol. 61 Berkhemer OA, van den Berg LA, Fransen PS,
2015.2371. et al. The effect of anesthetic management
during intra-arterial therapy for acute stroke in
49 Claude Hemphill J 3rd, Lam A. Emergency
MR CLEAN. Neurology 2016;87(7):656–664.
neurological life support: intracerebral
doi:10.1212/WNL.0000000000002976.
hemorrhage. Neurocrit Care 2017;27(suppl 1):
89–101. doi:10.1007/s12028-017-0453-0. 62 Schönenberger S, Uhlmann L, Hacke W, et al.
Effect of conscious sedation vs general
50 Hashimoto T, Matsumoto S, Ando M, et al.
anesthesia on early neurological improvement
Cerebral hyperperfusion syndrome after
among patients with ischemic stroke undergoing
endovascular reperfusion therapy in a
endovascular thrombectomy: a randomized
patient with acute internal carotid artery
clinical trial. JAMA 2016;316(19):1986–1996.
and middle cerebral artery occlusions.
doi:10.1001/jama.2016.16623.
World Neurosurg 2018;110:145–151.
doi:10.1016/j.wneu.2017.11.023. 63 Dziewas R, Ritter M, Schilling M, et al. Pneumonia
in acute stroke patients fed by nasogastric tube.
51 Siegal DM, Curnutte JT, Connolly SJ, et al.
J Neurol Neurosurg Psychiatry 2004;75(6):852–856.
Andexanet alfa for the reversal of factor Xa
inhibitor activity. N Engl J Med 2015;373(25): 64 Santoli F, De Jonghe B, Hayon J, et al. Mechanical
2413–2424. doi:10.1056/NEJMoa1510991. ventilation in patients with acute ischemic
stroke: survival and outcome at one year.
52 Vahedi K. Decompressive hemicraniectomy for
Intensive Care Med 2001;27(7):1141–1146.
malignant hemispheric infarction. Curr Treat
Options Neurol 2009;11(2):113–119. 65 Schönenberger S, Niesen WD, Fuhrer H, et al.
Early tracheostomy in ventilated stroke patients:
53 Jüttler E, Unterberg A, Woitzik J, et al.
study protocol of the international multicentre
Hemicraniectomy in older patients with
randomized trial SETPOINT2 (stroke-related early
extensive middle-cerebral-artery stroke.
tracheostomy vs. prolonged orotracheal
N Engl J Med 2014;370(12):1091–1100.
intubation in neurocritical care trial 2). Int J Stroke
doi:10.1056/NEJMoa1311367.
2016;11(3):368–379. doi:10.1177/1747493015616638.
54 Honeybul S, Ho KM, Blacker DW. ORACLE stroke
66 Coplin WM. Critical care management of acute
study: opinion regarding acceptable outcome
ischemic stroke. Continuum (Minneap Minn)
following decompressive hemicraniectomy for
2012;18(3, Critical Care Neurology):547–559.
ischemic stroke. Neurosurgery 2016;79(2):
doi:10.1212/01.CON.0000415427.53653.1b.
231–236. doi:10.1227/NEU.0000000000001115.
1682 DECEMBER 2018

REVIEW ARTICLE
Status Epilepticus, 
Refractory Status C O N T I N U UM A U D I O
ONLINE
Epilepticus, and
Super-refractory Status
Epilepticus CITE AS:
By Sarah E. Nelson, MD; Panayiotis N. Varelas, MD, PhD, FNCS, FAAN CONTINUUM (MINNEAP MINN)
1683–1707.
ABSTRACT Dr Sarah E. Nelson, Johns
Hopkins University, 600 N Wolfe
PURPOSE OF REVIEW: Status epilepticus, refractory status epilepticus, and St, Phipps 455, Baltimore, MD
super-refractory status epilepticus can be life-threatening conditions. This 21287, snelso43@jhmi.edu.
article presents an overview of the three conditions and discusses their
management and outcomes. Dr Nelson receives grant
support from the Johns Hopkins
RECENT FINDINGS: Status epilepticus was previously defined as lasting for Anesthesiology and Critical Care
Medicine (ACCM) Stimulating
30 minutes or longer but now is more often defined as lasting 5 minutes and Advancing ACCM Research
or longer. A variety of potential causes exist for status epilepticus, (StAAR) program. Dr Varelas
refractory status epilepticus, and super-refractory status epilepticus, serves on the board of directors
of the Neurocritical Care
but all three ultimately involve changes at the cellular and molecular Society, on the editorial board
level. Management of patients with status epilepticus generally of Neurocritical Care, and on an
advisory board of Portola
requires several studies, with EEG of utmost importance given the Pharmaceuticals, Inc.
pathophysiologic changes that can occur during the course of status Continued on page 1707
epilepticus. Status epilepticus is treated with benzodiazepines as
first-line antiepileptic drugs, followed by phenytoin, valproic acid, or UNLABELED USE OF
levetiracetam. If status epilepticus does not resolve, these are followed USE DISCLOSURE:
by an IV anesthetic and then alternative therapies based on limited Drs Nelson and Varelas discuss
data/evidence, such as repetitive transcranial magnetic stimulation, the unlabeled/investigational
use of allopregnanolone, deep
therapeutic hypothermia, immunomodulatory agents, and the ketogenic brain stimulation, desflurane,
diet. Scores have been developed to help predict the outcome of status diazepam, electroconvulsive
epilepticus. Neurologic injury and outcome seem to worsen as the duration therapy, fosphenytoin,
gabapentin, isoflurane, IV
of status epilepticus increases, with outcomes generally worse in immunoglobulin, ketamine,
super-refractory status epilepticus compared to status epilepticus and ketogenic diet, lacosamide,
levetiracetam, lidocaine,
sometimes also to refractory status epilepticus.
lorazepam, methylprednisolone,
midazolam, pentobarbital,
SUMMARY: Status epilepticus can be a life-threatening condition associated phenobarbital, phenytoin,
plasma exchange, propofol,
with multiple complications, including death, and can progress to pyridoxine, thiopental,
refractory status epilepticus and super-refractory status epilepticus. More topiramate, vagal nerve
studies are needed to delineate the best management of these three stimulation, and valproic acid
for the treatment of refractory
entities. status epilepticus.

of Neurology.

STATUS EPILEPTICUS
KEY POINTS
INTRODUCTION
S
● While in the past, tatus epilepticus is a common neurologic emergency that requires
tonic-clonic status prompt treatment to decrease morbidity and mortality. Best
epilepticus was defined as management continues to evolve for this condition. This article
continuous seizure activity
or two or more seizures
reviews the current definitions of status epilepticus, refractory status
without recovery of epilepticus, and super-refractory status epilepticus as well as their
consciousness lasting longer epidemiology, etiology, pathophysiology, diagnosis, management, and
than 30 minutes, outcomes. It is important to note that guidelines for the evaluation and treatment
tonic-clonic status
of status epilepticus have been published by the Neurocritical Care Society.1
epilepticus is now defined
as seizure activity lasting Treatment guidelines have also recently been published by the American
5 minutes or longer, with Epilepsy Society, but they stop short of addressing refractory status epilepticus
30 minutes being the cutoff and super-refractory status epilepticus.2
for development of
long-term consequences.
DEFINITIONS
● No standard definition for Typical seizures are fewer than 5 minutes in duration and are self-limited, but
nonconvulsive status those that are longer tend not to resolve on their own.1 Traditionally, status
epilepticus currently exists.
A working definition epilepticus was defined as continuous seizure activity or two or more seizures
differentiates diagnostic without recovery of consciousness lasting longer than 30 minutes. However, status
criteria based on whether epilepticus is now often defined by two time points, t1 and t2, identified as the time
epileptic encephalopathy beyond which seizures are likely to be prolonged and the time beyond which
is present.
seizures lead to long-term consequences, respectively; for tonic-clonic seizures,
● Refractory status t1 is 5 minutes and t2 is 30 minutes, but for focal status epilepticus or absence status
epilepticus is continuous these time points are different or unknown.2,3
seizure activity not Status epilepticus has several subtypes, including convulsive status
controlled by first-line and
epilepticus, epilepsia partialis continua, and nonconvulsive status epilepticus. In
second-line antiepileptic
drugs; it occurs in 9% to 43% convulsive status epilepticus, repetitive tonic-clonic movements occur, followed
of all cases of status by a postictal state. In epilepsia partialis continua, focal neurologic deficits such
epilepticus. as aphasia and motor dysfunction occur as a result of partial seizures, but altered
mental status is not present. Continuous or fluctuating mental status changes
● Super-refractory status
epilepticus is defined either occur in nonconvulsive status epilepticus.2 Although no definitive criteria for
as status epilepticus not nonconvulsive status epilepticus exist, a working definition is listed in
TABLE 5-1.
4,5
controlled by third-line
anesthetic agents or as Refractory status epilepticus is continuous seizure activity not controlled by
status epilepticus continuing
for 24 hours or longer after
first-line and second-line antiepileptic drugs (AEDs).6 Super-refractory status
anesthesia is administered. epilepticus is defined as status epilepticus not controlled by third-line agents.7
The exact incidence and Another definition posits that super-refractory status epilepticus exists if status
associated mortality of epilepticus continues for 24 hours or longer after anesthesia is administered.8
super-refractory status
epilepticus are unknown.
EPIDEMIOLOGY
● The annual incidence of The next section reviews the epidemiology of status epilepticus, refractory status
status epilepticus is
epilepticus, and super-refractory status epilepticus.
approximately 12.6 per
100,000 person-years and is
increasing over time. Seizures Status Epilepticus
or status epilepticus may The pooled crude annual incidence rate of status epilepticus is approximately 12.6
occur in up to 19% of patients
hospitalized in the intensive
per 100,000 person-years.9 In the intensive care unit (ICU) specifically, seizures
care unit. or status epilepticus may occur in up to 19% of patients, although studies
evaluating this frequency are limited by their retrospective nature and different
definitions.10 Of those with status epilepticus, 12% to 43% progress to refractory
status epilepticus and 10% to 15% progress to super-refractory status epilepticus.8
1684 DECEMBER 2018

Status epilepticus incidence appears to peak at ages older than 50 years
(approximately 28.4 per 100,000 per year) and younger than 10 years (14.3 per
100,000 per year) and seems to be greater in African Americans (13.7 per
100,000 per year) compared to whites (6.9 per 100,000 per year) and other
races (7.4 per 100,000 per year).11 Although previous studies found a greater
incidence of status epilepticus in males,12–14 recent estimates suggest that the
rates for both sexes are more similar than previously thought (11.1 per 100,000
person-years in females, 11.3 per 100,000 person-years in males).9 The overall
case fatality rate appears to be close to 15%, with a greater case fatality rate
occurring in the elderly (24.9%) and in patients with refractory status
epilepticus (33.3%).9
The incidence of status epilepticus appears to be increasing over time. In a
study evaluating data from the US National Hospital Discharge Survey, between
1979 and 2010, the incidence of status epilepticus was found to increase from 3.5
per 100,000 per year to 12.5 per 100,000 per year but with no significant change
in in-hospital mortality.11 In a 2015 study that used data from the Centers for
Disease Control and Prevention and the Nationwide Inpatient Sample, status
epilepticus hospitalizations increased by 56.4% from 1999 (8.9 per 100,000
persons) to 2010 (13.9 per 100,000 persons). Mortality also increased over
this same time period, but only by 5.6% (1.8 per 1 million persons to 1.9 per
1 million persons).15
In the United States, 120,000 to 180,000 convulsive status epilepticus
episodes occur annually,16 but the incidence of nonconvulsive status epilepticus
is not as well established.17 Several studies in the late 1990s and early 2000s more
clearly delineated the frequency of seizure types. In a prospective study of status
epilepticus in Richmond, Virginia, seizure types were determined to be
generalized in 74% and partial in 26% of cases.12 A study evaluating status
Criteria for Nonconvulsive Status Epilepticusa TABLE 5-1
Patients Without Epileptic Encephalopathyb

◆ Epileptic discharges greater than 2.5 Hz
OR
◆ Epileptic discharges greater than 2.5 Hz or rhythmic delta/theta activity and one of the
following:
◇ Subtle clinical ictal phenomena during the above EEG patterns
◇ Spatiotemporal evolution
◇ EEG and clinical improvement after IV antiepileptic drugs
Patients With Epileptic Encephalopathyb
◆ Increase in frequency or prominence of the above EEG findings compared to baseline
along with change in clinical state
◆ EEG and clinical improvement with IV antiepileptic drugs
EEG = electroencephalogram; IV = intravenous.

a
Modified with permission from Beniczky S, et al, Epilepsia.4 © 2013 John Wiley and Sons.
b
Epileptic encephalopathy indicates that epileptic activity contributes to behavioral and cognitive
problems in excess of what would be expected from the underlying pathology.5

STATUS EPILEPTICUS
KEY POINT epilepticus in French-speaking Switzerland found that the most common seizure
type was partial (44.8%), followed by generalized tonic-clonic (33.1%).13
● The etiology of status
epilepticus may be divided
into known or symptomatic Refractory and Super-refractory Status Epilepticus
causes and unknown or Refractory status epilepticus occurs in 9% to 43% of all cases of status
cryptogenic causes. In epilepticus.6,7 In one study of 395 patients with refractory status epilepticus
general, acute causes
treated in the ICU, the annual incidence was found to be 3.4 per 100,000.18 As
appear to be more common
than chronic causes. expected, prospective studies19 estimate a lower incidence than retrospective
studies.20–22 Predictors of refractory status epilepticus were lower level of
consciousness and new diagnosis of status epilepticus in one study19 and focal
motor seizures at onset and nonconvulsive status epilepticus in another.21
The exact incidence and associated mortality of super-refractory status
epilepticus have not yet been delineated, likely because of the low number of
patients with this condition and lack of prospective studies.23 One estimate is that
10% to 15% of all in-hospital cases of status epilepticus will evolve into
super-refractory status epilepticus.23 In a 2015 study that used the Finnish
Intensive Care Consortium database, 22% of patients with refractory status
epilepticus were categorized as having super-refractory status epilepticus, with
an annual incidence estimate of 0.7 per 100,000 persons.8 Of 98 patients in West
China diagnosed with status epilepticus, 12.2% had super-refractory status
epilepticus (by comparison, the percentages of patients with nonrefractory
status epilepticus and refractory status epilepticus were 67.3% and 20.4%,
respectively). In this study, convulsive status epilepticus was the main seizure
type in patients with super-refractory status epilepticus (as compared to
nonconvulsive status epilepticus).24
Another study examined 177 patients with convulsive status epilepticus in
India. While 105 (59.3%) patients had nonrefractory status epilepticus, 72
(40.7%) had refractory status epilepticus, of which 30 (16.9% of the total 177)
had super-refractory status epilepticus. Super-refractory status epilepticus was
more common in children and the elderly.25 In a study of patients with status
epilepticus who did not respond to first-line AEDs, seven episodes (20%) met the
authors’ definition of malignant status epilepticus (persistent clinical and/or
electrographic seizure activity that recurred within 5 days of weaning the
maximum dose of IV anesthetics that resulted in EEG burst suppression), which
appears to be similar to the definition of super-refractory status epilepticus.
Patients with malignant status epilepticus were younger than patients with
refractory status epilepticus.26
ETIOLOGY
Status epilepticus may have a structural, infectious, toxic-metabolic, or
autoimmune cause (TABLE 5-2).1 According to the International League Against
Epilepsy, the etiology of status epilepticus may be divided into two groups:
(1) known or symptomatic and (2) unknown or cryptogenic. The symptomatic
group can be subdivided into acute symptomatic, remote symptomatic, and
progressive symptomatic.3 In general, acute causes appear to be more common
than chronic causes.9 The underlying etiology for status epilepticus often
influences the likelihood of a patient’s mortality.15
Interestingly, etiologies seem to vary among different populations. In one
prospective population-based study of 150 patients with status epilepticus
conducted in Germany, the most common etiologies were a remote stroke
1686 DECEMBER 2018

(36.0%) or other remote symptomatic cause (26.7%).14 In a study analyzing
status epilepticus in Richmond, Virginia, the most common etiologies in adults
were low AED levels (34%), remote etiology (24%), and stroke (22%).12 A study
in Switzerland found that status epilepticus due to an acute etiology (eg, traumatic
brain injury, tumor, stroke, metabolic derangement; 62.7%) was more common
than status epilepticus due to a remote etiology (eg, chronic infection, autoimmune
disease, degenerative disease; 28.4%) or due to an unknown etiology (8.7%).13
The etiology of refractory status epilepticus appears to be similar overall
to that of nonrefractory status epilepticus. In one study of 54 patients who
experienced 63 episodes of refractory status epilepticus, low AED levels
(or missed doses) were the most common etiology, followed by metabolic causes
and central nervous system infection.27 Another study found that refractory
status epilepticus was more likely associated with encephalitis and nonrefractory
status epilepticus with low AED levels.20 Similarly, in a study examining the
differences between refractory status epilepticus and nonrefractory status
epilepticus, central nervous system infections were found in greater frequency in
patients with refractory status epilepticus; viral encephalitis, in particular, was
more common in this group.28
The etiology of super-refractory status epilepticus may be different than that
of status epilepticus and refractory status epilepticus.29 Several studies suggest
that encephalitis is a frequent cause of super-refractory status epilepticus
(CASE 5-1).24–26 Encephalitis was the most common etiology in super-refractory
status epilepticus at 66.7% (compared to 12.3% in nonrefractory status
epilepticus) in one study and was found to predict the progression of status
epilepticus to super-refractory status epilepticus.25 It was also identified as an
Possible Causes of Status Epilepticusa TABLE 5-2
Acute Causes
◆ Acute stroke (eg, ischemic stroke, intracerebral hemorrhage)
◆ Head trauma
◆ Central nervous system infections (eg, abscess, meningitis, encephalitis)
◆ Hypoxic brain injury
◆ Posterior reversible encephalopathy syndrome (PRES)
◆ Autoimmune and paraneoplastic etiologies (eg, anti–N-methyl-D-aspartate [NDMA] receptor
encephalitis)
◆ Sepsis
◆ Metabolic disturbances (eg, hypoglycemia, abnormal electrolytes)
◆ Drug withdrawal, toxicity, or noncompliance
Chronic Causes
◆ History of epilepsy (eg, due to noncompliance with antiepileptic drugs)
◆ Brain tumor
◆ Previous brain pathology (eg, due to trauma, stroke, cortical dysplasia)
a
Modified with permission from Brophy GM, et al, Neurocrit Care.1 © 2012 Springer Science+Business
Media, LLC.

STATUS EPILEPTICUS
CASE 5-1 A 24-year-old man presented with fever, nausea, headache, and
generalized tonic-clonic seizures. His past medical history was
unremarkable. He was admitted to the neurocritical care unit, where he
was placed on continuous EEG. CT and MRI of the brain and CSF analysis
(including multiple viral cultures and a paraneoplastic panel) were
unremarkable aside from a mild CSF pleocytosis. Body positron emission
tomography (PET) and testicular ultrasound were normal, but PET of the
brain showed occipital lobe hypometabolism (FIGURE 5-1), and this finding
can be seen in anti–N-methyl-D-aspartate (NMDA) receptor antibody
encephalitis, but despite extensive autoimmune testing he never tested
positive for any antibody.30 He was initially treated with 5 days of
methylprednisolone with no improvement; he was then given five
sessions of plasma exchange.
EEG initially showed electrographic seizures that started in the right
posterior temporal region with subsequent spread first to the rest of the
right hemisphere and then bilaterally (FIGURE 5-2). He had a clinical
correlate of left head deviation, nystagmus, and left face and limb tonic-
clonic activity with some of the seizures, but many had no clear clinical
correlate. He was monitored with continuous EEG for longer than a week
and required multiple medications, including IV levetiracetam,
phenytoin, lacosamide, and phenobarbital. He also required multiple
continuous anesthetic agents, including midazolam, propofol, and
ketamine, to achieve burst suppression. Because additional seizure
control was still needed, he was also started on the ketogenic diet; it was
discontinued 4 days later because of severe metabolic acidosis.
FIGURE 5-1
Positron emission tomography (PET) of the patient in CASE 5-1. The green areas show occipital
lobe hypometabolism, and normal brain metabolism is depicted in black and blue colors.
1688 DECEMBER 2018

FIGURE 5-2
EEG of the patient in CASE 5-1. Seizure onset in the right posterior temporal region (A) with CONTINUED ON
spread to the rest of the right hemisphere (B) and subsequently bilaterally (C). PAGE 1690

STATUS EPILEPTICUS
independent risk factor for malignant status epilepticus (which had a definition
similar to super-refractory status epilepticus in one study, as discussed above).26
In a status epilepticus cohort in China, the most common cause of super-refractory
status epilepticus was acute encephalitis (67.7% of super-refractory status
epilepticus cases).24
The term new-onset refractory status epilepticus (NORSE) has recently emerged
to define patients who have prolonged refractory status epilepticus with no
readily identifiable cause (although an autoimmune or viral encephalitis etiology
may later be found). In a multicenter retrospective study of patients for
whom seizure etiology was not known within 48 hours of admission, out of
675 cases of refractory status epilepticus, 130 cases fulfilled NORSE criteria
(19%). In 47% of these NORSE cases, an etiology was later determined, including
nonparaneoplastic autoimmune (40%), paraneoplastic (30%), and infectious
causes (16%). Anti–N-methyl-D-aspartate (NMDA) receptor antibodies were the
most common causes of both nonparaneoplastic autoimmune and paraneoplastic
etiologies, while herpes viruses (except herpes simplex virus type 1) were the
most frequent infectious cause. Approximately 64% of patients were females,
and 48% were older than 50 years of age. Prodromal symptoms, including
confusion, fever, fatigue, and headache, preceded NORSE onset in 46% of the
cases. Unilateral seizure onset on EEG was more common than bilateral
independent, multifocal, and generalized onsets, and EEG findings did not differ
between patients whose seizure etiology was determined and those whose was
not. MRI abnormalities were found in 62% of cases, most commonly in limbic or
neocortical structures, or both. CSF abnormalities were discovered in 73%, and
no difference was seen in MRI or CSF findings between those with and without
an identified etiology. The type and number of AEDs prescribed for both
cryptogenic and noncryptogenic cases were similar. Status epilepticus duration
was longer in patients whose seizures were of cryptogenic origin (8 days versus
4 days), but ICU and hospital stay duration were similar. Thirty-eight percent
of patients achieved a good or fair outcome (modified Rankin Scale [mRS]
score of 0 to 3) at discharge. Multivariate predictors of both poor outcome
(mRS score of >3) and mortality included the Status Epilepticus Severity Score
CONTINUED FROM After weaning from burst suppression, his EEG showed nearly
PAGE 1689 continuous multifocal sharp waves and diffuse generalized rhythmic
delta activity but no seizures. His examination steadily improved, and
he was successfully extubated a few days later. Antiepileptic drugs
were weaned, and he was eventually discharged on levetiracetam,
lacosamide, and a slow phenobarbital taper as well as oral prednisone for
his presumed diagnosis of autoimmune encephalitis. His EEG before
discharge showed intermittent left temporal slowing.
COMMENT This case illustrates that super-refractory status epilepticus from presumed
encephalitis may respond to immunomodulatory therapy and to multiple
antiepileptic drugs. It may need to be evaluated with extensive laboratory
workup and a PET scan in addition to traditional CSF studies and MRI of the brain.
1690 DECEMBER 2018

(STESS) (refer to the section on prognostic scores and the future of status KEY POINTS
epilepticus research later in this article), status epilepticus duration, and
● Encephalitis appears to
number of complications.31 be a common cause of both
refractory status epilepticus
PATHOPHYSIOLOGY and super-refractory status
At the cellular level, several changes seem to occur that allow a seizure to evolve epilepticus.
into and maintain as status epilepticus. Protein phosphorylation, the release of
● New-onset refractory
neurotransmitters, and the opening and closing of ion channels occur in status epilepticus (NORSE)
milliseconds to seconds after seizure onset. In the next seconds to minutes, has recently emerged as a
changes in receptor trafficking occur, including an increase in excitatory challenging disorder in
α-amino-3-hydroxy-5-methylisoxazole-4 propionic acid (AMPA) and NMDA which the etiology of
refractory status epilepticus
receptors and a decrease in inhibitory γ-aminobutyric acid (GABA)A β2, β3, and is not immediately apparent.
γ2 subunits. Resistance to benzodiazepines with time may be due to modulation
of these GABAA receptors. Next, over minutes to hours, alterations occur in ● Multiple changes at the
neuropeptide expression, including an increase in excitatory substance P and cellular and molecular level
occur in status epilepticus
insufficient presence of inhibitory neuropeptide Y, thus maintaining an excitable as time passes.
state. In the days to weeks after status epilepticus, genetic and epigenetic changes
occur, including alterations in expression of multiple genes. Changes in ● With increasing duration
microRNA regulation and DNA methylation may also occur.10 of seizure activity, the
frequency of systemic
With increasing duration of seizure activity, the frequency of systemic
complications, neurologic
complications, neurologic injury, and mortality increase.16 Neuronal injury injury, and mortality
occurring during status epilepticus has been demonstrated in animals.32 For increase.
example, neuronal injury in the neocortex, thalami, and hippocampi was
demonstrated in baboons in which convulsive seizures were induced.33 Neuronal
injury was also seen when baboons were paralyzed, suggesting that
nonconvulsive status epilepticus can also be harmful.34 In humans, a marker for
neuronal injury, serum neuron-specific enolase, has been shown to be increased
after convulsive and nonconvulsive status epilepticus.10 As demonstrated in
animal models, mechanisms thought to possibly contribute to neuronal injury
and cell death in status epilepticus include excitotoxicity, mitochondrial
dysfunction, necrosis, and apoptosis.10
Neuroimaging may reveal findings in the setting of status epilepticus that
likely reflect the pathophysiology of this condition. CT findings include cortical
edema and sulcal effacement, loss of gray-white matter differentiation, reduced
attenuation, and enhancement. On MRI, findings can include T2 hyperintensity,
restricted diffusion, and apparent diffusion coefficient changes similar to those
seen in a stroke. These changes can occur in the cortex, basal ganglia, thalami,
hippocampi, and corpus callosum.10 Leptomeningeal enhancement and crossed
cerebellar diaschisis (depression in metabolism and blood flow in the cerebellum
contralateral to a supratentorial lesion) can also be seen. Findings may resolve on
follow-up imaging (CASE 5-2), but hippocampal sclerosis and focal atrophy can
persist, thus suggesting permanent neuronal injury.10
DIAGNOSIS
The diagnosis of status epilepticus involves a combination of clinical suspicion as
well as laboratory testing, EEG, and imaging.
General Workup
Convulsive status epilepticus is a clinical diagnosis. When patients present
with suspected status epilepticus, recommended studies include vital signs,

STATUS EPILEPTICUS
CASE 5-2 A 78-year-old woman with a history of hypertension, coronary artery

disease, ischemic stroke with residual right hemiparesis, and a 10-year
history of seizures for which she was taking levetiracetam presented to
the emergency department after experiencing four consecutive
generalized seizures. Per her family, she had not taken her antiepileptic
drugs for 2 days. She was at a family member’s house celebrating her
78th birthday when she had a seizure. She had another seizure when
emergency medical services arrived and another two on the way to the
emergency department.
Her examination was significant for aphasia and right-sided weakness
that was worse than her baseline. She was given IV lorazepam and
levetiracetam while in the emergency department. She then underwent
an MRI of the brain that showed subtle cortical restricted diffusion in the
left parietal lobe involving the superior and inferior parietal lobules. It
also showed restricted diffusion in the posterior left thalamus in the
region of the pulvinar, with subtle corresponding fluid-attenuated
inversion recovery (FLAIR) hyperintensity in the thalamic abnormality, but
with only minimal FLAIR signal change involving the cortex (FIGURE 5-3).
Initial continuous EEG showed occasional brief electrographic seizures
that were maximal over the left hemisphere, frequent generalized
periodic discharges that were maximal over the left hemisphere, diffuse
background slowing in the delta-theta range, and background asymmetry
with higher amplitude and persistent focal slowing over the left
FIGURE 5-3
MRI of the patient in CASE 5-2. Axial diffusion-weighted (A) and fluid-attenuated inversion
recovery (FLAIR) (B) images showing slight restricted diffusion of the left parietal lobe and
left thalamus with corresponding subtle FLAIR abnormalities.
1692 DECEMBER 2018

FIGURE 5-4
EEGs of the patient in CASE 5-2. A, Initial EEG showing frequent generalized periodic
discharges, maximal on the left. B, Follow-up EEG showing left-sided periodic discharges.
hemisphere (FIGURE 5-4A). Her IV levetiracetam dose was increased, and,

with continued seizure activity on EEG, she was also loaded and started
on a maintenance dose of IV phenytoin. Because seizure activity
persisted, phenytoin was discontinued, however, and IV lacosamide and
valproic acid were added.
Follow-up EEG showed left-sided periodic discharges that eventually
decreased in frequency (FIGURE 5-4B). Her examination slowly improved,
and EEG was eventually discontinued. Repeat MRI brain showed that the
previous findings had resolved.
This case of status epilepticus illustrates that EEG does not have to be COMMENT
completely normal to discontinue EEG monitoring, especially if the clinical
examination is improving. In addition, this case demonstrates that
reversible MRI findings can be seen in the setting of status epilepticus.

STATUS EPILEPTICUS
laboratory tests (including complete blood cell count, basic metabolic

panel, calcium, magnesium, and glucose), AED levels (as indicated), head
CT, and EEG.1 Additional studies that may be helpful in determining the cause
of status epilepticus include a toxicology screen, brain MRI, lumbar puncture,
inborn errors of metabolism panel, additional imaging (eg, single-photon
emission CT [SPECT], positron emission tomography (PET) scan, magnetic
resonance spectroscopy), and a paraneoplastic and autoimmune
encephalitis workup.1,35
Electroencephalography
EEG is extremely important in the diagnosis of status epilepticus, especially in
cases of nonconvulsive status epilepticus, which may be suspected clinically
but proven only by EEG. For example, in one study of 570 patients in the ICU
who underwent continuous EEG for detection of subclinical seizures or
unexplained lower level of consciousness, seizures were detected in 19%; 92%
of these patients had only nonconvulsive seizures. Independent predictors for
discovering seizures on EEG were age younger than 18 years, convulsive seizures
before continuous EEG monitoring, coma, and a history of epilepsy.36 One
prospective study in a general hospital setting found that 19% of patients with
status epilepticus had nonconvulsive status epilepticus.37 Forty-seven percent of
patients in a tertiary care center were found to have nonconvulsive status
epilepticus.38 In a study of patients in a general ICU, nonconvulsive status
epilepticus was detected in 8% of patients who were comatose.39 A study of the
neurologic ICU population specifically found that 23 of 170 patients (13.5%) had
nonconvulsive status epilepticus.40
Importantly, studies evaluating the timing of seizures have found that
continuous EEG may not necessarily capture seizures in the first hours after it is
placed. In one study, while continuous EEG detected seizures within the first
24 hours of monitoring in 88% of patients who would eventually have seizures, in
a similar percentage (87%) of patients who were comatose, seizures were more
likely to be detected after more than 24 hours of monitoring, suggesting that
additional monitoring time may be required in patients who are comatose
longer.36 Another study found that approximately 97% of patients who were
hospitalized and noncritically ill had their first seizure within 24 hours of starting
continuous EEG.41
Imaging
The value of CT—and even more so MRI—in revealing a focal lesion as the
cause of status epilepticus has been well demonstrated.42 SPECT has also been
used to detect foci of status epilepticus. In several case series and case
reports of patients with status epilepticus, brain SPECT using technetium
99m–hexamethylpropyleneamine oxime (99mTc-HMPAO) or 99mTc–ethyl
cysteinate dimer (ECD) generally demonstrated focal hyperperfusion in areas
consistent with EEG findings.42–44 Compared to SPECT, PET (especially
PET/CT) can provide better resolution and the ability to perform quantitative
measurements.35 In one study of eight patients with focal status epilepticus,
fludeoxyglucose (FDG)-PET helped to support the diagnosis of status
epilepticus; determine localization of the epileptic focus for planning surgical
treatment; and clarify discordant findings among clinical, MRI, and EEG
findings.45
1694 DECEMBER 2018

MANAGEMENT KEY POINTS
Compared to other neurologic emergencies, such as stroke, treatment of status
● Continuous video-EEG is a
epilepticus is rife with poor-quality data supplied by only a small number of valuable tool given its ability
randomized controlled studies. However, the urgency of treating status to detect nonconvulsive
epilepticus is similar to that of stroke because the response to treatments and the status epilepticus.
prognosis diminishes with elapsed time. Therefore, a staged approach to
● Imaging modalities,
treatment has been advocated, with different drugs used in early (stage I),
including CT, MRI,
established (stage II), refractory (stage III), and super-refractory status single-photon emission CT,
epilepticus (stage IV).46 and positron emission
Several studies suggest that status epilepticus is an emergency that requires tomography, may help in the
immediate treatment. In a study of 48 patients with poor-grade aneurysmal diagnosis and management
of status epilepticus.
subarachnoid hemorrhage in a neurologic ICU with nonconvulsive seizures
who underwent multimodality monitoring, including intracortical EEG, ● Status epilepticus
intracortical seizures were detected in 38% of patients and surface seizures in 8%. requires urgent treatment
Increases in mean arterial pressure, heart rate, respiratory rate, and minute because the response to
treatments and the
ventilation were found to be associated with intracortical seizures, and trends of prognosis diminish with
association with greater intracranial pressure and cerebral perfusion pressure elapsed time. A staged
were seen.47 In a study of 34 patients with moderate to severe traumatic brain approach to treatment has
injury who received multimodality invasive monitoring, 21 patients had either thus been advocated, with
different medications used
pseudoperiodic discharges or seizures, all nonconvulsive. Metabolic crises, as
in early (stage I), established
characterized by elevated cerebral microdialysis lactate to pyruvate ratio and (stage II), refractory (stage
decreased glucose, were time-locked to episodes of pseudoperiodic discharges III), and super-refractory
and seizures.48 These data suggest that nonconvulsive epileptic phenomena may status epilepticus (stage IV).
not be benign.
Initial Management of Status Epilepticus

The efficacy of benzodiazepines as the initial treatment of status epilepticus is
well established. In a four-arm trial comparing (1) lorazepam, (2) phenobarbital,
(3) phenytoin, and (4) diazepam followed by phenytoin, lorazepam was more
successful than phenytoin in aborting status epilepticus in patients with
convulsive status epilepticus and in the combined group of patients with
convulsive status epilepticus and patients with nonconvulsive status epilepticus.
Comparisons among other arms of the trial were not significant.49
The efficacy of out-of-hospital administration of benzodiazepines in status
epilepticus was tested in a randomized trial in which adults with status epilepticus
were randomly assigned to IV lorazepam, IV diazepam, or placebo. The two
benzodiazepines were more successful in aborting status epilepticus by emergency
department arrival than placebo.50 Another study examining out-of-hospital use
of benzodiazepines found that administration of IM midazolam was noninferior to
IV lorazepam in terminating seizures before emergency department arrival in
patients with status epilepticus.51 A smaller and older randomized study found no
difference between the administration of IV diazepam and lorazepam in ending
seizure activity.52 Refer to TABLE 5-3 for recommended dosing of benzodiazepines
as first-line AEDs in status epilepticus.
The value of the other major AEDs used in status epilepticus has also been
investigated. A randomized open-label study of 79 patients with convulsive or
subtle convulsive status epilepticus found that lorazepam and levetiracetam had
similar efficacy in terminating clinical seizures.53 Patients with convulsive status
epilepticus randomly assigned to IV valproic acid or phenytoin experienced a
better seizure termination rate with valproic acid, but no difference was seen

STATUS EPILEPTICUS
between the two medications in terms of seizure freedom at 24 hours.54 A

randomized open-label study of 74 patients with status epilepticus or acute
repetitive seizures (defined as at least two seizures occurring over 5 to 6 hours
different from their usual pattern and not categorized as status epilepticus)
showed that valproic acid and phenytoin were equally efficacious at aborting
seizures, with no rescue medications needed.55 Another study suggested that
phenobarbital works more quickly to terminate generalized convulsive status
epilepticus than the combination of diazepam and phenytoin.56
Approximately 40% of patients with convulsive status epilepticus do not
respond to benzodiazepines and enter stage II (established status epilepticus).
Second-line AED therapy was the topic of two randomized studies. IV valproic
acid and continuous IV diazepam were found to be equally efficacious in one
study,57 while IV valproic acid and phenytoin were found to work similarly in
another study.58 In a meta-analysis of studies of benzodiazepine-resistant status
epilepticus, seizure cessation with valproate (75.7%, 95% confidence interval
63.7% to 84.8%) and phenobarbital (73.6%, 95% confidence interval 58.3% to
TABLE 5-3 Suggested Treatment Algorithm for Status Epilepticus Stages I and IIa,b
Initial Care
◆ Assess airway, breathing, circulation
◆ Monitor vital signs, including oxygenation
◆ Check finger stick blood glucose
◆ Draw metabolic profile, complete blood cell count, toxicology screen, antiepileptic drug
levels
If Seizures Continue, First-line Antiepileptic Drug
◆ Treat with one of the following
◇ IV lorazepam 0.1 mg/kg total dose, can repeat (maximum 4 mg/single dose)
◇ IV diazepam 0.15–0.2 mg/kg total dose, can repeat (maximum 10 mg/single dose)
◇ IM midazolam 10 mg
◆ Alternatives
◇ Rectal diazepam 0.2–0.5 mg/kg (maximum dose 20 mg)
◇ IV phenobarbital 15-20 mg/kg loading dose
◇ Intranasal or buccal midazolam
If Seizures Continue, Second-line Antiepileptic Drug
◆ Treat with one of the following
◇ IV phenytoin or fosphenytoin
◇ IV valproic acid
◇ IV levetiracetam
◇ Others (see TABLE 5-4)
IM = intramuscular; IV = intravenous.
a
Modified with permission from Glauser T, et al, Epilepsy Curr.2 © 2016 American Epilepsy Society.
b
See text for definition of status epilepticus stages.
1696 DECEMBER 2018

84.8%) was higher than with levetiracetam (68.5%, 95% confidence interval
56.2% to 78.7%) or phenytoin (50.2%, 95% confidence interval 34.2% to 66.1%).59
Based on these data, algorithms have been proposed for treating status
epilepticus. A typical algorithm is presented in TABLE 5-3. TABLE 5-4 details
medications (including dosing information and side effects) typically used as
first- and second-line agents to treat status epilepticus.60,61
Generally, the efficacy of each subsequent AED is less than those before it.
One randomized controlled trial that compared four different AED regimens
(phenytoin, lorazepam, phenobarbital, and diazepam followed by phenytoin)
showed that the efficacy of the first AED in aborting convulsive status epilepticus
was 55.5%, efficacy of the second AED 7.0%, and efficacy of the third AED 2.3%.2,49
Of note, the ESETT (Established Status Epilepticus Treatment Trial) study
is ongoing (although currently closed to enrollment for adults) to refine
Second-line Antiepileptic Drugsa TABLE 5-4
Typical Empiric
Medication Initial Dose Maintenance Dose Serious Adverse Effects Notes
Fosphenytoin 20 mg phenytoin 100 mg IV every Arrhythmia, hypotension Phenytoin and valproic

equivalents/kg IV, 8 hours acid interact by
maximum rate up to increasing their free
150 mg phenytoin levels60
equivalents/min
Lacosamide 200–400 mg IV 200 mg IV every PR prolongation, hypotension Minimal drug

12 hours interactions; has not
been used much in status
epilepticus
Levetiracetam 1000–3000 mg IV, up to 500–1500 mg IV Occasional behavioral issues61 Minimal drug interactions
a maximum dose of every 12 hours
4500 mg
Phenobarbital 20 mg/kg IV 50–100 mg IV every Respiratory depression, IV form contains

12 hours hypotension propylene glycol
Phenytoin 20 mg/kg IV 100 mg IV every Arrhythmia, hypotension, IV form contains

8 hours purple glove syndrome propylene glycol;
phenytoin and valproic
acid interact by
increasing their free
levels60
Topiramate 200–400 mg orally 300–1600 mg/d Metabolic acidosis Not available in IV form
orally (divided over
2–4 doses daily)
Valproic acid 20–40 mg/kg IV 500–750 mg IV every Thrombocytopenia, Phenytoin and valproic
8 hours gastrointestinal adverse acid interact by
effects (pancreatitis, increasing their free
hepatotoxicity), levels60
hyperammonemia
IV = intravenous.
a
Modified with permission from Brophy GM, et al, Neurocrit Care.1 © 2012 Springer Science+Business Media, LLC.

STATUS EPILEPTICUS
management of status epilepticus. ESETT is a National Institutes of Health

(NIH)–supported multicenter, randomized, blinded, comparative effectiveness
study of fosphenytoin, valproic acid, and levetiracetam for patients with
benzodiazepine-refractory status epilepticus.16,62
Management of Refractory Status Epilepticus and Super-refractory

Status Epilepticus
Management of refractory status epilepticus is primarily composed of seizure
control, treating the etiology of the seizures, and managing and preventing
complications.63 Interestingly, however, the extent of seizure suppression
needed to treat refractory status epilepticus is not entirely clear. In a
meta-analysis including 193 patients with refractory status epilepticus,
background EEG suppression was associated with fewer breakthrough seizures
as compared to merely suppressing seizures, although it had no effect on
mortality and a greater frequency of hypotension was seen.64 A retrospective
study of 47 patients with refractory status epilepticus showed that the level of
EEG suppression had no effect on outcome.22 However, another study of 63
refractory status epilepticus episodes showed that improved functional outcome
was associated with seizure suppression as compared to burst suppression or
isoelectric background.63
Continuous anesthetic agents are typically used to treat refractory status
epilepticus (TABLE 5-5), and continuous EEG is recommended while these
medications are being used. Although it is not known for how long anesthetic
treatment should be administered to control seizures, a continuous infusion
is typically maintained for 24 to 48 hours before it is weaned.1 In a study of
63 episodes of refractory status epilepticus in 54 patients, however,
anesthetic-induced coma occurred for an average of 11 days.27
The four major IV anesthetics used for refractory status epilepticus are
midazolam, propofol, pentobarbital (thiopental is often used outside the
TABLE 5-5 Continuous Infusions for Refractory Status Epilepticusa
Medication Initial Dose Maintenance Dose Serious Adverse Effects/Drawbacks
Midazolam 0.2 mg/kg 0.05–2 mg/kg/h Respiratory depression, hypotension, tachyphylaxis

after long use
Propofol 1–2 mg/kg 30–200 mcg/kg/min Respiratory depression, hypotension, propofol infusion
loading dose syndrome
Pentobarbital 5–15 mg/kg 0.5–5 mg/kg/h Cardiac and respiratory depression, hypotension, ileus,
loss of neurologic examination at high doses
Thiopental 2–7 mg/kg 0.5–5 mg/kg/h Cardiac and respiratory depression, hypotension
Ketamine 0.5–4.5 mg/kg Up to 5 mg/kg/h Hypertension, arrhythmia, anaphylaxis, pulmonary

edema
Isoflurane Not established End-tidal concentrations Cardiac and respiratory depression, infections
0.8–2% titrated to EEG
EEG = electroencephalography.
a
Data from Brophy GM, et al, Neurocrit Care1 and Hocker S, et al, Neurol Res.63
1698 DECEMBER 2018

United States), and ketamine. Midazolam, propofol, and barbiturates are GABA KEY POINTS
agonists, and propofol may also act as an NMDA antagonist; ketamine is an
● Treating status
NMDA antagonist. All four drugs are primarily metabolized in the liver.7 It is epilepticus generally begins
unclear if the anesthetics with a shorter half-life (midazolam or propofol) should with benzodiazepines
be used before those with a longer half-life (barbiturates). Some algorithms and is followed by IV
include pentobarbital and ketamine as treatment options for refractory status administration of
fosphenytoin, valproic
epilepticus and some algorithms include these drugs only for treatment of
acid, or levetiracetam.
super-refractory status epilepticus, but the ultimate goal is to stop status
epilepticus as soon as possible. In one meta-analysis, pentobarbital appeared to ● The extent of suppression
be associated with decreased breakthrough seizures, less short-term treatment needed to treat refractory
failure, and fewer changes to another medication infusion compared to midazolam status epilepticus (burst
suppression versus merely
and propofol.64 In refractory status epilepticus cases in which barbiturates were suppressing seizures) is
administered, EEG burst or total suppression was achieved more frequently than not known.
in refractory status epilepticus cases without administration of barbiturates.22
However, barbiturates may be linked to a longer hospital stay.22 In a small ● Continuous anesthetic
agents are used to treat
randomized trial of propofol versus pentobarbital, time spent on mechanical refractory status epilepticus
ventilation was longer in patients treated with pentobarbital, but return to baseline and are typically maintained
and mortality were similar.65 One problem with midazolam is that tachyphylaxis for 24 to 48 hours before
can develop, requiring progressively higher doses.63 Another potential problem is being weaned.
propofol infusion syndrome, which is characterized by circulatory collapse, lactic
● Several alternative
acidosis, rhabdomyolysis, and hypertriglyceridemia. It can be a life-threatening therapies can be used in
condition and is typically associated with high doses and long duration of patients with refractory
propofol use.6 status epilepticus and,
Ketamine has recently emerged as an alternative to traditional IV anesthetic especially, super-refractory
status epilepticus, including
agents. However, knowledge about ketamine and its potential usefulness is surgical resection,
limited since it is often added to other continuous infusions.7 A meta-analysis of repetitive transcranial
110 adult patients revealed that ketamine may have helped control refractory magnetic stimulation,
status epilepticus in about 57% of patients.66 A review of 95 patients treated with immunosuppression or
immunomodulation, and
ketamine for refractory status epilepticus or super-refractory status epilepticus the ketogenic diet.
showed that seizures resolved in 68%, but outcomes were variable: good
outcomes were observed in 19 patients (including discharges to home or
rehabilitation), death in 30, and other/unknown deficits in the remaining
patients.67 Although the side effects of ketamine are not well delineated,
concerns include psychiatric symptoms (eg, hallucinations, delirium, dreams) in
awake patients, increased intracranial pressure, increased intraocular pressure,
increased secretion of saliva, arrhythmias, respiratory depression,
and neurotoxicity.67
In addition to general anesthetics, lacosamide has also been tried in refractory
status epilepticus (this drug was not included in the aforementioned meta-
analysis of the drugs used for benzodiazepine-resistant established status
epilepticus).59 The most commonly used IV bolus dose of lacosamide was 200 mg
to 400 mg, followed by a daily dose of approximately 200 mg to 400 mg in a
review of 136 patients with refractory status epilepticus treated with the drug.
This regimen led to control of seizures in 56% of cases of refractory status
epilepticus. Adverse events occurred in 25% of cases; sedation was most
common, followed by hypotension; allergic skin reaction; and one case each of
possible angioedema, pruritus, and third-degree atrioventricular block and
paroxysmal asystole.68
Several other approaches have been attempted in refractory status epilepticus
and even more in super-refractory status epilepticus. Resective neurosurgery can

STATUS EPILEPTICUS
be considered if a definitive seizure focus causing status epilepticus can be

located in a noneloquent brain region.6 Surgeries can include focal, lobar, or
multilobar resection; corpus callosotomy; hemispherectomy; and multiple
subpial transections with or without focal resection. A review of 23 patients
undergoing surgery for refractory status epilepticus showed that 78.3% were
seizure free over a follow-up of 4 months to 5 years.69
Repetitive transcranial magnetic stimulation (rTMS), which provides
intracranial electric current in a noninvasive manner, has also been attempted.
A 2015 systematic review on the use of rTMS in status epilepticus and refractory
status epilepticus identified 11 studies (all case series or case reports) with
21 patients included. In the setting of rTMS, seizure control or reduction
occurred in 71.4%, although seizures recurred in 73.3% of patients who had
initially responded. An adverse event (lower extremity sensory symptoms that
resolved) was reported in only one patient.70
Electroconvulsive therapy has also been used. One systematic review found
14 articles (all retrospective case reports or series) consisting of 19 patients who
underwent electroconvulsive therapy for refractory status epilepticus. In 57.9%
of patients, seizure reduction or control was seen. Four studies discussed adverse
events: Two studies reported no adverse events; in the other two studies, three
patients were described as having transient amnesia or lethargy.71
Therapeutic hypothermia has also been posited as a possible treatment for
refractory status epilepticus and super-refractory status epilepticus. After several
case reports suggested a possible benefit for therapeutic hypothermia in
refractory status epilepticus,6 a 2016 study randomly assigned 270 patients in
the ICU who were on mechanical ventilation and in convulsive status epilepticus
to standard care alone or standard care plus hypothermia (32oC [89.6°F] to 34oC
[93.2°F] for 24 hours). The study did not show better outcomes in the patients
treated with hypothermia (primary outcome was defined as a Glasgow Outcome
Scale score of 5 at 90 days). Interestingly, progression to EEG-confirmed status
epilepticus (a secondary outcome) was more common in the standard care–only
group.72
Immunomodulatory agents, such as plasma exchange, IV immunoglobulin
(IVIg), steroids, and adrenocorticotropic hormone (ACTH), could be considered
in status epilepticus cases suspected to be caused by an immunologic process
after infection is excluded.6 Other adjunctive therapies include IV magnesium,
inhalational anesthetic agents, vagal nerve stimulation, and deep brain
stimulation.1,6,63
The ketogenic diet has recently shown promise as a treatment for
super-refractory status epilepticus. In a 2017 prospective multicenter phase
1/2 study of adult patients with super-refractory status epilepticus treated with
the ketogenic diet, all 15 patients enrolled in the study achieved ketosis after a
median of 2 days on the diet, and 78.6% of patients who completed treatment
with the diet achieved resolution of super-refractory status epilepticus at a
median of 5 days.73 Side effects of the ketogenic diet include metabolic acidosis,
hyperlipidemia, hypoglycemia, constipation, weight loss, and hyponatremia.73
These results suggest the need for future randomized trials to determine the
ketogenic diet’s safety and efficacy in super-refractory status epilepticus.
Allopregnanolone is a neuroactive steroid and positive allosteric modulator of
GABAA receptors that has shown success in reducing seizure activity in multiple
animal models.74 The STATUS (SAGE-547 Treatment as Adjunctive Therapy
1700 DECEMBER 2018

Utilized in Status Epilepticus) trial, a phase 3 randomized, double-blind, KEY POINT
placebo-controlled trial designed to evaluate the efficacy and safety of this
● In some cases, seizures
medication in super-refractory status epilepticus, however, was recently and status epilepticus may
completed and showed no difference between the allopregnanolone and be epiphenomena for severe
placebo arms.75 brain injury rather than the
As previously mentioned, many of the management principles used to treat primary offender. It is
unknown whether treating
refractory status epilepticus are also used for super-refractory status epilepticus.
these conditions will
TABLE 5-6 describes one possible treatment algorithm for refractory status improve outcomes.
epilepticus and super-refractory status epilepticus.
Several complications can occur in status epilepticus that could require
additional attention and management, including the possible need for
tracheostomy, hypotension, arrhythmias, acid-base disorders, renal failure,
thrombocytopenia, pulmonary embolus or deep vein thrombosis, infections such
as pneumonia or sepsis, drug rash, rhabdomyolysis, and critical illness
neuropathy or myopathy.63 Risk factors for these complications include not only
immobility from status epilepticus itself and potentially from IV anesthetic
agents causing coma but also any adverse effects or immunosuppression
resulting from the medications used to treat status epilepticus.63
Finally, it should be acknowledged that limited data exist regarding patients
who are critically ill to guide AED withdrawal after cessation of status epilepticus
or to predict which patients may not tolerate weaning. Although no randomized
controlled trials exist to guide the timing of AED weaning in adults who are
seizure free,76 in the first study evaluating early weaning of AEDs after seizure
cessation in patients who are critically ill, weaning AEDs after seizure cessation
did not lead to more recurrent seizures as compared to not weaning AEDs.77
Recently, independent predictors of drug-resistant epilepsy after convulsive
status epilepticus have been found to include a history of epilepsy, duration of
status epilepticus of 24 hours or more, and cortical or hippocampal abnormalities
on neuroimaging.78
CLINICAL OUTCOMES
Seizures and status epilepticus can lead to poor outcomes; however, it is unclear
whether detecting and treating seizures has an effect on outcomes since, in some
cases, seizures are epiphenomena for severe brain injury rather than the primary
offender.10 Mortality with status epilepticus may reach up to 30% in adults.2
Age older than 60 years, female sex, treatment in smaller-sized hospitals, the
presence of comorbidities (eg, hypertension, diabetes mellitus, previous stroke),
status epilepticus complications (eg, respiratory failure, sepsis), and etiologies
such as status epilepticus postcardiopulmonary resuscitation have been
associated with worse discharge outcomes.79
Outcomes in refractory status epilepticus can be worse, with mortality
reaching 16% to 39%.6 In one study of 54 patients and 63 episodes of refractory
status epilepticus, the mean length of hospital stay was 27.7 days. Poor outcome
at discharge (mRS score of 4 to 6) was noted in 76.2% of patients, and in-hospital
mortality occurred in 31.8%. Mechanical ventilation was required in 90.5% of
patients. Prolonged mechanical ventilation was associated with mortality. Poor
functional outcome was associated with greater CSF white blood cell count, more
days under anesthetic agents, the need for intervention for cardiac arrhythmias,
pneumonia, lack of seizure control on EEG (eg, requiring isoelectric or burst
suppression), and prolonged hospital stay. Seizure control without need for deep

STATUS EPILEPTICUS
TABLE 5-6 Suggested Treatment Algorithm for Refractory Status Epilepticus (Stage III)
and Super-refractory Status Epilepticus (Stage IV)a
Stage III Refractory Status Epilepticus

◆ Patients who are intubated and mechanically ventilated, on complete hemodynamic
support, and under continuous EEG recording
◆ Continue all antiepileptic drugs already started; use IV formulations if available
◆ Anesthetics for 24–48 hours:
◇ Midazolam 0.2 mg/kg IV bolus, which can be repeated every 5–10 minutes up to 2 mg/kg
total and start infusion 0.1–0.2 mg/kg/h OR
◇ Propofol 2 mg/kg IV bolus and 150 mcg/kg/min infusion OR
◇ Thiopental 4 mg/kg IV loading dose and 0.3–0.4 mg/kg/min infusion OR
◇ Pentobarbital 10 mg/kg IV loading dose, which can be repeated to burst suppression
effect (20–30 second suppression interval); start infusion at 1 mg/kg/h and titrate up to
10 mg/kg/h OR
◇ Ketamine 0.5–4.5 mg/kg IV bolus and start infusion up to 5 mg/kg/h
◇ Monitor and aggressively treat hypotension, sepsis, atelectasis or pneumonia, and deep
venous thrombosis; may need total parenteral nutrition
Stage IV.I Super-refractory Status Epilepticus
◆ If seizure control fails or seizures recur after tapering the doses, use the same as above for
longer period (eg, 1 week is suggested by the authors of this Continuum article) or proceed
directly to stage IV.II
Stage IV.II Super-refractory Status Epilepticus
◆ If seizures are still not controlled or recur, use one or more of the following alternative
therapies:
◇ Isoflurane or desflurane or gabapentin or levetiracetam (in acute intermittent porphyria)
◇ Topiramate 300–1600 mg/d per orogastric tube (if no increased stomach residuals)
◇ Magnesium 4 g bolus IV and 2–6 g/h infusion (keep serum levels <6 mEq/L)
◇ Pyridoxine 100–600 mg/d IV or via orogastric tube
◇ Methylprednisolone 1 g/d IV for 5 days, followed by prednisone 1 mg/kg/d orally for 1 week
◇ IVIg 0.4 g/kg/d IV for 5 days
◇ Plasma exchange for 5 sessions
◇ Ketogenic diet 4:1 (fat:carbohydrate and protein grams)
◇ Neurosurgical resection of epileptogenic focus
◇ Electroconvulsive therapy
◇ Vagal nerve stimulation or deep brain stimulation or repetitive transcranial magnetic
stimulation
Stage IV.III Super-refractory Status Epilepticus
◆ If several weaning attempts have failed over a period of weeks, consider an ethics
consultation or palliative care discussion with family or surrogate decision maker based on
patient’s wishes, with subsequent autopsy (if no etiology has been found)
EEG = electroencephalogram; IM = intramuscular; IV = intravenous; IVIg = intravenous immunoglobulin.

a
Modified with permission from Cuero MR, Varelas PN, Curr Neurol Neurosci Rep.23 © 2015 Springer
Science+Business Media, LLC.
1702 DECEMBER 2018

suppression on the EEG (isoelectric or burst suppression) was associated with KEY POINT
good functional recovery. Common complications in this cohort included cardiac
● Up to 50% of patients with
arrhythmias (35%), pulmonary edema (36.7%), acid-base disorders (71.4%), super-refractory status
hypotension (79%), hypoxia (35.7%), and pneumonia (69.6%).27 In another epilepticus die, and
study, fever was found to be the only independent predictor of outcome after outcomes are worse in
adjusting for acute symptomatic etiology, viral encephalitis etiology, fever, super-refractory status
epilepticus compared to
septicemia, and acidosis.28 In one study of 395 patients with refractory status
nonrefractory status
epilepticus treated in an ICU, in-hospital mortality was 7.4% and mortality at epilepticus.
1 year was 25.4%. On multivariate analysis, only the Sequential Organ Failure
Assessment (SOFA) score was independently associated with in-hospital
mortality. In terms of mortality at 1 year, independent predictors were older age,
premorbid nonindependence in activities of daily living, SOFA score, and
development of super-refractory status epilepticus.18
The differences between patients with refractory status epilepticus and
patients with nonrefractory status epilepticus have been investigated. Fever,
septicemia, and acidosis have been more frequently found in patients with
refractory status epilepticus as compared to patients with nonrefractory status
epilepticus.28 Patients with refractory status epilepticus have also been shown to
have a longer duration of seizure activity, mechanical ventilation, neurologic ICU
stay, and hospital stay.20,28 A 2017 meta-analysis found that the case fatality rate
for refractory status epilepticus was greater than the overall status epilepticus
case fatality rate (33% versus 15%).9 Another study also showed greater mortality
in patients with refractory status epilepticus compared to patients with
nonrefractory status epilepticus (42.2% versus 6.2%).28 Although in another
study mortality was similar in patients with nonrefractory status epilepticus and
those with refractory status epilepticus, increased neurologic ICU length of stay,
increased hospital length of stay, and decreased Glasgow Outcome Scale from
admission to discharge were found in patients with refractory status
epilepticus.21
Little has been reported in the literature regarding the outcome of
super-refractory status epilepticus. Long-term mortality of super-refractory
status epilepticus seems to be approximately 30% to 50%. In one study of
87 patients with super-refractory status epilepticus, mortality at 1 year was 36%,8
but mortality reached 50% in another study.24 In yet another status epilepticus
cohort, overall mortality was 19.2%, but mortality was up to 40% in the
super-refractory status epilepticus group, which was significantly greater than in
the nonrefractory status epilepticus (6.7%) group. At 6 months, a Glasgow
Outcome Scale score of 4 to 5 was achieved in 33.3% of patients with
super-refractory status epilepticus, which was significantly less than in patients
with nonrefractory status epilepticus (79.1%) but similar to patients with
refractory status epilepticus (57.1%).25 Compared to patients with refractory
status epilepticus, patients who met the definition of super-refractory status
epilepticus in another study had longer stays in both the neurologic ICU and in the
hospital and were more likely to be functionally dependent at hospital discharge.26
PROGNOSTIC SCORES AND THE FUTURE OF STATUS

EPILEPTICUS RESEARCH
Scores to predict outcome in status epilepticus have recently been developed.
The Status Epilepticus Severity Score (STESS) was first published in 2006. It is
based on four factors: age, seizure type, level of consciousness, and history of

STATUS EPILEPTICUS
seizures.80 A subsequent study found that STESS was a predictor of survival and
ability to achieve baseline clinical condition. This study also suggested that
patients who had favorable STESS scores generally appeared to survive
regardless of whether or not they received coma induction for their status
epilepticus.81 This score was further externally validated in a study of 171 patients
in which the STESS score identified survivors better than nonsurvivors.82 The
Epidemiology-Based Mortality Score in Status Epilepticus (EMSE) was created
using epidemiologic data in hopes of being superior to the STESS score. Points
were developed based on mortality rates in the literature for factors that were
thought to be predictive. Items eventually included in the score were etiology,
age, comorbidities, and EEG findings. This score explained individual mortality
in almost 90% of cases and was found to be superior to STESS.83
More recently, the END-IT score (the letters of which are an acronym for the
score’s components, encephalitis, nonconvulsive status epilepticus, diazepam
resistance, image abnormalities, and tracheal intubation) was created as an
outcome prediction tool in 132 patients with convulsive status epilepticus.
Independent predictors of unfavorable outcome (mRS score of 3 to 6) at
3 months after discharge were encephalitis, nonconvulsive status epilepticus
(defined as subtle status epilepticus in which myoclonic jerks or nystagmus
occurred in insufficiently treated convulsive status epilepticus), diazepam
resistance, imaging abnormalities (unilateral lesions, bilateral lesions, or diffuse
cerebral edema), and intubation. One point was given for each category except
for imaging (in which 1 point was given for unilateral lesions and 2 points for
diffuse cerebral edema or bilateral lesions). The greater the sum of these
categories, the greater the probability of unfavorable outcome. An END-IT score
of 3 or greater seemed to be the cutoff point to provide the best sensitivity and
specificity for predicting unfavorable outcome.84
CONCLUSION
Status epilepticus is a neurologic emergency and can progress to refractory status
epilepticus and super-refractory status epilepticus. Although much progress
has been made in diagnosing and treating status epilepticus, more studies are
needed to delineate optimal management and to improve outcomes.
REFERENCES
1 Brophy GM, Bell R, Claassen J, et al. Guidelines 4 Beniczky S, Hirsch LJ, Kaplan PW, et al. Unified
for the evaluation and management of status EEG terminology and criteria for nonconvulsive
epilepticus. Neurocrit Care 2012;17(1):3–23. status epilepticus. Epilepsia 2013;54(suppl 6):
doi:10.1007/s12028-012-9695-z. 28–29. doi:10.1111/epi.12270.
2 Glauser T, Shinnar S, Gloss D, et al. Evidence- 5 Berg AT, Berkovic SF, Brodie MJ, et al. Revised
based guideline: treatment of convulsive status terminology and concepts for organization of
epilepticus in children and adults: report of the seizures and epilepsies: report of the ILAE
guideline committee of the American Epilepsy Commission on Classification and Terminology,
Society. Epilepsy Curr 2016;16(1):48–61. 2005–2009. Epilepsia 2010;51(4):676–685.
doi:10.5698/1535-7597-16.1.48. doi:10.1111/j.1528-1167.2010.02522.x.
3 Trinka E, Cock H, Hesdorffer D, et al. A definition 6 Rossetti AO, Lowenstein DH. Management of
and classification of status epilepticus—report of refractory status epilepticus in adults: still more
the ILAE Task Force on classification of status questions than answers. Lancet Neurol 2011;
epilepticus. Epilepsia 2015;56(10):1515–1523. 10(10):922–930. doi:10.1016/S1474-4422(11)70187-9.
doi:10.1111/epi.13121.
1704 DECEMBER 2018

7 Reznik ME, Berger K, Claassen J. Comparison of 21 Mayer SA, Claassen J, Lokin J, et al. Refractory
intravenous anesthetic agents for the treatment status epilepticus: frequency, risk factors, and
of refractory status epilepticus. J Clin Med 2016; impact on outcome. Arch Neurol 2002;59(2):
5(5):pii:E54. doi:10.3390/jcm5050054. 205–210. doi:10.1001/archneur.59.2.205.
8 Kantanen AM, Reinikainen M, Parviainen I, et al. 22 Rossetti AO, Logroscino G, Bromfield EB.
Incidence and mortality of super-refractory Refractory status epilepticus: effect of
status epilepticus in adults. Epilepsy Behav 2015; treatment aggressiveness on prognosis. Arch
49:131–134. doi:10.1016/j.yebeh.2015.04.065. Neurol 2005;62(11):1698–1702. doi:10.1001/
archneur.62.11.1698.
9 Lv RJ, Wang Q, Cui T, et al. Status epilepticus-
related etiology, incidence and mortality: a 23 Cuero MR, Varelas PN. Super-refractory status
meta-analysis. Epilepsy Res 2017;136:12–17. epilepticus. Curr Neurol Neurosci Rep 2015;15(11):
doi:10.1016/j.eplepsyres.2017.07.006. 1–7. doi:10.1007/s11910-015-0594-5.
10 Betjemann JP, Lowenstein DH. Status epilepticus 24 Tian L, Li Y, Xue X, et al. Super-refractory status
in adults. Lancet Neurol 2015;14(6):615–624. epilepticus in West China. Acta Neurol Scand
doi:10.1016/S1474-4422(15)00042-3. 2015;132(1):1–6. doi:10.1111/ane.12336.
11 Dham BS, Hunter K, Rincon F. The epidemiology 25 Jayalakshmi S, Ruikar D, Vooturi S, et al.
of status epilepticus in the United States. Determinants and predictors of outcome in
Neurocrit Care 2014;20(3):476–483. doi:10.1007/ super refractory status epilepticus—a developing
s12028-013-9935-x. country perspective. Epilepsy Res 2014;108(9):
1609–1617. doi:10.1016/j.eplepsyres.2014.08.010.
12 DeLorenzo RJ, Hauser WA, Towne AR, et al. A
prospective, population-based epidemiologic 26 Holtkamp M, Othman J, Buchheim K, et al. A
study of status epilepticus in Richmond, Virginia. “malignant” variant of status epilepticus. Arch
Neurology 1996;46(4):1029–1035. doi:10.1212/ Neurol 2005;62(9):1428–1431. doi:10.1001/
WNL.46.4.1029. archneur.62.9.1428.
13 Coeytaux A, Jallon P, Galobardes B, Morabia A. 27 Hocker SE, Britton JW, Mandrekar JN, et al.
Incidence of status epilepticus in French- Predictors of outcome in refractory status
speaking Switzerland: (EPISTAR). Neurology epilepticus. JAMA Neurol 2013;70(1):72–77.
2000;55(5):693–697. doi:10.1212/WNL. doi:10.1001/jamaneurol.2013.578.
55.5.693.
28 Vooturi S, Jayalakshmi S, Sahu S, Mohandas S.
14 Knake S, Rosenow F, Vescovi M, et al. Incidence Prognosis and predictors of outcome of
of status epilepticus in adults in Germany: a refractory generalized convulsive status
prospective, population-based study. Epilepsia epilepticus in adults treated in neurointensive
2001;42(6):714–718. doi:10.1046/j.1528-1157. care unit. Clin Neurol Neurosurg 2014;126:7–10.
2001.01101.x. doi:10.1016/j.clineuro.2014.07.038.
15 Betjemann JP, Josephson SA, Lowenstein DH, 29 Hocker S, Tatum WO, LaRoche S, Freeman WD.
Burke JF. Trends in status epilepticus-related Refractory and super-refractory status
hospitalizations and mortality. JAMA Neurol epilepticus—an update. Curr Neurol Neurosci
2015;72(6):650–655. doi:10.1001/ Rep 2014;14(6):452. doi:10.1007/s11910-014-0452-x.
jamaneurol.2015.0188.
30 Leypoldt F, Buchert R, Kleiter I, et al.
16 Bleck T, Cock H, Chamberlain J, et al. The Fluorodeoxyglucose positron emission
established status epilepticus trial 2013. Epilepsia tomography in anti-N-methyl-D-aspartate
2013;54(suppl 6):89–92. doi:10.1111/epi.12288. receptor encephalitis: distinct pattern of
disease. J Neurol Neurosurg Psychiatry 2012;
17 Sutter R, Semmlack S, Kaplan PW. Nonconvulsive
83(7):681–686. doi:10.1136/jnnp-2011-301969.
status epilepticus in adults—insights into the
invisible. Nat Rev Neurol 2016;12(5):281–293. 31 Gaspard N, Foreman BP, Alvarez V, et al. New-onset
doi:10.1038/nrneurol.2016.45. refractory status epilepticus: etiology, clinical
features, and outcome. Neurology 2015;85(18):
18 Kantanen AM, Kälviäinen R, Parviainen I, et al.
1604–1613. doi:10.1212/WNL.0000000000001940.
Predictors of hospital and one-year mortality in
intensive care patients with refractory status 32 Meldrum BS, Horton RW. Physiology of status
epilepticus: a population-based study. Crit Care epilepticus in primates. Arch Neurol 1973;28(1):
2017;21(1):71. doi:10.1186/s13054-017-1661-x. 1–9. doi:10.1001/archneur.1973.00490190019001.
19 Novy J, Logroscino G, Rossetti AO. Refractory 33 Meldrum BS, Brierley JB. Prolonged epileptic
status epilepticus: a prospective observational seizures in primates. Ischemic cell change and
study. Epilepsia 2010;51(2):251–256. doi:10.1111/ its relation to ictal physiological events. Arch
j.1528-1167.2009.02323.x. Neurol 1973;28(1):10–17. doi:10.1001/archneur.1973.
00490190028002.
20 Holtkamp M, Othman J, Buchheim K, Meierkord
H. Predictors and prognosis of refractory status 34 Meldrum BS, Vigouroux RA, Brierley JB. Systemic
epilepticus treated in a neurological intensive factors and epileptic brain damage. Prolonged
care unit. J Neurol Neurosurg Psychiatry 2005; seizures in paralyzed, artificially ventilated
76(4):534–539. doi:10.1136/jnnp.2004.041947. baboons. Arch Neurol 1973;29(2):82–87.
doi:10.1001/archneur.1973.00490260026003.

STATUS EPILEPTICUS
35 Sarikaya I. PET studies in epilepsy. Am J Nucl Med 49 Treiman DM, Meyers PD, Walton NY, et al. A
Mol Imaging 2015;5(5):416–430. comparison of four treatments for generalized
convulsive status epilepticus. Veterans Affairs
36 Claassen J, Mayer SA, Kowalski RG, et al.
Status Epilepticus Cooperative Study Group.
Detection of electrographic seizures with
N Engl J Med 1998;339(12):792–798. doi:10.1056/
continuous EEG monitoring in critically ill
NEJM199809173391202.
patients. Neurology 2004;62(10):1743–1748.
doi:10.1212/01.WNL.0000125184.88797.62. 50 Alldredge BK, Gelb AM, Isaacs SM, et al. A
comparison of lorazepam, diazepam, and
37 Dunne JW, Summers QA, Stewart-Wynne EG.
placebo for the treatment of out-of-hospital
Non-convulsive status epilepticus: a prospective
status epilepticus. N Engl J Med 2001;345(9):
study in an adult general hospital. Q J Med 1987;
631–637. doi:10.1056/NEJMoa002141.
62(238):117–126. doi:10.1093/oxfordjournals.
qjmed.a068084. 51 Silbergleit R, Durkalski V, Lowenstein D, et al.
Intramuscular versus intravenous therapy for
38 Rudin D, Grize L, Schindler C, et al. High
prehospital status epilepticus. N Engl J Med 2012;
prevalence of nonconvulsive and subtle status
366(7):591–600. doi:10.1056/NEJMoa1107494.
epilepticus in an ICU of a tertiary care center: a
three-year observational cohort study. Epilepsy 52 Leppik IE, Derivan AT, Homan RW, et al. Double-
Res 2011;96(1–2):140–150. doi:10.1016/j.eplepsyres. blind study of lorazepam and diazepam in status
2011.05.018. epilepticus. JAMA 1983;249(11):1452–1454.
39 Towne AR, Waterhouse EJ, Boggs JG, et al. 53 Misra UK, Kalita J, Maurya PK. Levetiracetam versus
Prevalence of nonconvulsive status epilepticus lorazepam in status epilepticus: a randomized,
in comatose patients. Neurology 2000;54(2): open labeled pilot study. J Neurol 2012;259(4):
340–345. doi:10.1212/WNL.54.2.340. 645–648. doi:10.1007/s00415-011-6227-2.
40 Laccheo I, Sonmezturk H, Bhatt AB, et al. Non- 54 Misra UK, Kalita J, Patel R. Sodium valproate vs
convulsive status epilepticus and non-convulsive phenytoin in status epilepticus: a pilot study.
seizures in neurological ICU patients. Neurocrit Neurology 2006;67(2):340–342. doi:10.1212/01.
Care 2015;22(2):202–211. doi:10.1007/s12028- wnl.0000224880.35053.26.
014-0070-0.
55 Gilad R, Izkovitz N, Dabby R, et al. Treatment
41 Betjemann JP, Nguyen I, Santos-Sanchez C, et al. of status epilepticus and acute repetitive
Diagnostic yield of electroencephalography in a seizures with i.v. valproic acid vs phenytoin.
general inpatient population. Mayo Clin Proc Acta Neurol Scand 2008;118(5):296–300.
2013;88(4):326–331. doi:10.1016/j.mayocp. doi:10.1111/j.1600-0404.2008.01097.x.
2012.12.013.
56 Shaner DM, McCurdy SA, Herring MO, Gabor AJ.
42 Bauer J, Stefan H, Huk WJ, et al. CT, MRI and Treatment of status epilepticus: a prospective
SPECT neuroimaging in status epilepticus with comparison of diazepam and phenytoin versus
simple partial and complex partial seizures: phenobarbital and optional phenytoin.
case report. J Neurol 1989;236(5):296–299. Neurology 1988;38(2):202–207. doi:10.1212/
doi:10.1007/BF00314460. WNL.38.2.202.
43 Tatum WO, Alavi A, Stecker MM. Technetium- 57 Chen WB, Gao R, Su YY, et al. Valproate versus
99m-HMPAO SPECT in partial status epilepticus. diazepam for generalized convulsive status
J Nucl Med 1994;35(7):1087–1094. epilepticus: a pilot study. Eur J Neurol 2011;18(12):
1391–1396. doi:10.1111/j.1468-1331.2011.03420.x.
44 Kutluay E, Beattie J, Passaro EA, et al. Diagnostic
and localizing value of ictal SPECT in patients 58 Agarwal P, Kumar N, Chandra R, et al.
with nonconvulsive status epilepticus. Epilepsy Randomized study of intravenous valproate and
Behav 2005;6(2):212–217. doi:10.1016/j.yebeh. phenytoin in status epilepticus. Seizure 2007;
2004.12.001. 16(6):527–532. doi:10.1016/j.seizure.2007.04.012.
45 Siclari F, Prior JO, Rossetti AO. Ictal cerebral 59 Yasiry Z, Shorvon SD. The relative effectiveness
positron emission tomography (PET) in focal of five antiepileptic drugs in treatment of
status epilepticus. Epilepsy Res 2013;105(3): benzodiazepine-resistant convulsive status
356–361. doi:10.1016/j.eplepsyres.2013.03.006. epilepticus: a meta-analysis of published
studies. Seizure 2014;23(3):167–174. doi:10.1016/
46 Trinka E, Kälviäinen R. 25 Years of advances in the
j.seizure.2013.12.007.
definition, classification and treatment of status
epilepticus. Seizure 2017;44:65–73. doi:10.1016/ 60 Perucca E, Hebdige S, Frigo GM, et al. Interaction
j.seizure.2016.11.001. between phenytoin and valproic acid: plasma
protein binding and metabolic effects. Clin
47 Claassen J, Perotte A, Albers D, et al. Nonconvulsive
Pharmacol Ther 1980;28(6):779–789. doi:10.1038/
seizures after subarachnoid hemorrhage:
clpt.1980.235.
Multimodal detection and outcomes. Ann Neurol
2013;74(1):53–64. doi:10.1002/ana.23859. 61 Cramer JA, De Rue K, Devinsky O, et al. A
systematic review of the behavioral effects of
48 Vespa P, Tubi M, Claassen J, et al. Metabolic
levetiracetam in adults with epilepsy, cognitive
crisis occurs with seizures and periodic
disorders, or an anxiety disorder during clinical
discharges after brain trauma. Ann Neurol 2016;
trials. Epilepsy Behav 2003;4(2):124–132.
79(4):579–590. doi:10.1002/ana.24606.
doi:10.1016/S1525-5050(03)00005-2.
1706 DECEMBER 2018

62 Clinicaltrials.gov. Established Status Epilepticus 74 Rogawski M, Loya C, Reddy K, et al. Neuroactive
Treatment Trial (ESETT): NCT01960075. steroids for the treatment of status epilepticus.
clinicaltrials.gov/ct2/show/NCT01960075. Epilepsia 2013;54(suppl 6):93–98. doi:10.1111/
Updated January 5, 2018. Accessed October 5, epi.12289.
2018.
75 Sage Therapeutics Reports Top-Line Results
63 Hocker S, Wijdicks EF, Rabinstein AA. Refractory from Phase 3 STATUS Trial of Bexanolone in
status epilepticus: new insights in presentation, Super-Refractory Status Epilepticus. investor.
treatment, and outcome. Neurol Res 2013;35(2): sagerx.com/news-releases/news-release-
163–168. doi:10.1179/1743132812Y.0000000128. details/sage-therapeutics-reports-top-line-
results-phase-3-status-trial. Updated
64 Claassen J, Hirsch LJ, Emerson RG, Mayer SA.
September 12, 2017. Accessed October 5, 2018.
Treatment of refractory status epilepticus with
pentobarbital, propofol, or midazolam: a 76 Strozzi I, Nolan SJ, Sperling MR, et al. Early versus
systematic review. Epilepsia 2002;43(2):146–153. late antiepileptic drug withdrawal for people
doi:10.1046/j.1528-1157.2002.28501.x. with epilepsy in remission. Cochrane Database
Syst Rev 2015;(2):CD001902. doi:10.1002/
65 Rossetti AO, Milligan TA, Vulliémoz S, et al. A
14651858.CD001902.pub2.
randomized trial for the treatment of refractory
status epilepticus. Neurocrit Care 2011;14(1):4–10. 77 Creed JA, Son J, Farjat AE, Swisher CB. Early
doi:10.1007/s12028-010-9445-z. withdrawal of non-anesthetic antiepileptic drugs
after successful termination of nonconvulsive
66 Zeiler FA, Teitelbaum J, Gillman LM, West M.
seizures and nonconvulsive status epilepticus.
NMDA antagonists for refractory seizures.
Seizure 2018;54:45–50. doi:10.1016/j.seizure.
2017.12.001.
s12028-013-9939-6.
78 Yuan F, Jia R, Gao Q, et al. Early predictors of
67 Fang Y, Wang X. Ketamine for the treatment of
drug-resistant epilepsy development after
refractory status epilepticus. Seizure 2015;30:
convulsive status epilepticus. Eur Neurol 2018;
14–20. doi:10.1016/j.seizure.2015.05.010.
79(5-6):325–332. doi:10.1159/000490900.
68 Höfler J, Trinka E. Lacosamide as a new treatment
79 Tiamkao S, Pranboon S, Thepsuthammarat K,
option in status epilepticus. Epilepsia 2013;54(3):
Sawanyawisuth K. Incidences and outcomes of
393–404. doi:10.1111/epi.12058.
status epilepticus: a 9-year longitudinal national
69 Lhatoo SD, Alexopoulos AV. The surgical study. Epilepsy Behav 2015;49:135–137.
treatment of status epilepticus. Epilepsia 2007; doi:10.1016/j.yebeh.2015.04.040.
48(suppl 8):61–65. doi:10.1111/j.1528-
80 Rossetti AO, Logroscino G, Bromfield EB. A
1167.2007.01353.x.
clinical score for prognosis of status epilepticus
70 Zeiler FA, Matuszczak M, Teitelbaum J, et al. in adults. Neurology 2006;66(11):1736–1738.
Transcranial magnetic stimulation for status doi:10.1212/01.wnl.0000223352.71621.97.
epilepticus. Epilepsy Res Treat 2015;2015:678074.
81 Rossetti AO, Logroscino G, Milligan TA, et al.
doi:10.1155/2015/678074.
Status Epilepticus Severity Score (STESS): a tool
71 Zeiler FA, Matuszczak M, Teitelbaum J, et al. to orient early treatment strategy. J Neurol 2008;
Electroconvulsive therapy for refractory status 255(10):1561–1566. doi:10.1007/s00415-008-0989-1.
epilepticus: a systematic review. Seizure 2016;35:
82 Sutter R, Kaplan PW, Rüegg S. Independent
23–32. doi:10.1016/j.seizure.2015.12.015.
external validation of the status epilepticus
72 Legriel S, Lemiale V, Schenck M, et al. severity score. Crit Care Med 2013;41(12):
Hypothermia for neuroprotection in convulsive e475–e479. doi:1097/CCM.0b013e31829eca06.
status epilepticus. N Engl J Med 2016;375(25):
83 Leitinger M, Höller Y, Kalss G, et al.
2457–2467. doi:10.1056/NEJMoa1608193.
Epidemiology-based mortality score in status
73 Cervenka MC, Hocker S, Koenig M, et al. Phase epilepticus (EMSE). Neurocrit Care 2015;22(2):
I/II multicenter ketogenic diet study for adult 273–282. doi:10.1007/s12028-014-0080-y.
superrefractory status epilepticus. Neurology
84 Gao Q, Ou-Yang TP, Sun XL, et al. Prediction of
2017;88(10):938–943. doi:10.1212/
functional outcome in patients with convulsive
WNL.0000000000003690.
status epilepticus: the END-IT score. Crit Care
2016;20:46. doi:10.1186/s13054-016-1221-9.
DISCLOSURE
Continued from page 1683
Dr Varelas has received personal compensation for Therapeutics, Inc; Marinus Pharmaceuticals, Inc;
speaking engagements for Portola Pharmaceuticals, the National Institutes of Health; the Patient-
Inc and UCB SA and receives publishing royalties Centered Outcomes Research Institute; and Portola
from Springer. Dr Varelas receives research/grant Pharmaceuticals, Inc
support from Bard Pharmaceuticals Limited; Edge

REVIEW ARTICLE

Coma and Brain Death
C O N T I N UU M A UD I O By Alejandro A. Rabinstein, MD, FAAN
ONLINE

S U P P L E M E N T AL D I G I T A L ABSTRACT
CONTENT (SDC) PURPOSE OF REVIEW: This article discusses the diagnostic and therapeutic
A VA I L A B L E O N L I N E
approach to patients who are comatose and reviews the current
knowledge on prognosis from various causes of coma. This article also
provides an overview of the principles for determination of brain death as
well as advice on how to avoid common pitfalls.
RECENT FINDINGS: Technologic advances have refined our understanding

of the physiology of consciousness and the spectrum of disorders of
consciousness; they also promise to improve our prognostic accuracy. Yet
the clinical principles for the evaluation and treatment of coma remain
unaltered. The clinical standards for determination of death by neurologic
criteria (ie, brain death) are also well established, although variabilities in
local protocols and legal requirements remain a problem to be resolved.
SUMMARY: Effective evaluation of coma demands a systematic approach

relying on clinical information to ensure rational use of laboratory and
imaging tests. When the cause of coma is deemed irreversible in the setting
of a catastrophic brain injury and no clinical evidence exists for brain and
brainstem function, patients should be evaluated for the possibility of
brain death by following the clinical criteria specified in the American
Academy of Neurology guidelines.
CITE AS:
CONTINUUM (MINNEAP MINN) INTRODUCTION
C
2018;24(6, NEUROCRITICAL CARE): onsciousness is the cerebral function that defines our identity,
1708–1731.
yet the definition of this primordial function is elusive, and the
Address correspondence to Dr
complex mechanisms underlying its existence and preservation
Alejandro A. Rabinstein, 200 1st are just beginning to be unraveled. More is known about loss of
St SW, Rochester, MN 55905, consciousness, which results in the state known as coma. However,
rabinstein.alejandro@mayo.edu.
clinical evaluation of consciousness is far from straightforward. For example,
RELATIONSHIP DISCLOSURE: growing understanding of the spectrum of chronic disorders of consciousness
Dr Rabinstein receives royalties demonstrates that minimally preserved consciousness may not be apparent even
from Elsevier and Oxford
University Press and has received to the examiner at the bedside.1
research support from DJO Catastrophic brain injury can result in irreversible coma with cessation of all
Global, Inc.
brain (and brainstem) function. When this occurs, patients can be legally
UNLABELED USE OF declared dead by neurologic criteria (ie, brain death) and become eligible as
PRODUCTS/INVESTIGATIONAL organ donors. Needless to say, determination of brain death is an enormous
USE DISCLOSURE:
Dr Rabinstein reports no
responsibility. It demands a conscientious and detailed evaluation, which has
disclosure. been carefully described in guidelines published by the American Academy of
Neurology (AAN).2
© 2018 American Academy Coma and brain death are two core topics in any neurologic training
of Neurology. curriculum, and having in-depth knowledge of these clinical scenarios is essential
1708 DECEMBER 2018

for every neurologist practicing in a hospital setting. Nonetheless, KEY POINTS
misinformation and misunderstandings are not uncommon in evaluation and
● Coma is a state of
prognosis of coma and determination of brain death. This article provides an unresponsiveness in which
updated, practical, and systematic approach to clarify key concepts and the patient is not awake and
highlights potential pitfalls that may result in serious medical errors. cannot interact with the
environment, even after
vigorous stimulation.
COMA
Coma is commonly defined as a state of unresponsiveness in which the patient is ● Bilateral diffuse
not awake and cannot interact with the environment, even after vigorous alterations in cortical
function or severe
alerting stimulation. It lies at one end of a continuum of depression in the level of
diencephalic or brainstem
consciousness, a spectrum that also comprises less severe states of decreased dysfunction can produce
responsiveness such as drowsiness (awakening to verbal stimulation) and stupor coma.
(interaction with the environment elicited by alerting noxious stimulation).
These terms can be helpful for communication, but specifically describing the
patient’s responses to various forms of stimulation is often clearer.
It is important to remember that consciousness is best conceptualized as having
two domains: level and content. Level of consciousness refers to the degree
of arousal (ie, is the patient fully awake or does he or she require stimulation to
awaken and respond?), while content of consciousness refers to the degree
of awareness (ie, is the patient coherent or is he or she confused, inattentive, or
delusional?). Both domains should be evaluated, except when the severity of
depression in the level of consciousness precludes evaluation of its content.
Pathophysiologic Mechanisms
The anatomic and physiologic underpinnings of human consciousness are
incompletely understood, yet we know that bilateral diffuse alterations in
cortical function or severe alterations in diencephalic or brainstem function can
result in coma.
It is common to distinguish between coma caused by structural brain damage
and coma caused by diffuse cerebral dysfunction.3 This distinction, while
pragmatically useful, is sometimes imprecise. Patients with a brain tumor can
become comatose from mass effect and tissue shift or from seizures. After
prolonged refractory status epilepticus, patients may remain comatose because
of cortical injury even after resolution of the seizures that were responsible for
the initial loss of responsiveness. Nonetheless, it is always useful to try to localize
the cause of coma when possible.
A basic anatomic classification of coma is presented in TABLE 6-1.
Anatomic Classification of Coma TABLE 6-1
◆ Hemispheric brain lesion with tissue shift and herniation

◆ Diffuse bilateral hemispheric damage
◆ Bilateral diencephalic lesions
◆ Cerebellar lesion causing brainstem compression
◆ Intrinsic brainstem lesion
◆ Hydrocephalus

COMA AND BRAIN DEATH
Supratentorial lesions typically cause coma by mass effect, tissue shift, and
herniation. Lateral tissue displacement causes subfalcine herniation, while
vertical tissue shift produces transtentorial (uncal) and eventually transforaminal
herniation. However, the gradient induced by the mass effect is not unidirectional,
and, in most cases, lateral and vertical displacements are combined.
Another way to categorize coma is by the type of condition responsible for its
occurrence (TABLE 6-2).3,4 Coma can be the end result of many forms of severe
brain dysfunction, including cerebrovascular events, trauma, anoxia, infections,
noninfectious inflammation, hydrocephalus, seizures, intoxications, metabolic
disturbances, sepsis, and extreme alterations in body temperature. Notably, several
causes can be combined. Common combinations include anoxia and seizures;
trauma and anoxia, sepsis, drug effects, or metabolic disturbances; and intracranial
hemorrhage and hydrocephalus. In these cases, the relative contribution of each
TABLE 6-2 Common Causes of Coma
Global Anoxia/Ischemia
Cerebrovascular Disease
◆ Massive hemispheric infarction or hematoma
◆ Brainstem ischemic stroke from basilar artery occlusion
◆ Brainstem hemorrhage
◆ Cerebellar infarction or hematoma with mass effect
◆ Poor-grade aneurysmal subarachnoid hemorrhage
◆ Dural venous sinus thrombosis
Trauma
◆ Diffuse axonal injury
◆ Brain edema and intracranial hypertension
◆ Epidural or subdural hematoma
◆ Hemorrhagic brain contusions with mass effect
◆ Fat embolism
Infection
◆ Acute bacterial meningitis
◆ Acute viral encephalitis
◆ Fungal or mycobacterial meningoencephalitis
◆ Brain abscesses
◆ Empyema
Inflammatory
◆ Autoimmune encephalitis
◆ Posterior reversible encephalopathy syndrome (PRES)
◆ Vasculitis
◆ Acute disseminated demyelination
1710 DECEMBER 2018

cause may not be initially apparent, but recognition of every possible cause and
treatment of those that are treatable are crucial to optimize the chances of recovery.
Evaluation
Although many practitioners rush to order tests when confronted with a
patient who is comatose, the main clues to the etiology of the coma can often be
obtained by a focused history and a skillful physical examination.4 Furthermore,
because the causes of coma are so many and so varied, a shotgun approach to
testing can generate more confusion by offering abnormal results of unclear
relevance. Thus, testing, with the exception of capillary glucose, should be
guided by the information provided by the history and the examination.
History taking should focus on obtaining the details of the presentation,
pertinent past medical history, and recent exposures. A crucial question is
Tumor
◆ Large hemispheric tumors with edema and mass effect
◆ Brainstem tumors
◆ Cerebellar tumors with mass effect
◆ Infiltrative tumors
◆ Pituitary apoplexy
Acute Hydrocephalus
Seizures
Toxins
◆ Prescription drug overdose
◆ Recreational drug overdose
◆ Drug interactions
◆ Poisoning
Metabolic and Endocrine Abnormalities
◆ Hyponatremia
◆ Hypercalcemia
◆ Acidosis
◆ Renal failure and uremia
◆ Hepatic failure and hyperammonemia
◆ Myxedema
◆ Adrenal insufficiency
Hypothermia

whether the onset was witnessed. Acuteness of onset, preceding symptoms,

abnormal movements, urinary incontinence, and signs that breathing may have
been impaired are also essential pieces of information. Diabetes mellitus,
obstructive pulmonary disease, epilepsy, previous brain injury, brain tumor,
recent neurosurgery, immunosuppression, and conditions requiring
anticoagulation are among the most relevant comorbid conditions that should be
specifically considered. Clues to the possibility of intoxication with prescription
or recreational drugs should be explored.4
The physical examination should begin by ensuring the adequacy of the
airway, breathing, and circulation. Fever can denote underlying sepsis or central
nervous system (CNS) infection. Hypothermia, when severe, can be solely
responsible for the coma. Dyspnea and cyanosis may indicate that the altered
consciousness can be reversed by correction of hypercapnia. Shock is a common
cause of altered mentation and even coma and demands its own evaluation. The
TABLE 6-3 Glasgow Coma Scalea
Response Scoreb
Eye response (E)
No eye opening 1
Eye opening to pain 2
Eye opening to verbal command 3
Spontaneous eye opening 4
Verbal response (V)c
No verbal response 1
Incomprehensible sounds 2
Inappropriate words 3
Confused 4
Oriented 5
Motor response (M)
No motor response 1
Stereotyped extension to pain 2
Stereotyped flexion to pain 3
Withdrawal from pain 4
Localizes pain 5
Follows commands 6
a
Data from Teasdale G, Jennett B, Lancet.6
b
Maximal sum score is 15 points. Coma is usually defined as a score of 8 points or less.
c
By convention, when the patient is intubated, the verbal response score is reported as 1 and T
(endotracheal tube or intubated) and is added to qualify the sum score. Alternatively, the verbal response
score can be estimated from the other two scores, as is shown in TABLE 6-4.7,8
1712 DECEMBER 2018

general examination should include inspection of the skin; for example, a KEY POINTS
petechial rash can alert the clinician to the possibility of fulminant meningitis.
● Focused history and
The neurologic examination of the patient who is comatose, although physical examination very
deceivingly simple and rarely diagnostic by itself, can offer extremely useful often provide clues to the
information.4 Attention should be paid to the degree of unresponsiveness, etiology of coma.
brainstem reflexes, position of the eyes, muscle tone, focal or lateralizing signs,
● Coma scales are useful
adventitious movements of the eyes and limbs, and meningeal signs.
but should not replace a
Differentiating the very common abnormal movements associated with toxic more complete neurologic
and metabolic causes from motor manifestations of seizures can be challenging, examination.
especially for less seasoned clinicians. While the former are typically arrhythmic
myoclonic jerks and asterixis and the latter are characteristically rhythmic and ● The differential diagnosis
and diagnostic testing
usually respect a more consistent pattern, EEG is indicated when uncertainty should be guided by the
persists after careful observation. history, physical
Using a validated scale to categorize the degree of coma is useful for examination, and location
triaging, subsequent monitoring through serial examinations, and prognosis. of the patient (out of
hospital, in the emergency
The Glasgow Coma Scale (GCS) was originally designed for acute evaluation of department, on a hospital
alterations of consciousness from traumatic brain injury in the field5 but was floor, or in the intensive
later applied to all other causes of coma (TABLE 6-3).6 The GCS is widely used, care unit).
but it has noticeable shortcomings: (1) it loses one of its three components (ie, the
● CT is a reasonable and
verbal component) when the patient is intubated; (2) it does not incorporate
practical first imaging study
an evaluation of the brainstem reflexes or the breathing pattern, which are when structural damage is
crucial elements in the assessment of any patient who is comatose; and (3) it is suspected, but MRI may be
skewed toward the motor response, particularly in patients who are intubated, more useful for various
for whom it accounts for 6 of 10 possible points, although it may be possible to causes of coma.
estimate the verbal component based on motor and eye scores to overcome this
limitation in patients who are comatose and intubated (TABLE 6-3 and TABLE 6-4).7,8
The Full Outline of UnResponsiveness (FOUR) score, developed by Wijdicks
and colleagues,9 addresses these deficiencies of the GCS and has been shown to
be comparable or superior to the GCS for prognostication across various types of
severe brain injury (FIGURE 6-1 and TABLE 6-5).10–12
Having a checklist in mind when assessing a patient who is comatose is
recommended (TABLE 6-6). Essentially, all patients who are comatose will require
some blood tests to exclude major primary or secondary metabolic abnormalities,
but beyond that, the laboratory investigations should be individualized.
A toxicologic screen is reasonable when the cause of coma is uncertain. Blood
urea nitrogen and serum ammonia should be measured in most cases.
Brain imaging is often, but not always, necessary. It becomes indispensable
after head trauma, when the physical examination shows focal deficits or
asymmetric motor responses, and when meningitis is suspected to ensure that
lumbar puncture can be performed safely. A head CT is often obtained first and
has good sensitivity to detect acute intracranial hemorrhage, hydrocephalus, and
brain tissue shift from a mass (FIGURE 6-2). Contrast administration is useful
when infection or tumor is possible.
Brain MRI is more sensitive for the recognition of most other causes of
coma and also helps clarify causes when the CT is abnormal but the diagnosis
remains indeterminate. Examples of coma etiologies that can be reliably
diagnosed with MRI include herpes simplex virus type 1 encephalitis, brainstem
infarction, and posterior reversible encephalopathy syndrome (PRES), the
latter in most but not all cases (FIGURE 6-3). Brain MRI can also show cortical and
basal ganglia damage after global brain anoxia-ischemia, diffuse axonal injury

after trauma, small infarctions from an embolic shower (including cerebral fat
embolism), venous infarctions in cases of dural or deep venous sinus thrombosis,
cortical injury after prolonged status epilepticus, pontine and supratentorial
demyelination after rapid osmotic changes, and signs of infiltrative brain tumor
that might not be appreciable on head CT. However, not every case of coma
from a structural brain injury will have an abnormal MRI. For example, some
forms of viral encephalitis (such as those caused by arboviruses), a minority of
cases with hypertensive encephalopathy, and some cases of persistent coma after
cardiopulmonary resuscitation may have normal appearance on brain MRI.
Noninvasive vessel imaging should be guided by clinical suspicion of a
particular diagnosis, most notably acute basilar artery occlusion or dural venous
sinus thrombosis. Ordering CT angiography (CTA) or magnetic resonance
angiography (MRA) and magnetic resonance venography (MRV) is not necessary
in most cases of coma, and the author has found these studies to be grossly
overused for this indication. Catheter angiography is very rarely indicated.
A lumbar puncture can be diagnostic in patients with CNS infection and
provides helpful clues in some other instances, such as in autoimmune
encephalitis. Low glucose levels can be seen after severe hypoglycemia (it takes
longer for the glucose level to normalize in the CSF than in blood) and with
fungal infections. High protein is quite common and nonspecific; it can be seen
with inflammation, infection, venous thrombosis, and seizures, among others.
The EEG is always abnormal in patients who are comatose but is far less
commonly useful for reaching a specific diagnosis or assisting in management. It
is definitely indicated in patients who fail to awaken after a prolonged clinical
seizure or multiple clinical seizures and when ongoing subtle clinical
manifestations of seizures are suspected. In fact, when seizures are strongly
suspected, it is advisable to extend the period of monitoring for at least a few
hours, if possible, before concluding that the coma is not being perpetuated by
TABLE 6-4 Glasgow Coma Scale Predictive Verbal Scorea,b
Glasgow Coma Scale Eye Score
Glasgow Coma
Scale Motor Score 1 2 3 4
1 1 1 1 2
2 1 1 1 1
3 1 1 1 2
4 2 1 2 2
5 3 2 3 3
6 4 4 4 5
a
Reprinted with permission from Rutledge R, et al, J Trauma.7 © 1996 Williams & Wilkins.
b
This table can be used in patients who are intubated to predict what the verbal score would be if the
patient were not intubated. It relies on the combination of the eye and motor scores of the Glasgow Coma
Scale to estimate the verbal score. For instance, a patient who is intubated with a motor score of 4 and an
eye score of 3 would be predicted to have a verbal score of 2.
1714 DECEMBER 2018

KEY POINT
● The value of EEG for the

evaluation of patients who
are comatose without
clinical seizures remains to
be established.
FIGURE 6-1
Full Outline of UnResponsiveness (FOUR) score. Additional explanations on scoring
are provided in TABLE 6-5.
E = eye response; M = motor response; B = brainstem reflexes; R = respiration.
Used with the permission of Mayo Foundation for Medical Education and Research © 2005.

intermittent seizures. Conversely, the value of EEG in patients who are comatose
in general is much less clear. Some advocate EEG monitoring for all patients in
coma, and some observational studies have reported a relatively high yield of
epileptiform abnormalities by following this approach.13 However, we are just
learning about the significance of these abnormalities in the ictal-interictal
continuum, and improvement in clinical outcomes from treating epileptiform
abnormalities with antiseizure medications in patients who are comatose has not
been proven. These abnormalities may just be manifestations of the brain
TABLE 6-5 The Full Outline of UnResponsiveness (FOUR) Scorea
Score Response
Eye response
4 Eyelids open or opened, tracking, or blinking to command
3 Eyelids open but not tracking
2 Eyelids closed but open to loud voice
1 Eyelids closed but open to pain
0 Eyelids remain closed with pain
Motor response
4 Thumbs-up, fist, or peace sign
3 Localizing to pain
2 Flexion response to pain
1 Extension response to pain
0 No response to pain or generalized myoclonus status
Brainstem reflexes
4 Pupil and corneal reflexes present
3 One pupil wide and fixed
2 Pupil or corneal reflexes absent
1 Pupil and corneal reflexes absent
0 Absent pupil, corneal, and cough reflex
Respiration
4 Not intubated, regular breathing pattern
3 Not intubated, Cheyne-Stokes breathing pattern
2 Not intubated, irregular breathing
1 Breathes above ventilatory rate
0 Breathes at ventilator rate or apnea
a
Modified with permission from Wijdicks EF, et al, Ann Neurol.9 © 2005 John Wiley and Sons.
1716 DECEMBER 2018

Checklist for the Evaluation of Acute Coma TABLE 6-6
History
◆ Sudden onset? Witnessed onset?
◆ Previous level of function
◆ Comorbid conditions
◆ Medications and toxic exposures
◆ Previous episodes of altered consciousness?
◆ Location (out of hospital, hospital floor, intensive care unit)
Physical Examination
◆ Vital signs
◆ Brainstem reflexes
◆ Funduscopy
◆ Motor responses to pain
◆ Lateralizing or other focal findings?
◆ Meningeal signs?
◆ Adventitious movements?
Blood (and Urine) Tests
◆ Glucose, electrolytes, blood urea nitrogen/creatinine, complete blood cell count
◆ Consider pH, PaCO2, liver enzymes, ammonia, ethanol, toxicologic screen (urine and
sometimes serum), levels of prescribed drugs, thyroid-stimulating hormone (TSH), cortisol
◆ Infectious workup (when pertinent)
◆ Autoimmune encephalitis panel (when pertinent)
Brain Imaging?
◆ Brain parenchyma (CT versus MRI)
◆ Brain vessels (arteries, veins)
Lumbar Puncture?
◆ Glucose, protein, nucleated cells
◆ Infectious workup (when pertinent)
◆ Autoimmune encephalitis panel (when pertinent)
EEG?
◆ Spot versus continuous
CT = computed tomography; EEG = electroencephalogram; MRI = magnetic resonance imaging; PaCO2 =

partial pressure of carbon dioxide, arterial.

FIGURE 6-2
Select causes of coma demonstrated by CT scan. Axial noncontrast images show epidural
hematoma (A, lower arrow) causing lateral brain displacement with upper brainstem
compression (A, upper arrow), malignant middle cerebral artery territory infarction with
subfalcine herniation (B, arrow), subarachnoid hemorrhage with global brain edema (C), and
right caudate hemorrhage with extensive intraventricular extension and hydrocephalus (D).
disorder that is causing the comatose state, in which case initiating treatment to
correct them would be futile, if not detrimental. Until more definitive
information is acquired, it is advisable to be conservative with the interpretation
of EEG abnormalities in patients who are comatose.
Differential Diagnosis
My mantra when evaluating coma of unclear etiology is “Think of treatable causes
first!” Many treatable causes of coma require emergency treatment to avoid poor
outcomes; therefore, failure to identify these causes in a timely manner can have
1718 DECEMBER 2018

KEY POINTS
● When evaluating coma of

unclear etiology, always
think of treatable causes
first.
● Do not assume coma is

irreversible when the cause
remains indeterminate.
FIGURE 6-3
Select causes of coma demonstrated by axial MRI. A, Diffusion-weighted image (DWI) shows
diffuse cortical injury after prolonged cardiac arrest. B, Fluid-attenuated inversion recovery
(FLAIR) image shows vasogenic edema in the posterior head regions in a patient with
posterior reversible encephalopathy syndrome (PRES). C, FLAIR image shows right
temporal swelling in a patient with herpes simplex virus type 1 encephalitis. D, FLAIR
image shows pituitary apoplexy (arrow).
devastating consequences (CASE 6-1).4,14 Some of the most notable examples are
basilar artery occlusion, status epilepticus, fulminant bacterial meningitis, severe
herpes simplex virus encephalitis, hydrocephalus, venous sinus thrombosis, some
intoxications, and brain herniation from an excisable mass or extraaxial fluid collection.
Recognizing severe irreversible damage and spontaneously reversible causes
has prognostic implications and may avoid unnecessary additional testing and
empirical treatment attempts. Because outcomes generally should not be affected
by a delayed etiologic diagnosis, or the evolution of the case, or the emergence of
new information, testing can be more conservative in these instances. In fact,
time itself may clarify the etiology. Coma should never be assumed to be
irreversible as long as the cause remains indeterminate. Failure to recognize

CASE 6-1 A 58-year-old man with multiple poorly controlled vascular risk factors
was brought to the emergency department with acute diplopia,
dysarthria, and right hemiparesis. He arrived 45 minutes after symptom
onset, and, shortly after arrival, he developed left hemiparesis, then
became comatose and was emergently intubated.
Examination showed slight anisocoria with preserved reactivity, absent
corneal and oculocephalic reflexes, and bilateral extensor responses
to pain (Full Outline of UnResponsiveness [FOUR] score E0, M1, B2, R1).
Head CT showed no acute parenchymal changes, but CT angiography
confirmed the clinical suspicion of basilar artery occlusion.
He was immediately taken to the angiographic suite, where
recanalization was rapidly achieved (FIGURE 6-4). By the following day, he
was fully awake and had a moderate right hemiparesis and a left sixth
nerve palsy. Subsequently, he was successfully extubated but required a
percutaneous gastrostomy. Three months later, he was swallowing well,
his diplopia was nearly resolved, and his hemiparesis had improved.
FIGURE 6-4
Angiogram of the patient in CASE 6-1. Before (A) and after (B) recanalization of the basilar
artery occlusion.
COMMENT Basilar artery occlusion is a paradigmatic example of a treatable cause of

coma compatible with a favorable prognosis but only if treated without any
delay. The progression of the deficits and the physical findings in this case
were sufficient to have a very strong suspicion of the correct diagnosis.
Focused testing and an immediate therapeutic decision resulted in a good
recovery; however, if not recognized and diagnosed early, major damage
to the patient inevitably occurs.
1720 DECEMBER 2018

spontaneously reversible causes, such as residual effects of sedative drugs in KEY POINTS
patients who are critically ill with impaired clearance, may result in misguided
● Treatable causes of coma
prognosis and premature withdrawal of life-sustaining measures. most often represent a
The differential diagnosis of coma differs considerably depending on where medical emergency.
the patient is when coma is first noticed. In patients who are found comatose,
a cerebrovascular event, unwitnessed trauma, seizure, cardiopulmonary arrest, ● The value of
pharmacologic or
fulminant CNS infection, sepsis, and intoxications should be primarily
nonpharmacologic
suspected. In patients who become comatose in the hospital, the most common neurostimulation for coma
causes are toxic (eg, sedatives, opiates) and metabolic derangements, which may recovery is unproven.
be compounded by sepsis and hypoperfusion from shock. Postoperative coma
is also often due to residual effects of medications, but anoxic-ischemic damage, ● The prognosis of coma
depends mostly on its cause
stroke, and nonconvulsive seizures also deserve careful consideration. After and duration.
neurosurgery, tension pneumocephalus should be added to the differential.
Cerebral fat embolism can occur with orthopedic surgeries involving long bones,
while air embolism can be a complication of complex cardiovascular operations.
Therapeutic Management
After ensuring that the patient’s ventilation and oxygenation are adequate and
the blood pressure is sufficient for organ perfusion, acute management depends
on the known or suspected cause of the coma. Treatable causes demand
emergent therapeutic interventions (TABLE 6-7).4,14 Other general principles to
follow include giving thiamine before administering dextrose, remembering that
administration of flumazenil (to reverse benzodiazepine effect) may be
complicated with seizures, and being aware that the effects of naloxone (to
reverse opiate effect) and flumazenil are transient.
Patients who remain comatose but have conditions that may improve over time
need intensive care to prevent additional brain injury and systemic complications.
Secondary brain injury can result from anoxia, ischemia, herniation, intracranial
hypertension, prolonged seizures, hypoglycemia, prolonged high fever, and any
other derangements that may induce cerebral energy failure (ie, energy demand
that is greater than the energy supply). Targeted temperature management to 33°C
(91.4°F) or 36°C (96.8°F) (both temperature targets having been found equivalent
in a solid randomized controlled trial)15 followed by strict avoidance of fever
should be part of the standard care of patients who remain comatose after
cardiopulmonary resuscitation.16 Patients who are comatose are at increased risk
of multiple systemic complications, including infections (most commonly
aspiration and ventilator-associated pneumonia, urinary tract infections related to
indwelling catheters, and bacteremia related to central venous catheters), venous
thromboembolism from immobility, skin breakdown, eye injury from exposure,
and malnutrition, among others.
The alternative of trying some form of neurostimulation to improve alertness is
a subject of debate. In a randomized controlled trial, amantadine was shown to
accelerate recovery of consciousness after traumatic brain injury but without
improving overall outcomes upon further follow-up.17 Other medications, such as
zolpidem, fluoxetine, methylphenidate, and other dopaminergic agents, are only
supported by anecdotal reports. The value of deep brain stimulation remains unproven.
Prognosis
The prognosis of coma depends on its cause and duration. Prognosis after
anoxic brain injury is generally poorer than for other causes of coma,

particularly trauma. Yet, prognosis after cardiac arrest is improving. For more
information on management of coma after cardiac arrest, refer to the article
“Management of Comatose Survivors of Cardiac Arrest” by David B. Seder,
MD, FCCP, FCCM, FNCS,18 in this issue of Continuum. Rapidly reversible
causes, even if initially severe, can have an excellent prognosis if patients have
no substantial residual brain injury (CASE 6-2). In general terms, it is true that
the longer the duration of coma, the lower the chances of regaining
consciousness. Yet, some patients may have a good outcome after a delayed
TABLE 6-7 Main Treatable Causes of Acute Coma and Their Treatments
Cause Treatment
Basilar artery occlusion IV thrombolysis, mechanical thrombectomy
Massive hemispheric infarction Hemicraniectomy (if indicated)
Status epilepticus Antiseizure medications
Intraparenchymal mass lesions Surgery, osmotherapy, corticosteroids (only if edema is vasogenic)
Extraaxial collections Surgery
Acute hydrocephalus Ventricular drainage
Bacterial meningitis Antibiotics and corticosteroids
Herpes simplex virus encephalitis Acyclovir; consider antiseizure medications
Autoimmune encephalitis Methylprednisolone, plasma exchange, or IV immunoglobulin (IVIg)
Venous sinus thrombosis Anticoagulation, consider local infusion of thrombolytic agent
Posterior reversible encephalopathy Blood pressure control, discontinuation of possible culprit drugs, antiseizure
syndrome (PRES) medications (if seizures)
Air embolism Consider hyperbaric oxygen (benefit possible but unproven)
Drug intoxications Stop offending drugs and administer antidotes when available
Hypercapnia Mechanical ventilation
Carbon monoxide poisoning Normobaric or hyperbaric oxygen
Hypoglycemia Dextrose preceded by thiamine
Diabetic coma Insulin, fluids
Uremia Dialysis
Hyperammonemia Lactulose, rifaximin
Pituitary apoplexy Corticosteroids, possible surgery
Myxedema Thyroid hormone
Wernicke encephalopathy Thiamine
Extreme hypothermia Slow rewarming
IV = intravenous.
1722 DECEMBER 2018

recovery of consciousness, most notably young patients with drug
intoxications.19 Favorable functional outcome is possible even after very
prolonged refractory status epilepticus.20 Some patients with autoimmune
encephalitis may be comatose for many days and then experience a dramatic
recovery with aggressive immune treatment.21 Clinicians should be
particularly cautious when estimating prognosis after severe traumatic brain
injury in young patients; in these situations, progressive recovery of
consciousness remains possible months after the injury.22
Prognostication in coma should be done cautiously and with distinct attention
to the inciting cause, secondary complications (especially anoxia-ischemia),
proven structural damage, and the patient’s age and previous condition.
Discussions with families regarding the optimal level of care for each patient
should always incorporate any previously expressed patient wishes or,
alternatively, the most likely patient preference in such a situation as understood
by the family.
A 35-year-old woman presented to the emergency department with CASE 6-2

headache and nausea. She was treated with analgesic and antiemetic
medications and was discharged home. Over the next 24 hours, she
developed worsening headache and neck pain, vomiting, and rapid
decline in her level of consciousness. She was taken back to the same
emergency department. A head CT scan was unremarkable, and she was
transferred to a tertiary care center. Before transportation, she needed
to be intubated for airway protection.
On arrival at the tertiary care center, she was febrile, tachycardic, and
hypotensive. Neurologic examination showed coma with preserved
brainstem reflexes and bilateral withdrawal responses to pain in her arms
with diffusely increased muscle tone (Full Outline of UnResponsiveness
[FOUR] score E0, M2, B4, R1), and she had marked neck stiffness. She was
immediately started on ceftriaxone, vancomycin, acyclovir, and
dexamethasone. After review of the initial head CT scan, she had a
lumbar puncture, and the CSF demonstrated pleocytosis with high
protein and mildly low glucose. Gram stain demonstrated gram-positive
cocci, and cultures rapidly grew Streptococcus pneumoniae. Within
24 hours of starting antibiotics, the patient began to improve, and she
eventually made a full recovery.
Coma from central nervous system infection is a medical emergency. In this COMMENT
case, prompt treatment of the acute bacterial meningitis allowed a
complete recovery, but any delay in diagnosis or empiric treatment could
have been catastrophic. While signs of sepsis and neck stiffness were clear
indicators of the correct diagnosis in this case, I have seen cases of severe
acute meningitis in which sepsis was attributed to the wrong source and
the stiffness was incorrectly blamed on another cause, such as a presumed
drug intoxication. This case also illustrates that a focused history and
examination can confidently guide additional testing.

KEY POINTS Prolonged and Chronic Disorders of Consciousness

Patients with impaired consciousness lasting more than 28 days are operationally
● Prolonged disorders of
consciousness (longer than
categorized as having a prolonged disorder of consciousness. These patients
28 days) are typically the typically have widespread cerebral damage and develop substantial brain
result of widespread brain atrophy over time.23 Depending on severity, the disorder can be classified as
damage. vegetative state or minimally conscious state. Recently, the term unresponsive
wakefulness syndrome has been proposed to replace the designation of vegetative
● Prolonged disorders of
consciousness constitute a state.24 Vegetative state, or unresponsive wakefulness syndrome, is characterized
spectrum that includes by spontaneous eye opening and sleep-wake cycles without any purposeful
vegetative state (also behavior suggestive of awareness of self or the environment. When such
referred to as unresponsive purposeful behavior becomes apparent, the patient has recovered consciousness.
wakefulness syndrome) and
minimally conscious state. If the patient continues to have severe alteration of consciousness but exhibits
purposeful behaviors, even if briefly, subtly, or inconsistently, the patient is said
● Although the problem is to be in a minimally conscious state. Subcategories of the minimally conscious
improving, practice state have been recently introduced: minimally conscious state plus, when the
variations in the
determination of brain death
patient regains language function (receptive, expressive, or both), and minimally
continue to occur across conscious state minus, when the signs of conscious awareness do not include
different states and language (but can include, for example, visual tracking, object manipulation, or
countries. affective reactions to external stimuli). Emergence from a minimally conscious
state is characterized by recovery of functional communication (verbal or
● The diagnosis of brain
death demands an gestural) or functional use of objects.
established cause of The distinction between vegetative state/unresponsive wakefulness syndrome
irreversible coma, the and minimally conscious state can be challenging at the bedside. There are
absence of confounding several standardized neurobehavioral assessment tools (such as the Coma
factors, the complete
absence of motor response Recovery Scale-Revised)25 that have been validated for prolonged disorders of
and brainstem reflexes, and consciousness, and they should be preferentially used and serially repeated for
no breathing efforts on a the evaluation of these patients (SDC APPENDIX A, links.lww.com/CONT/A256).26
formal apnea test. Furthermore, recent research using functional MRI (fMRI) and fludeoxyglucose
positron emission tomography (FDG-PET) imaging indicates that a small
minority of patients classified as vegetative state/unresponsive wakefulness
syndrome actually show changes in brain metabolism during mental activation
tasks compatible with a minimally conscious state.1,27 These findings might
predict that a patient may later have behaviorally evident recovery of
consciousness.26 Yet, it is important to clarify that the potential value of these
investigations has only been shown in subjects with prolonged disorders of
consciousness and not in patients with acute coma.
When medically stable, patients with prolonged disorders of consciousness
should be referred to centers with multidisciplinary teams specialized in the
monitoring of delayed complications and the rehabilitative care of these types
of cases.26
BRAIN DEATH
Brain death is the term often used for death determined by neurologic criteria. It
refers to a condition in which systemic circulation is preserved but no evidence of
any brain or brainstem function is present. The diagnosis does not permit
ambiguity. The implications are final, and, therefore, the diagnosis must be
definite and unquestionable.
Unfortunately, while the concept of brain death is broadly accepted across the
world, the conditions required for its determination vary in different countries
and even, to a lesser degree, across US states.28,29 Uniformity in hospital policies
1724 DECEMBER 2018

is growing across large institutions in the United States, but some variability
persists.30 Thus, physicians should be cognizant of the local policies and specific
legal requirements where they practice. Various national and international health
organizations, including the AAN, are actively working to achieve a consensus
on how to determine brain death regardless of where the patient is physically
located. Also, while one examination is sufficient to declare brain death in adults
in most US states, a second examination by a different physician is required in
pediatric cases, with the two examinations separated by 24 hours when the
patient’s age is between 37 weeks gestation and 1 month and separated by
12 hours when the patient’s age is between 1 month and 18 years.31
Clinical Criteria
In most instances, the diagnosis of brain death can and should be established by
the findings of the bedside neurologic examination. The AAN guidelines provide
a clear description of the steps that are necessary to determine brain death with
full reliability (FIGURE 6-5) (SDC APPENDIX B, links.lww.com/CONT/A257).2
Before proceeding with the clinical examination, two prerequisites must
always be met: (1) establish an irreversible cause of coma and (2) confirm the
absence of confounding factors, such as prescription and recreational drugs that
can depress brain function, pharmacologic neuromuscular blocking effect,
FIGURE 6-5
Schematic representation of the steps required for the determination of brain death.

hypothermia, hypotension, and major metabolic derangements (electrolytes,

acid base, or endocrine) (CASE 6-3).
The neurologic examination to assess for brain death should be performed in a
rigorous stepwise manner, with strict attention to detail.32
CONFIRM COMA. The patient should show no evidence of responsiveness to

any stimulation, whether eye opening, eye movements, or any other motor
response that cannot be explained by spinal reflexes.
CONFIRM ABSENCE OF BRAINSTEM REFLEXES. The patient’s pupils should be

midsized and nonreactive to bright light. Corneal stimulation should not elicit
blinking or any other response. On oculocephalic (only if cervical spine integrity
allows it) and oculovestibular (ie, cold caloric) reflex testing, the eyes should
not move. Facial muscles should not contract upon pressure applied to the
temporomandibular joints, gag reflex should not be activated by stimulation
of the posterior pharyngeal wall, and cough reflex should be not be elicited
despite stimulation of the carina with a suctioning catheter.
CONFIRM APNEA. Complete absence of a breathing drive despite high PaCO2

should be established by a formal apnea test. The apnea test itself has a set of
prerequisites and rigorous steps that must be respected, as detailed in TABLE 6-8.
CASE 6-3 A 72-year-old man was brought to the emergency department after a
witnessed cardiac arrest from ventricular fibrillation. After multiple
electric shocks and prolonged advanced cardiopulmonary resuscitation
efforts, he regained spontaneous circulation but remained comatose. He
was treated with a targeted temperature management protocol during
the first 24 hours using a target temperature of 33°C (91.4°F). Because
of shivering, he was given neuromuscular blocking agents and propofol.
EEG performed through rewarming showed a burst-suppression pattern.
He had a history of diabetic nephropathy and developed superimposed
acute kidney injury. Elevated liver transaminases were also noticed. Early
on day 3 after the arrest, the cardiology team questioned whether the
patient might be brain dead.
COMMENT This case illustrates the common scenario of a patient who is comatose
with very poor prognosis for whom a team of non-neurologists brings up
the possibility of brain death without considering the presence of
confounding factors. After therapeutic hypothermia, the neurologic
examination is often contaminated by the lingering effects of sedative
drugs, which may be prolonged because of delayed clearance from kidney
and liver dysfunction and hypothermia itself. In the case presented, brain
death could not be determined even if the examination had shown the
absence of brainstem reflexes and motor responses to pain. Thus, in this
situation, it would be incorrect to proceed with a formal evaluation for
brain death.
1726 DECEMBER 2018

Apnea tests are feasible in the majority of patients with irreversible coma, even
among most patients with severe cardiopulmonary disease, when these steps are
carefully followed.33 Ensuring steady delivery of oxygen and having immediate
access to vasopressors (most of these patients are already on vasopressors by
the time of the apnea test) are essential conditions for successful completion
of the test. In the rare instances that the test must be aborted because of
hypotension or hypoxemia, it is reasonable to consider a new attempt after
adequate preparation. The time of conclusion of the apnea test, which is when
the PaCO2 is shown to be sufficiently high, represents the legal time of
expiration. After that, state laws require the physician to inform the local organ
procurement organization.
Ancillary Tests
Several ancillary tests can be used to confirm brain death, but all have limitations.34
They can complement the neurologic examination but should never replace it.
Their most appropriate role is when certain aspects of the neurologic examination
cannot be performed (eg, when an apnea test cannot be safely completed in a
patient with severe pulmonary contusions and acute respiratory distress
syndrome). However, some countries and a few US states require one of these
tests for confirmatory purposes in all cases of brain death determination.
EEG, cerebral scintigraphy with nuclear scan, and catheter cerebral
angiography are the traditional ancillary tests, but transcranial Doppler and, more
recently, CTA have been added to the list.35 Specific considerations apply to the
performance of all these tests for the indication of possible brain death
The Apnea Test TABLE 6-8
Prerequisites
◆ PaCO2 between 35 mm Hg and 45 mm Hg
◆ Systolic blood pressure ≥100 mm Hg with or without vasopressors
◆ Administer 100% oxygen for at least 10 minutes (ideal PaO2 >200 mm Hg with positive
end-expiratory pressure ≤5 cm H2O)
◆ Absence of clinical signs of intravascular volume contraction
Steps
◆ Disconnect the patient from the ventilator
◆ Deliver oxygen at 6 L/min through a catheter advanced through the tracheal tube until close
to the carina
◆ Look carefully for any respiratory movements while monitoring pulse oximetry and blood
pressure
◆ Abort and reconnect to the ventilator if evidence of respiratory movements, refractory
hypotension (systolic blood pressure <90 mm Hg) or worsening hypoxemia (pulse
oximetry <85%)
◆ If no respiratory movements after approximately 8 minutes, obtain arterial blood gases
◆ Apnea is established if PaCO2 ≥60 mm Hg (or 20 mm Hg greater than baseline)
PaCO2 = partial pressure of carbon dioxide, arterial; PaO2 = partial pressure of oxygen, arterial.

(TABLE 6-9). False negatives and, most concerningly, false positives (ie, results
that would falsely support the diagnosis of brain death) have been reported with
these tests when compared to the gold standard of physical examination.
Therefore, it is essential that their pitfalls be carefully considered.34
Common Pitfalls and Persistent Controversies

In addition to the limitations and potentially false results of ancillary tests, several
other pitfalls must be kept in mind. Inadequate consideration of confounding
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 artifacts (common in the
electric activity, including lack of intensive care unit); mostly
Interelectrode distance ≤10 cm reactivity to intense, painful, visual, evaluates the cortex
and auditory stimulation
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
Nuclear Isotope injection within No brain perfusion (hollow skull) Incorrect injection (can be
medicine 30 minutes of reconstitution; avoided by confirming uptake in
scana anterior and bilateral planar image the liver)
counts upon injection and after
30 minutes, 1 hour, and 2 hours
Transcranial Bilateral transtemporal and Reverberating arterial flow or small Lack of reliable signal because of
Doppler transforaminal insonation; peaks in early systole poor temporal bone window;
transorbital window insonation highly dependent on skill of
can be considered operator; absence of flow is not
reliable because it may be due to
poor windows or poor technique
Catheter Contrast injection in the arch and Absence of flow in intracranial Inadequate pressure upon
angiography under high pressure arteries injection; partial filling of
intracranial arteries without
reaching perfusing branches
CT Contrast injection from a Absence of flow in distal middle May be unreliable in low-flow
angiography peripheral vein with a pressure cerebral arteries states (delayed perfusion may be
injector; arterial and venous missed by usual timing of image
phases should be imaged acquisition); sensitivity is limited
when only using arterial filling as
diagnostic criterion; absent flow
in internal cerebral veins may
increase sensitivity
CT = computed tomography; EEG = electroencephalogram.

a
Cerebral scintigraphy with technetium-99m-hexamethylpropylene amine oxime (99mTc-HMPAO).
1728 DECEMBER 2018

factors, especially residual effects of drugs (CNS depressants and neuromuscular KEY POINTS
blocking agents), is probably the most common. When in doubt, measurement
● Angiography, nuclear
of serum levels of administered drugs is always advisable. Misidentification of scans, EEG, and transcranial
spinal reflexes as motor responses is another common problem. Movements Doppler are ancillary tests
generated by spinal reflexes are provoked by neck flexion or stimulation below that can be used to confirm
the neck but not by stimulation above the neck. They can have various brain death, but they all
have limitations.
appearances (most frequent are leg triple flexion, finger flexion or extension, or
slight head turning),36 but in the same patient, they are always stereotypical. ● The most appropriate
Therefore, repeated testing may aid in their differentiation from motor responses use of ancillary tests to
generated higher in the neuraxis; however, stereotypy itself is not confirm brain death is to
pathognomonic of spinal reflexes because it can also be observed with extensor complement the neurologic
examination when certain
or flexor posturing. Experience is necessary for reliable distinction. So-called aspects of the examination
“brain death mimics” exist (eg, Guillain-Barré syndrome, organophosphate (such as the apnea test)
poisoning, lidocaine toxicity, baclofen overdose), but attentive clinicians should cannot be performed.
be able to recognize them. Another common problem is inadequate preparation
● Not recognizing
of the patient before starting the apnea test. Insufficient preoxygenation, lack of confounding factors
access to titratable vasopressors, and failure to correct volume depletion can (especially the residual
result in failure to complete the apnea test while exposing organs to potential effects of central nervous
injury from hypoperfusion and hypoxemia. system depressants) is a
common pitfall in brain
Lingering controversies persist. Some consider that ancillary tests are
death evaluations.
indicated to evaluate supratentorial function in patients with devastating
brainstem damage. The ancillary test would be necessary to confirm that these ● The concept of brain
patients are not in a complete locked-in state. In these cases, supratentorial death is accepted by all
blood flow may be present initially but becomes absent within a few days at the major religions, but small
cultural and religious groups
latest.37 Exceptional cases of erroneous clinical diagnosis of brain death have continue to challenge the
been reported, but in all instances, confounding factors that failed to be validity of the concept.
recognized could explain the diagnostic error.38 While all major religions across
the world endorse the concept of brain death, small cultural and religious
groups continue to challenge the validity of the concept.39 Clearer and more
consistent laws are the only way these rare and very problematic disputes may
be avoided in the future.
CONCLUSION
The evaluation of the patient who is comatose requires attention to detail and a
methodical approach to avoid missing treatable causes and secondary injurious
complications. History and examination should guide additional testing to
navigate across the multiple possible causes of coma. Prognosis depends
primarily on the cause of the coma and should be delivered cautiously because
delayed recovery is possible in various situations.
Brain death can only be diagnosed after concluding that a patient has a known
cause of irreversible coma and proving the complete absence of all signs of brain
and brainstem function. Confounding factors (such as drug effects and
hypothermia) must always be considered and specifically excluded. In most
instances, brain death can and should be determined based on neurologic
examination; use of ancillary tests is generally unnecessary.

REFERENCES
1 Jox RJ, Bernat JL, Laureys S, Racine E. Disorders 13 Struck AF, Osman G, Rampal N, et al.
of consciousness: responding to requests for Time-dependent risk of seizures in critically ill
novel diagnostic and therapeutic interventions. patients on continuous electroencephalogram.
Lancet Neurol 2012;11(8):732–738. doi:10.1016/ Ann Neurol 2017;82(2):177–185. doi:10.1002/
S1474-4422(12)70154-0. ana.24985.
2 Wijdicks EF, Varelas PN, Gronseth GS, et al. 14 Rabinstein AA, Wijdicks EF. Management of the
Evidence-based guideline update: determining comatose patient. Handb Clin Neurol 2008;90:
brain death in adults: report of the Quality 353–367. doi:10.1016/S0072-9752(07)01720-4.
Standards Subcommittee of the American
15 Nielsen N, Wetterslev J, Cronberg T, et al; TTM
Academy of Neurology. Neurology 2010;74(23):
Trial Investigators. Targeted temperature
1911–1918. doi:10.1212/WNL.0b013e3181e242a8.
management at 33°C versus 36°C after cardiac
3 Wijdicks EF. Management of the comatose arrest. N Engl J Med 2013;369(23):2197–2206.
patient. Handb Clin Neurol 2017;140:117–129. doi:10.1056/NEJMoa1310519.
doi:10.1016/B978-0-444-63600-3.00008-8.
16 Geocadin RG, Wijdicks E, Armstrong MJ, et al.
4 Edlow JA, Rabinstein A, Traub SJ, Wijdicks EF. Practice guideline summary: reducing brain injury
Diagnosis of reversible causes of coma. following cardiopulmonary resuscitation: report
Lancet 2014;384(9959):2064–2076. doi:10.1016/ of the guideline development, dissemination,
S0140-6736(13)62184-4. and implementation subcommittee of the
American Academy of Neurology. Neurology
5 Jennett B, Teasdale G, Braakman R, et al.
2017;88(22):2141–2149. doi:10.1212/
Predicting outcome in individual patients after
WNL.0000000000003966.
severe head injury. Lancet 1976;1(7968):1031–1034.
doi:10.1016/S0140-6736(76)92215-7. 17 Giacino JT, Whyte J, Bagiella E, et al. Placebo-
controlled trial of amantadine for severe
6 Teasdale G, Jennett B. Assessment of coma and
traumatic brain injury. N Engl J Med 2012;366(9):
impaired consciousness:a practical scale.
819–826. doi:10.1056/NEJMoa1102609.
Lancet 1974;304(7872):81–84. doi:10.1016/
S0140-6736(74)91639-0. 18 Seder DB. Management of comatose survivors of
cardiac arrest. Continuum (Minneap Minn) 2018;
7 Rutledge R, Lentz CW, Fakhry S, Hunt J.
24(6 Neurocritical Care):1732–1752.
Appropriate use of the Glasgow Coma Scale in
intubated patients: a linear regression prediction 19 Braksick SA, Rabinstein AA, Wijdicks EF, et al.
of the Glasgow verbal score from the Glasgow Post-ischemic myoclonic status following cardiac
eye and motor scores. J Trauma 1996;41(3):514–522. arrest in young drug users. Neurocrit Care 2017;
26(2):280–283. doi:10.1007/s12028-016-0317-z.
8 Meredith W, Rutledge R, Fakhry SM, et al. The
conundrum of the Glasgow Coma Scale in 20 Cooper AD, Britton JW, Rabinstein AA. Functional
intubated patients: a linear regression prediction and cognitive outcome in prolonged refractory
of the Glasgow verbal score from the Glasgow status epilepticus. Arch Neurol 2009;66(12):
eye and motor scores. J Trauma 1998;44(5): 1505–1509. doi:10.1001/archneurol.2009.273.
839–844.
21 Mittal MK, Rabinstein AA, Hocker SE, et al.
9 Wijdicks EF, Bamlet WR, Maramattom BV, et al. Autoimmune encephalitis in the ICU: analysis of
Validation of a new coma scale: the FOUR score. phenotypes, serologic findings, and outcomes.
Ann Neurol 2005;58(4):585–593. doi:10.1002/ Neurocrit Care 2016;24:240–250. doi:10.1007/
ana.20611. s12028-015-0196-8.
10 Wijdicks EF, Kramer AA, Rohs T Jr, et al. 22 Vedantam A, Robertson CS, Gopinath SP.
Comparison of the full outline of Clinical characteristics and temporal profile of
UnResponsiveness score and the Glasgow Coma recovery in patients with favorable outcomes at
Scale in predicting mortality in critically ill 6 months after severe traumatic brain injury.
patients. Crit Care Med 2015;43(2):439–444. J Neurosurg 2018;129:234–240. doi:10.3171/2017.3.
doi:10.1097/CCM.0000000000000707. JNS162720.
11 Wijdicks EF, Rabinstein AA, Bamlet WR, 23 Guldenmund P, Soddu A, Baquero K, et al.
Mandrekar JN. FOUR score and Glasgow Coma Structural brain injury in patients with disorders
Scale in predicting outcome of comatose of consciousness: a voxel-based morphometry
patients: a pooled analysis. Neurology 2011;77(1): study. Brain Inj 2016;30(3):343–352. doi:10.3109/
84–85. doi:10.1212/WNL.0b013e318220ac06. 02699052.2015.1118765.
12 Fugate JE, Rabinstein AA, Claassen DO, et al. The 24 Laureys S, Celesia GG, Cohadon F, et al.
FOUR score predicts outcome in patients after Unresponsive wakefulness syndrome: a new
cardiac arrest. Neurocrit Care 2010;13(2):205–210. name for the vegetative state or apallic syndrome.
doi:10.1007/s12028-010-9407-5. BMC Med 2010;8:68. doi:10.1186/1741-7015-8-68.
1730 DECEMBER 2018

25 Giacino JT, Kalmar K, Whyte J. The JFK Coma 32 Wijdicks EF. Determining brain death. Continuum
Recovery Scale-Revised: measurement (Minneap Minn) 2015;21(5 Neurocritical Care):
characteristics and diagnostic utility. Arch Phys 1411–1424. doi:10.1212/CON.0000000000000221.
Med Rehabil 2004;85(12):2020–2029.
33 Datar S, Fugate J, Rabinstein A, et al. Completing
26 Giacino J, Katz DI, Schiff ND, et al. Practice the apnea test: decline in complications.
guideline update recommendations summary: Neurocrit Care 2014;21(3):392–396. doi:10.1007/
disorders of consciousness: report of the s12028-014-9958-y.
Guideline Development, Dissemination, and
34 Wijdicks EF. The case against confirmatory
Implementation Subcommittee of the American
tests for determining brain death in adults.
Academy of Neurology; the American Congress
Neurology 2010;75(1):77–83. doi:10.1212/
of Rehabilitation Medicine; and the National
WNL.0b013e3181e62194.
institute on Disability, Independent Living, and
Rehabilitation Research. Neurology 2018;91(10): 35 Kramer AH, Roberts DJ. Computed tomography
450–460. doi:10.1212/WNL.0000000000005926. angiography in the diagnosis of brain death:
a systematic review and meta-analysis.
27 Stender J, Gosseries O, Bruno MA, et al.
Diagnostic precision of PET imaging and
s12028-014-9997-4.
functional MRI in disorders of consciousness: a
clinical validation study. Lancet 2014;384(9942): 36 Saposnik G, Bueri JA, Mauriño J, et al.
514–522. doi:10.1016/S0140-6736(14)60042-8. Spontaneous and reflex movements in brain
death. Neurology 2000;54(1):221–223. doi:10.1212/
28 Wahlster S, Wijdicks EF, Patel PV, et al. Brain
WNL.54.1.221.
death declaration: practices and perceptions
worldwide. Neurology 2015;84(18):1870–1879. 37 Varelas PN, Brady P, Rehman M, et al. Primary
doi:10.1212/WNL.0000000000001540. posterior fossa lesions and preserved
supratentorial cerebral blood flow: implications
29 Greer DM, Wang HH, Robinson JD, et al.
for brain death determination. Neurocrit Care
Variability of brain death policies in the
2017;27(3):407–414. doi:10.1007/s12028-017-0442-3.
United States. JAMA Neurol 2016;73(2):213–218.
doi:10.1001/jamaneurol.2015.3943. 38 Wijdicks EF, Varelas PN, Gronseth GS, Greer DM.
There is no reversible brain death. Crit Care
30 Wang HH, Varelas PN, Henderson GV, et al.
Med 2011;39(9):2204–2205; author reply 2206.
Improving uniformity in brain death
doi:10.1097/CCM.0b013e318222724e.
determination policies over time. Neurology
2017;88(6):562–568. doi:10.1212/ 39 Lewis A, Greer D. Current controversies in brain
WNL.0000000000003597. death determination. Nat Rev Neurol 2017;13(8):
505–509. doi:10.1038/nrneurol.2017.72.
31 Nakagawa TA, Ashwal S, Mathur M, et al.
Guidelines for the determination of brain death
in infants and children: an update of the 1987 task
force recommendations-executive summary.
Ann Neurol 2012;71(4):573–585. doi:10.1002/
ana.23552.

REVIEW ARTICLE

Management of
ONLINE
Comatose Survivors
of Cardiac Arrest
By David B. Seder, MD, FCCP, FCCM, FNCS
ABSTRACT
PURPOSE OF REVIEW: Because the whole-body ischemia-reperfusion insult
associated with cardiac arrest often results in brain injury, neurologists
perform an important role in postresuscitation cardiac arrest care.
This article provides guidance for the assessment and management of
brain injury following cardiac arrest.
RECENT FINDINGS: Neurologists have many roles in postresuscitation cardiac

arrest care: (1) early assessment of brain injury severity to help inform triage
for invasive circulatory support or revascularization; (2) advocacy for the
maintenance of a neuroprotective thermal, hemodynamic, biochemical,
and metabolic milieu; (3) detection and management of seizures;
(4) development of an accurate, multimodal, and conservative approach to
prognostication; (5) application of shared decision-making paradigms
around the likely outcomes of therapy and the goals of care; and
(6) facilitation of the neurocognitive assessment of survivors. Therefore,
optimal management requires early neurologist involvement in patient
CITE AS:
care, a detailed knowledge of postresuscitation syndrome and its complex
CONTINUUM (MINNEAP MINN) interactions with prognosis, expertise in bringing difficult cases to their
optimal conclusions, and a support system for survivors with cognitive
1732–1752.
deficits.
Dr David B. Seder, Maine Medical SUMMARY: Neurologists have a critical role in postresuscitation cardiac
Center, 22 Bramhall St, Portland,
ME 04103, sederd@mmc.org.
arrest care and are key participants in the treatment team from the time of
first restoration of a perfusing heart rhythm through the establishment of
RELATIONSHIP DISCLOSURE: rehabilitation services for survivors.
Dr Seder receives grant support
from the Patient-Centered
Outcomes Research Institute
(CER-1602-34137).
INTRODUCTION
C
UNLABELED USE OF ardiac arrest is the final common pathway by which patients may
die from a myriad of disease processes; as such, the underlying
USE DISCLOSURE:
Dr Seder discusses the etiology of an arrest varies widely among those who survive
unlabeled/investigational use cardiopulmonary resuscitation. Determining and reversing the
of the bispectral index in
determining the severity of brain
cause of the arrest to prevent circulatory collapse and rearrest as
injury after cardiac arrest. well as maintaining a neuroprotective hemodynamic and biochemical milieu are
top treatment priorities in the early hours after resuscitation. However, not all
© 2018 American Academy therapies are appropriate for all patients, and the severity of the brain injury is a
of Neurology. key determinant of whether maximal resources, such as the use of extracorporeal
1732 DECEMBER 2018

membrane oxygenation (ECMO), mechanical circulatory support, or even KEY POINTS
multivessel coronary artery bypass grafting, are advisable or appropriate. During
● Top treatment priorities in
the early process of stabilization, neurologists should assist with the evaluation of the early hours after
brain injury and advise the emergency medicine, critical care, cardiology, and resuscitation include
surgical teams on resource-intensive decisions about the extent of circulatory determining and reversing
support to be attempted. This article discusses and reviews the critical the cause of cardiac arrest
to prevent circulatory
pathophysiology of the postresuscitation state.
collapse and rearrest as
well as maintaining a
APPROACH TO MANAGEMENT neuroprotective
Following resuscitation, an ischemia-reperfusion syndrome occurs, in which hemodynamic and
biochemical milieu.
systemic blood flow is reestablished, the damaged endothelial matrix of the
cerebral microcirculation may be restored or fail, and injured cells and tissues ● Restoration and
either initiate repair or apoptosis. These processes are profoundly influenced by maintenance of a favorable
blood flow, temperature, arterial oxygen tension, glucose levels, serum pH, thermal, hemodynamic,
seizure activity, and the underlying condition of the microcirculation. The most biochemical, and metabolic
milieu after cardiac arrest is
sensitive tissues are in the brain, and neurologic recovery is dependent on of great concern and should
prompt restoration of systemic homeostasis. As such, restoration and maintenance be a key element of urgent
of a favorable thermal, hemodynamic, biochemical, and metabolic milieu is of recommendations.
critical importance and should be a key element of urgent recommendations.
● Because many patients
Depending on the patient case mix of an individual center, seizures and
who undergo targeted
myoclonus occur in up to one-third of patients after resuscitation,1,2 and, temperature management
although both findings have been associated with poor functional outcomes, after cardiac arrest receive
neither offers definitive evidence of a catastrophic brain injury. Many patients neuromuscular blockade to
control shivering (and
with postresuscitation status epilepticus or myoclonus after an arrest have been
therefore never have
successfully treated and achieved favorable functional outcomes,2–4 suggesting convulsions) and because
that seizures may be more appropriately viewed as an ongoing source of injury to others experience primarily
be suppressed rather than a prognostic marker.5,6 Seizures and myoclonus should nonconvulsive seizures, it is
be evaluated with EEG, and ongoing seizure activity should be pharmacologically ideal to monitor for seizures
with continuous EEG.
suppressed. No data are available regarding the role of prophylactic antiepileptic
medications in this population. Because many patients who undergo targeted ● Prognostication after
temperature management after resuscitation receive neuromuscular blockade to cardiac arrest is
control shivering (and therefore never have convulsions) and because others problematic. Many of the
frequently cited studies
experience primarily nonconvulsive seizures, it is ideal to monitor for seizures describing prognostic
with continuous EEG. Seizures that arise after resuscitation from cardiac arrest factors after cardiac arrest
differ morphologically and temporally from seizures that occur in the healthy are flawed, and imprecise
brain and can be difficult to control7; the input of a neurologist for both detection early prognostication can
lead to unnecessary deaths.
and treatment is therefore crucial.
Prognostication after cardiac arrest is problematic because of the confounding
effects of sedating medications,8 impaired drug metabolism after resuscitation
and during hypothermia therapy,9 inconsistent methodology and findings of
clinical studies, and the effects of self-fulfilling prophecy on retrospective data
analyses.10 Many of the frequently cited studies describing prognostic factors
after cardiac arrest are flawed, and imprecise early prognostication can lead to
unnecessary deaths.11 It is therefore necessary to carefully review prognostic
criteria using recent guidelines12 and to advocate for the collection of prognostic
data in real time at the patient’s bedside, such as raw or processed EEG at 6 to
18 hours after resuscitation13,14; neuron-specific enolase levels measured at 24 to
72 hours after resuscitation15; diffusion-weighted imaging, apparent diffusion
coefficient, and fluid-attenuated inversion recovery (FLAIR) MRI sequences in
patients who are persistently comatose at 72 to 120 hours after resuscitation16,17;

COMATOSE SURVIVORS OF CARDIAC ARREST
recurrent neurologic examinations at standard time points off sedation (with

adequate time and circumstances for the clearance of sedating drugs); and
somatosensory evoked potentials in selected patients.18 In the absence of such data,
prognostication is imprecise, especially in patients receiving sedating medications
to suppress seizures or who have been managed, even days earlier, with
benzodiazepine infusions.19 After the application of diagnostic testing over
3 to 7 days, patients will fall into one of three categories: those likely to awaken and
make a good neurologic recovery, those unlikely to ever awaken, and those with
indeterminate findings. This discussion must be presented to distressed family
members by neurologists fluent in the meaning of the test results, who understand
how the neurologic findings intersect with medical issues such as age and organ
dysfunction, and who can skillfully and empathetically help families determine the
right course of action for an individual patient based on his or her prognosis and
likely wishes, premorbid functional status, and other patient-specific factors. This
process of shared decision making, in which clinicians, families, and other
stakeholders come together to determine a rational course of action, is the standard
of care20 and relies upon an engaged neurologist as part of the treatment team.
Finally, cognitive dysfunction is a significant source of morbidity among
survivors of cardiac arrest21,22 and should be directly addressed by the
CASE 7-1 A 37-year-old woman was brought to the emergency department by

emergency medical services (EMS) after she reported dyspnea and
collapsed in front of her family following a 2-week viral illness. Her family
members immediately began chest compressions, and when EMS
arrived, her first recorded cardiac rhythm was ventricular fibrillation. She
received a total of 32 minutes of cardiopulmonary resuscitation by her
family and EMS in addition to epinephrine, bicarbonate, and magnesium
before restoration of a perfusing heart rhythm.
On arrival to the emergency department, her blood pressure was
75/40 mm Hg despite infusion of norepinephrine and dopamine, serum
pH was 7.08, serum bicarbonate was 11 mmol/L, and venous lactate level
was 12 mmol/L. Echocardiography showed dilated cardiomyopathy, with
very weak biventricular contraction. Six hours after resuscitation, with an
epinephrine infusion running, and low-dose propofol and fentanyl
infused to assure comfort during temperature management, her blood
pressure and tissue perfusion were only slightly improved. Cardiology
and cardiothoracic surgery consultants convened with the critical care
team at the bedside to determine her suitability for extracorporeal
membrane oxygenation (ECMO) as a bridge to transplantation or a
mechanical ventricular assistance device, and they requested urgent
neurology consultation to help determine the severity of the brain injury
and for recommendations in managing her neurologic injury.
On neurologic evaluation, the patient was undergoing targeted
temperature management at 36°C (96.8°F). After fentanyl and propofol
infusions were discontinued for 60 minutes, the patient did not follow
commands, orient to voice, or attend. Brainstem reflexes were present,
and her Full Outline of UnResponsiveness (FOUR) score23 was E0 (no eye
1734 DECEMBER 2018

neurologist. Many seemingly recovered survivors of cardiac arrest exhibit
debilitating deficits in memory, learning, and executive function. Neurologists
should screen for these, and larger centers should have established pathways for
neurocognitive counseling and follow-up.
EARLY NEUROLOGIC RISK ASSESSMENT FOLLOWING RESUSCITATION

Postresuscitation cardiac arrest care has become increasingly complex and raises
questions regarding assessment of neurologic risk and recovery (CASE 7-1).
Although the mechanism of the arrest varies (eg, pulmonary embolism, acute
coronary syndrome, primary arrhythmia, idiopathic cardiomyopathy, or some
other insult), the technology to support a patient with a failing or failed
systemic circulation is now widely available, and neurologists are increasingly
being asked to provide an assessment of neurologic risk, meaning the
likelihood of death by neurologic criteria, severe neurologic disability, or
death due to the withdrawal of life-sustaining therapies because of a poor
neurologic prognosis in patients who will require mechanical circulatory
support to survive. Resource-intensive therapy such as ECMO is unreasonable
to apply to a patient who has already sustained a severe brain injury and
is unlikely to ever recover but is very reasonable to offer a young patient
opening to pain), M3 (localizing to pain), B4 (pupil and corneal reflexes

present), and R1 (breathes above the ventilator rate). (For more
information on the FOUR score, refer to the article “Coma and Brain
Death” by Alejandro A. Rabinstein, MD, FAAN,24 in this issue of
Continuum.) She moved her extremities spontaneously and withdrew to
pain. EEG showed a continuous background without epileptiform activity,
with no evidence of subtle seizure activity. Fentanyl and propofol were
restarted for comfort and to suppress her vigorous shivering.
This patient was determined to be an appropriate candidate for ECMO
based on her stratification as having a low to intermediate risk of poor
neurologic outcome. This assessment was based on her FOUR score,
motor examination, continuous EEG background early after resuscitation
(low risk category), and a relatively prolonged total ischemic time
(moderate risk category).
One week after being placed on ECMO, her cardiac function recovered.
She was weaned from sedation and the ECMO circuit, awoke, and was
extubated. At 3 months postarrest, she had returned to the community
with a steadily improving functional status and no serious cognitive deficits.
This case is representative of the typical consultation in which neurologists COMMENT

are called to provide expert opinion on questions such as what data help
guide urgent neurologic evaluation after resuscitation from cardiac arrest,
what additional testing (if any) should be obtained, whether a reasonable
possibility of neurologic recovery exists, and what supportive measures will
maximize the likelihood of a good outcome.

who was previously healthy and has a moderate or high likelihood of

neurologic recovery.25
Unlike traditional prognostication after a cardiac arrest, the goal of early
neurologic evaluation is to classify patients into categories of high, medium,
and low risk of a poor neurologic outcome in a time frame that allows for triage
to individualized treatment pathways (TABLE 7-1). For most patients at this
stage, the severity of the neurologic injury is still in evolution and may be directly
impacted by clinical decisions and the quality of critical care provided. Such
risk categories are easily understood by cardiologists, surgeons, other clinicians
involved in the shared decision-making paradigm, and family members. As
such, they are useful in risk-benefit assessments of mechanical circulatory
support, early percutaneous or surgical coronary artery revascularization when
necessary, and other key therapeutic decisions. Rapid diagnostic testing
modalities that allow for accurate early neurologic triage include assessment
of the total ischemic time (from collapse to the return of spontaneous
circulation), the no-flow (from collapse to onset of cardiopulmonary
resuscitation) and low-flow (cardiopulmonary resuscitation duration) intervals,
neurologic examination (including the Full Outline of UnResponsiveness
[FOUR] score), CT, EEG background, continuity and reactivity at different time
TABLE 7-1 Early Neurologic Risk Categories After Cardiac Arrest
Neurologic
Triage Risk Class Neurologic Riska Characteristicsb
Low <20% No-flow intervalc <3 minutes, total ischemic time (collapse to return of
spontaneous circulation) <20 minutes, Glasgow Coma Scale (GCS) motor
subscore 4+, total GCS >5T, preserved gray-white matter junctiond on head CT,
EEG continuous and reactive, initial bispectral index score after neuromuscular
blockadee >20, or suppression ratio (percent fully suppressed EEG) at 6 hours
postresuscitationf <10%, no seizures or myoclonus
Medium 20–80% No-flow intervalc 3–8 minutes, total ischemic time >20 minutes, GCS motor
subscore 2–3, total GCS 5T, gray-white matter junction preserved,d continuous
EEG background, initial bispectral index after neuromuscular blockadee 5–20,
suppression ratio (percent fully suppressed EEG) at 6 hours postresuscitationf
11–69%, may have early seizures or multifocal cortical myoclonus
High >80% No-flow intervalc >8 minutes, GCS motor subscore 1, total GCS 3–4T, no pupillary
reaction and no corneal reflex at admission, loss of gray-white matter junction
on CT,d nonreactive and discontinuous or near-isoelectric EEG background,
bispectral indexe <10, suppression ratio (percent fully suppressed EEG) at
6 hours postresuscitationf >70, seizures arising from a discontinuous
background or early myoclonus with an EEG correlate
CT = computed tomography; EEG = electroencephalography.

a
Neurologic risk refers to risk due to brain death, poor neurologic outcome, or withdrawal of life-sustaining therapies because of poor neurologic
prognosis.
b
Findings are present during the initial 12 hours after resuscitation.
c
No-flow interval is the time from arrest to onset of cardiopulmonary resuscitation.
d
The loss of gray-white differentiation on head CT has been described quantitatively as the ratio of Hounsfield units <1.2 measured in deep gray
matter contrasted to the adjacent white matter tracts.26
e
Bispectral index measured in a patient who is chemically paralyzed as soon as possible after return of spontaneous circulation (median 4.5 hours).
f
Suppression ratio is the percentage of each EEG epoch that is isoelectric or very low voltage. There are prognostic implications to the
suppression ratio measured 6 hours after resuscitation.
1736 DECEMBER 2018

points after resuscitation, and the bispectral index and suppression ratio KEY POINTS
(processed, quantitative EEG modalities widely used in anesthesia and critical
● Cognitive dysfunction is a
care units that integrate various EEG parameters and have been shown to significant source of
correlate with neurologic outcomes after resuscitation) (TABLE 7-227–36). morbidity among survivors
Neurologic risk stratification after cardiac arrest is a strategy to facilitate triage of cardiac arrest.
to the most favorable treatment paradigm for an individual patient immediately
● Unlike traditional
after resuscitation when the brain is believed to be highly vulnerable. It rests
prognostication after a
upon a few principles: cardiac arrest, the goal of
early neurologic evaluation
is to classify patients into
u Patients whose risk assessment suggests mild or moderate brain injury should be
categories of high, medium,
considered for initiation of aggressive strategies for circulatory support, as they are
and low risk of a poor
likely to benefit, up to and including ECMO or surgical revascularization, if necessary,
neurologic outcome in a
and should undergo an accelerated diagnostic workup to identify the cause of the arrest
time frame that allows for
with the intention of maintaining adequate cerebral blood flow and preventing rearrest.
triage to individualized
u Patients whose risk assessment suggests severe brain injury should be offered treatment pathways.
resource-intensive modes of circulatory support such as ECMO only if they demonstrate
recovery of cortical function. Because the likelihood of benefit is very low, evidence ● Rapid diagnostic testing
of neurologic improvement is appropriate in these cases before resource-intensive modalities that allow for
measures intended to provide long-term or permanent circulatory support are employed. accurate early neurologic
triage after cardiac arrest
u Patients with severe brain injury should preferentially be considered for experimental
include assessment of the
neuroprotective protocols.
total ischemic time, the
u Seizures after resuscitation are more common in patients with severe brain injury, but no-flow and low-flow
they are also an important cause of secondary brain injury. Because some patients with intervals, neurologic
seizures after cardiac arrest do not have a severe injury, all hypoxic-ischemic examination, CT, EEG
encephalopathy–related seizures should be suppressed unless a determination of background, continuity and
futility using other metrics has been made or the discontinuation of life support measures reactivity at different time
is intended. points after resuscitation,
and the bispectral index and
u All patients who are comatose after cardiac arrest should be treated in a maximally
suppression ratio.
neuroprotective milieu, as described in the following section.
● Patients whose risk
assessment suggests mild or
STANDARD NEUROPROTECTIVE MEASURES FOLLOWING CARDIAC ARREST
moderate brain injury should
The vulnerability of patients resuscitated from cardiac arrest is greatest early on, be considered for initiation
during a chaotic time when neuroprotective measures are difficult to provide.36 of aggressive strategies for
This section focuses on three critical areas of concern in the postresuscitation circulatory support, as
they are likely to benefit,
brain: blood flow, temperature, and metabolic stress. Because management of
up to and including
these elements is critical to neuroprotection, neurologists have the greatest extracorporeal membrane
opportunity to make a positive impact on care if they are involved early after oxygenation or surgical
resuscitation and function as advocates for neuroprotective measures. revascularization, if
necessary, and should
undergo an accelerated
Threat to Cerebral Blood Flow diagnostic workup to
Adequate cerebral blood flow after resuscitation is critical to prevent the identify the cause of the
transition of ischemic neurons to infarction but is compromised by a series of arrest with the intention of
factors that conspire to create a “perfect storm” scenario in the brain: maintaining adequate
cerebral blood flow and
preventing rearrest.
u Microcirculatory failure is characterized on electron microscopy by endothelial flaps,
platelet plugs, and sludging of erythrocytes37 and in laboratory analyses by high plasma ● All patients who are
levels of circulating syndecans and other markers of endothelial injury.38 Blood flow comatose after cardiac
through these threatened vascular beds must be preserved to prevent capillary dropout arrest should be treated in a
with tissue infarction and necrosis, but vascular resistance is markedly increased.39 maximally neuroprotective
milieu.
u Impaired or absent cerebral autoregulation in the large vessels of the brain reduces its
capacity to direct flow to regions of ischemia and to compensate for systemic
hypotension.40

u Irreversible cardiac damage may be caused by myocardial infarction and may be a

precipitant in up to half of out-of-hospital cardiac arrests.41
u Reversible cardiac injury or “stunning” may occur, caused by ischemia, epinephrine
administration, defibrillations, cardiopulmonary resuscitation, and acidosis.42
u A global vasodilatory state characterized by high levels of inflammatory cytokines,
activated neutrophils, and other inflammatory mediators begins during the
early reperfusion period, peaks first during the initial 4 to 12 hours after
resuscitation, and then peaks again during rewarming from targeted temperature
management.43
TABLE 7-2 Modalities for Very Early Assessment of Neurologic Risk After Cardiac
Arresta,b,c
Data
Modality Neurologic Riskb Confounders Utility Quality
No-flow interval28,d <3 Minutes: low Unwitnessed arrest, Moderate Low

diabetes mellitus (shorter
3–8 Minutes: low-medium
no-flow intervals may still
>8 Minutes: medium-high result in severe injury)
Total ischemic <25 Minutes: low-medium Unwitnessed arrest, Low Low

time (no-flow quality of
>25 Minutes: medium-high
plus low-flow)29 cardiopulmonary
resuscitation
Motor examination after Glasgow Coma Scale (GCS) Sedation, neuromuscular Moderate Low
resuscitation30 motor subscore 4+: low blockade, seizures
Total GCS score ≥5:
low-medium
GCS score 3–4: medium-high
FOUR score brainstem 0–1 (pupil and corneal reflex Atropine, neuromuscular Low Low
subscore at admission23 absent): high blockade
CT findings at the time of Gray/white matter Scan quality and Moderate Low
intensive care unit admission31 differentiation effacede: interpretation
medium-high
Gray/white matter
differentiation severely
effacede: very high
Subarachnoid hemorrhage:
obtain urgent CT angiogram
and neurosurgery consultation
EEG background at 4–18 hours Reactive: low-medium Sedation, seizures and High Moderate
post–return of spontaneous epileptiform activity,
Continuous: low-medium
circulation32–34 timing, technique,
Nonreactive: high interobserver variability in
interpretation
Suppression-burst: high
Isoelectric: very high
Low-voltage: unreliable
1738 DECEMBER 2018

u Hyperventilation decreases cerebral blood flow by the effects of pH on cerebrovascular
smooth muscle and is independently associated with worse outcomes after cardiac
arrest.44,45 Many patients are inadvertently hyperventilated in the hours after cardiac
arrest, and clinicians should actively monitor pH and end-tidal carbon dioxide, aiming to
ventilate to a normal or slightly low pH to maintain or augment cerebral blood flow.46
Because the cerebral microcirculation depends on maintenance of an adequate

perfusion pressure in the hours after resuscitation to prevent progression of
vulnerable cortical and subcortical regions from ischemia to necrosis, the
Data
Modality Neurologic Riskb Confounders Utility Quality
Processed EEG in a patient who Initial bispectral index score Sedation, seizures and High Moderate
received neuromuscular blockade after neuromuscular blockadef epileptiform activity,
at 4–18 hours post–return of 0: very high timing of assessment
spontaneous circulation13,14,35
Initial bispectral indexf
>20: low
11–20: medium
<10: high
Suppression ratio at 6 hours
after resuscitationg
<10%: low
10–40%: medium
40–80%: high
>80%: very high
36
Seizures Early severe myoclonus with Sedation, EEG Moderate Moderate
EEG correlate: high interpretation
Frequent periodic epileptiform
discharges or seizures arising
from a suppressed
background: high
CT = computed tomography; EEG = electroencephalography; FOUR = Full Outline of UnResponsiveness.

a
Modified with permission from Pitcher J, Seder DB.27 © 2018 Oxford University Press.
b
Neurologic risk refers to risk due to brain death, poor neurologic outcome, or withdrawal of life-sustaining therapies because of poor
neurologic prognosis.
c
Very early assessment means within 6 to 12 hours of return of spontaneous circulation.
d
No-flow interval is the time from arrest to onset of cardiopulmonary resuscitation.
e
Effacement of the gray-white junction can be quantified by comparing the ratio of Hounsfield units: the deep gray matter structures contrasted
to the adjacent white matter tracts. A ratio of 1.1–1.2 signifies effacement, and a ratio of <1.1 is considered severe effacement.
f
Bispectral index measured in a patient who is chemically paralyzed as soon as possible after return of spontaneous circulation (median 4.5 hours).
g
Suppression ratio is the percentage of each EEG epoch that is isoelectric or very low voltage. There are prognostic implications to the
suppression ratio measured 6 hours after resuscitation.

maintenance of adequate blood pressure and cardiac output is crucial. Although

the ideal postresuscitation blood pressure targets are unknown, hypotension is
associated with worse outcomes and must be avoided. Clinicians should insist on
maintenance of a minimum mean arterial pressure of 65 mm Hg, and many
centers opt for a higher target.47,48
Effects of Body Temperature on Outcome

Higher temperatures following resuscitation have repeatedly been shown to be
associated with increased brain injury in both human and laboratory models,
while hypothermia is neuroprotective. Targeted temperature management after
cardiac arrest is standard care, based on a series of trials conducted in Europe
and Australia that showed the superiority of mild therapeutic hypothermia to
no temperature management49 and the equivalence of mild hypothermia (33°C
[91.4°F]) to controlled normothermia (36°C [96.8°F]) in the 24 to 72 hours after
resuscitation50 in an unselected population of survivors of cardiac arrest unable
to follow verbal commands after resuscitation. Although controversy remains
about whether subgroups of patients may benefit from different temperature
targets, the control of body temperature after resuscitation is recommended in
both American47 and European51 guidelines.
Body temperature should be controlled as soon as possible following
resuscitation to a target temperature between 32°C (89.6°F) and 36°C (96.8°F) for
24 to 72 hours after resuscitation, with careful management of the complications
of therapy. These complications may include shivering, discomfort, electrolyte
disturbances, decreased cardiac contractility and vascular tone, bleeding, delayed
metabolism of medications, decreased gastrointestinal motility, immune suppression,
and interference with neurologic prognostication. As such, temperature
management and general postresuscitation cardiac arrest care should be performed
in experienced centers with protocols and adequate equipment—particularly
servo-regulated cooling devices that effectively regulate body temperature—to
perform temperature management with minimal adverse effects.52
Although it is likely that a subpopulation of survivors of cardiac arrest may
benefit from a temperature target of 32°C (89.6°F) to 34°C (93.2°F), that
subpopulation has not been clearly defined. Conversely, some data suggest the
superiority of a 35°C (95°F) to 36°C (96.8°F) target for patients with severe
shock,53 and the higher target should also be considered in patients with active
bleeding or life-threatening infections. Rewarming should always be performed
slowly to minimize adverse consequences of the inflammatory and vasodilatory
effects of warming.
Reducing Metabolic Stress to Favor Neurologic Recovery

Reperfusion injury in the brain is potentiated by hyperglycemia and hyperoxia,
both of which augment oxidative stress in injured tissues and are independently
associated with worse outcomes.54,55 Hyperglycemia is common, even in those
who are not diabetic, and is caused by an extreme stress response, with both
endogenous and exogenous catecholamines causing insulin resistance at the
tissue level. Because of the strong independent association of hyperglycemia
and poor outcome, plasma glucose levels should be normalized by insulin
administration.
Blood hyperoxia leads to increased free oxygen radical formation in
damaged tissues and worsens neurologic injuries and outcomes in both animal
1740 DECEMBER 2018

and human models.56–58 However, hypoxemia also worsens ischemia and KEY POINTS
outcomes. Although the exact thresholds for injury are unknown, a “safe range”
● Because the cerebral
of PaO2 of 80 mm Hg to 300 mm Hg is reasonable based on retrospective microcirculation depends on
studies.56,59 Maintenance of an oxyhemoglobin saturation of 94% to 99% after maintenance of an adequate
resuscitation usually achieves this PaO2 level, and a starting FIO2 (fraction of perfusion pressure in the
inspired oxygen) of 0.5 is recommended to prevent hyperoxia in the early period hours after resuscitation to
prevent progression of
after resuscitation before arterial blood gas analysis has been performed.
vulnerable cortical and
It is a primary function of the neurologist to advocate for the rapid restoration subcortical regions from
of a neuroprotective milieu after cardiac arrest. This includes the maintenance of ischemia to necrosis, the
adequate cerebral blood flow using high-normal blood pressure targets and maintenance of adequate
blood pressure and cardiac
avoidance of hyperventilation, the maintenance of low or normal body
output is crucial.
temperature for 24 to 72 hours after resuscitation, and the normalization of
oxygenation and serum glucose levels. ● Hypotension after
resuscitation is associated
with worse outcomes and
INTERPLAY OF SEIZURES, MYOCLONUS, AND NEUROLOGIC must be avoided.
PROGNOSTICATION AFTER RESUSCITATION
Hypoxic-ischemic encephalopathy–related seizures (sometimes called postanoxic ● Body temperature should
seizures) have long been associated with poor outcome after cardiac arrest be controlled as soon as
possible following
(CASE 7-2). Management of hypoxic-ischemic encephalopathy–related seizures resuscitation to a target
may be challenging and confusing: temperature between 32°C
(89.6°F) and 36°C (96.8°F)
u Hypoxic-ischemic encephalopathy–related seizures are both a marker of brain injury for 24 to 72 hours after
severity and a cause of ongoing secondary injury, yet some patients with seizures after resuscitation, with careful
cardiac arrest can recover and do well with aggressive seizure control. Unlike a management of the
suppressed or discontinuous EEG background, intermittent epileptiform discharges do complications of therapy.
not correlate with biomarkers of brain injury and are probably not an indicator of the
severity of brain injury after cardiac arrest.5 ● Data suggest the
superiority of a 35°C (95°F)
u Myoclonus is common and not always associated with seizure activity.2 to 36°C (96.8°F) target for
u Some hypoxic-ischemic encephalopathy–related seizures are nonconvulsive, requiring patients with severe shock,
continuous EEG monitoring to identify and treat. and the higher target should
also be considered in
u Because of their neurotoxicity, frequent (>1 Hz) periodic epileptiform discharges are patients with active
often considered to warrant a suppressive approach equivalent to seizures in this bleeding or life-threatening
population.7,60 infections.
u Hypoxic-ischemic encephalopathy–related seizures are difficult to suppress.
● Rewarming after targeted
u Treatment of hypoxic-ischemic encephalopathy–related seizures may interfere with temperature management
the assessment of prognosis. should always be performed
slowly to minimize adverse
Because of the overlap and intersection of these issues as well as the consequences of the
controversies involved, the following paradigm for managing seizures and inflammatory and vasodilatory
prognostication is suggested: effects of warming.
● Because of the strong

u Patients who are comatose after cardiac arrest should be monitored with continuous
independent association of
EEG, at least through the process of rewarming to normal body temperature and the
hyperglycemia and poor
resolution or waning of epileptiform activity.
outcome after cardiac
u Persistent hypoxic-ischemic encephalopathy–related seizures should be treated as arrest, plasma glucose
any other form of status epilepticus to prevent ongoing brain injury. Treatment of levels should be normalized
hypoxic-ischemic encephalopathy–related seizures that arise from a continuous and by insulin administration.
reactive EEG background may result in a good outcome.
u Because of the need to suppress hypoxic-ischemic encephalopathy–related seizures,
prognostication must sometimes proceed without the neurologic examination or use
of EEG modalities or must be done on a very delayed basis, requiring the use of biomarkers
and MRI.

u Because the etiology and significance of myoclonus remain poorly understood, its value
as a prognosticator is poor,61 although two recent studies suggest that myoclonus arising
from a continuous EEG background, with multifocal myoclonic movements and
polymorphic vertex discharges, may have a better prognosis than uniform discharges
arising from a burst-suppression or low-voltage background; the data on subcortical
myoclonus are conflicting.2,4,62 The physical examination finding of myoclonus should
be treated symptomatically, and electrographic seizures related to myoclonus should be
suppressed as with any other seizures.
u In patients who are persistently comatose, neuron-specific enolase levels should be
sent to a reliable laboratory at 24, 48, and 72 hours after resuscitation,63 and MRI should
be performed in patients who remain comatose at 72 to 120 hours after resuscitation.
CASE 7-2 A 54-year-old man was brought to the emergency department after
he collapsed at a park. Following his collapse, he was pulseless and
apneic, and bystanders initiated cardiopulmonary resuscitation. When
emergency medical services arrived, his initial heart rhythm was
ventricular fibrillation. Cardiopulmonary resuscitation and advanced
cardiac life support were continued for 24 minutes before the
return of spontaneous circulation. On arrival in the emergency
department, his ECG showed ST-segment elevation myocardial
infarction. Following guidelines,47,51 he underwent urgent cardiac
catheterization and stenting of a critical left anterior descending artery
occlusion before neurologic assessment and was admitted to the
intensive care unit with a Glasgow Coma Scale score of 5 (Glasgow Coma
Scale motor subscore of 3).
Targeted temperature management at 33°C (91.4°F) was initiated.
Because of severe whole-body jerking that started about 12 hours
post–return of spontaneous circulation, worse after stimulation and
accompanied by epileptiform activity, hypothermia was prolonged to
60 hours. EEG initially showed generalized periodic epileptiform
discharges and then left-dominant lateralized periodic discharges
evolving from a slowed background. EEG also showed periods of
periodic lateralized epileptiform discharges and generalized periodic
epileptiform discharges greater than 1 Hz. IV levetiracetam, valproate,
and midazolam infusions were administered. His blood pressure,
blood glucose level, arterial PaO2, and systemic pH were rapidly
normalized.
By protocol, neuron-specific enolase levels were drawn at 24, 48, and
72 hours post–return of spontaneous circulation. The neuron-specific
enolase peak level was 35 ng/mL at 48 hours post–return of spontaneous
circulation, and a brain MRI performed 82 hours post–return of
spontaneous circulation showed mild cortical hyperintensity on
diffusion-weighted imaging, verified on apparent diffusion coefficient
imaging and most prominent in the occipital and temporal lobes
bilaterally (FIGURE 7-1).
These prognostic findings were indeterminate, and a multidisciplinary
family meeting emphasizing principles of shared decision making was
held. Based on the prognostic uncertainty, the family’s assessment of his
1742 DECEMBER 2018

u Patients with hypoxic-ischemic encephalopathy–related seizures who demonstrate very
high levels of neuron-specific enolase15 and evidence of extensive brain injury by MRI17
are extremely unlikely to awaken or make a meaningful neurologic recovery.
u Patients with hypoxic-ischemic encephalopathy–related seizures who do not have
very high neuron-specific enolase levels and extensive brain injury on MRI may make
a good recovery, and ongoing supportive therapy is appropriate unless circulatory
collapse or multiple organ failure preclude ongoing aggressive care.
u Although the absence of bilateral N20 responses on somatosensory evoked potentials
is highly predictive of poor neurologic outcome even among patients receiving sedation,
the use of this electrophysiologic modality in patients receiving suppressive levels of
benzodiazepines, barbiturates, and other general anesthetic agents should be discouraged.
personality, and his prior comments about end-of-life care, his family
requested that aggressive care be continued for up to 21 days post–return
of spontaneous circulation before consideration of withdrawal of life-
sustaining measures, with tracheostomy and surgical feeding tube
placement if needed. Midazolam was weaned without further seizures.
On hospital day 12, he followed commands and was extubated. On
hospital day 20, he was interactive, ambulatory with assistance, taking a
full diet, and participating in therapy activities. Physical medicine and
rehabilitation at an acute inpatient rehabilitation center found him to be
90% independent in activities of daily living at 60 days postarrest, and, on
an unscheduled follow-up telephone call, he asked clinicians about
complex details of his care, suggesting intact memory.
FIGURE 7-1
Imaging of patient in CASE 7-2. Brain MRI performed 82 hours post–return of spontaneous
circulation showed mild cortical hyperintensity, most prominent in the occipital and
temporal lobes bilaterally, on diffusion-weighted imaging (A) and was verified on apparent
diffusion coefficient imaging (B).
Reprinted with permission from Pitcher J, Seder DB.27 © 2018 Oxford University Press.
This case shows the complexity of care and prognostication of comatose COMMENT
survivors of cardiac arrest. The relationship between seizures, myoclonus,
and neurologic prognostication is complex.

PROGNOSTICATION, END-OF-LIFE CARE, AND SHARED

DECISION MAKING
Neuroprognostication is complex and is plagued by potential confounders and
the risk of inaccuracy. Patients, families, and the treatment team depend on a
high-quality prognostication process that does not decrease the chance of survival,
prolong a painful death, provide false hope to vulnerable family members, or
result in many neurologically devastated survivors. When performed too soon
or without adequate diagnostic testing, prognostication can cost some patients
their lives.11 Yet unnecessary delays and prolongation of painful, expensive,
and futile life support measures worsen the patient and family experience at
great expense to the health care system and serve no one. Because the stakes are
so high, expertise in neurologic prognostication after resuscitation from cardiac
arrest is a great service neurologists can provide to their patients and the health
care community.
CASE 7-3 A 45-year-old man with type 2 diabetes mellitus and a family history of
premature coronary artery disease reported chest pain and then
collapsed, apneic and pulseless. Bystanders initiated cardiopulmonary
resuscitation after a 3- to 4-minute delay, and on the arrival of emergency
medical services after 12 minutes, he was in ventricular fibrillation.
After a total of 42 minutes of cardiopulmonary resuscitation; line
placement; multiple defibrillations; intubation; and administration of
epinephrine, amiodarone, sodium bicarbonate, and magnesium, a
perfusing rhythm was reestablished, and he was transferred by ground to
a local hospital, where saline, norepinephrine, heparin, and aspirin were
administered. Chest x-ray confirmed appropriate placement of the
endotracheal tube; ECG showed anterior and septal Q waves and diffuse
ST-segment depressions.
He was transferred by helicopter to a tertiary cardiac intensive care
unit, where on arrival at 6 hours post–return of spontaneous circulation,
he received a maximal dose of norepinephrine with a blood pressure of
82/52 mm Hg, had no urine output and serum lactate of 9.2 mmol/L,
and exhibited mottled extremities. ECG was unchanged, and
echocardiography showed hypokinesis of the anterior wall and profoundly
decreased left ventricular ejection fraction. Neurology was paged
emergently to bedside to confer with the cardiac intensive care team in
consideration of mechanical circulatory support, urgent coronary
angiography, and possible revascularization.
On neurologic evaluation, absent any sedating medications, the
patient had a Glasgow Coma Scale score of 3T and a Full Outline of
UnResponsiveness (FOUR) score of E0, M0, B2, R0, with weakly reactive
3-mm pupils, no corneal reflexes, and no cough to deep suctioning. (For
more information on the FOUR score, refer to the article “Coma and Brain
Death” by Alejandro A. Rabinstein, MD, FAAN,24 in this issue of Continuum.)
The EEG was severely suppressed. Based on these findings, he was
characterized as being at very high risk of poor neurologic outcome, with
1744 DECEMBER 2018

Multimodal prognostication is standard after cardiac arrest (CASE 7-3).
Prognostication is roughly categorized as based on the physical examination,
EEG, serum biomarkers, and neuroimaging. These tools and their most appropriate
timing have been described in recent reviews and European guidelines,12,65 and
neurologists should be familiar with not only the data on prognostic accuracy,
but also the most effective and practical way to employ these tests.
The neurologic prognosis is further influenced by the circumstances of the
arrest and the patient’s age, code status,66 comorbid medical and neurologic
conditions, premorbid functional status, and other organ system failures. The
effect of these factors, however, is difficult to quantitate; in clinical practice, they
add a qualitative element to the discussion and inform shared decision making
rather than increasing the quantitative accuracy of the prognostic testing.
Although some patients will quickly be determined to be neurologically
devastated after resuscitation and early therapy and others awaken shortly
recommendations to delay mechanical circulatory support and coronary

angiography until neurologic recovery was demonstrated; normalize blood
pressure, glucose, oxygen, and pH; control temperature at 35°C (95°F) to
36°C (96.8°F) for at least 72 hours; monitor with EEG for seizures; draw
neuron-specific enolase levels at 24, 48, and 72 hours; and obtain MRI of
the brain at 72 to 120 hours post–return of spontaneous circulation.
Over the next 48 hours, hemodynamics, oxygenation, and systemic organ
perfusion improved. His organ function began to normalize, and the EEG
pattern evolved to burst suppression. Occasional myoclonic jerking was
noted and was treated with intermittent boluses of lorazepam. EEG
remained discontinuous and nonreactive, he had no recovery of corneal
reflexes, and, although pupils remained reactive and a very weak cough
reflex persisted, he remained flaccid and without autonomic response to
pain stimulus. The peak neuron-specific enolase level was 184 ng/mL, and
brain MRI showed diffuse and severe bilateral injury involving all regions of
cortex, the thalami and putamen, and the cerebellar hemispheres. On
hospital day 7, a family meeting was held to discuss the goals of care, in
which the evidence of a very severe brain injury was discussed. A
multidisciplinary team of clinicians uniformly felt he had no real possibility
of a functional neurologic recovery. The family decided to withdraw
64
life-sustaining measures.
This case illustrates a complex clinical scenario in which an early neurologic COMMENT
risk assessment of a patient with severe brain injury following cardiac
arrest helped determine the limits of early treatment, but multimodal
prognostication was delayed until reasonable certainty existed to inform
shared decision making about the goals of care. This case involved a
combination of prognostic tools performed at the optimal timing after
resuscitation and in the absence of important confounders.

after sedation is discontinued, many patients will nonetheless have an

uncertain prognosis. This uncertainty is an inherent part of medicine. When
the prognosis is uncertain, the principles of shared decision making are
employed to help the family and treatment team determine a patient- and
family-centered pathway of care that offers comfort, support, hope, and
direction.20,67 Palliative care specialists are often involved, and, although many
families choose to transition to comfort-based care without prolongation of
life-sustaining measures, others will opt for tracheostomy, feeding tube
placement, and a prolonged period of observation and therapy. This is
consistent with wide variations in end-of-life cultural values and attitudes
within the population and therefore should not be a source of frustration
to clinicians.
Confounding Effect of Sedation on Prognostic Testing

A 70-kg (154-lb) man undergoing targeted temperature management under
the (commonly employed) protocol used in the first landmark trial of
hypothermia for cardiac arrest would have received a total of 280 mg midazolam,
4480 mcg fentanyl, and 224 mg pancuronium (an intermediate-acting
nondepolarizing neuromuscular blockade agent) during the initial 32 hours of
care.49 Temperature management is frequently performed for 72 hours or longer
in patients post–cardiac arrest, and benzodiazepine infusions are common.
These massive doses of analgesia and sedation and muscle relaxants are poorly
metabolized after an arrest because of temperature management, organ
dysfunction, distribution into fatty tissues, and drug interactions, and they
profoundly affect physical examination and EEG-based prognostic testing.8
A patient who receives a continuous infusion of benzodiazepines for 2 to 3 days
may require a week (or even several weeks when renal failure is present)
for medications to be eliminated to provide examination conditions that allow
for accurate prognostication by neurologic examination and EEG. Because of
this, a low-intensity sedation regimen is preferred during targeted
temperature management (as discussed below); when patients have received
high-dose sedative infusions, clinicians must rely on serum biomarkers,
neuroimaging, and delayed functional testing to support their
prognostic evaluation.
Pragmatic Approach to Prognostication After Cardiac Arrest

A standardized approach to prognostication is recommended.
MINIMIZE THE CONFOUNDING OF ANALGESIA AND SEDATION DURING AND AFTER

TEMPERATURE MANAGEMENT. During targeted temperature management, a
short-acting regimen of fentanyl or remifentanil infusion and low-dose
propofol, titrated for comfort, blood pressure, and control of shivering,
should be employed. Patients with unstable hemodynamics commonly do not
tolerate propofol, in which case the opioid infusion should be increased and
bolus-dose IV midazolam administered as needed for agitation or anxiety.
When this regimen is used and a continuous infusion of benzodiazepines
avoided, less buildup of sedation occurs, patients awaken earlier, and
prognostication is markedly less confounded.68 This regimen is typically not
adequate when seizure suppression is required, in which case benzodiazepine,
barbiturate, and propofol infusions may be required and prognostication must
rely on non-EEG and nonexamination-based testing.
1746 DECEMBER 2018

PROTOCOLIZE APPROPRIATE PROGNOSTIC TESTING AT THE OPTIMAL TIME KEY POINTS
POINTS. Certain elements of prognostication should be protocolized, such as an
EEG background, suppression, and reactivity assessment at 6 to 12 hours ● Hypoxic-ischemic
encephalopathy–related
post–return of spontaneous circulation5,13,14,65; serum neuron-specific enolase seizures are both a marker
levels at 24, 48, and 72 hours after resuscitation in patients who remain of brain injury severity and
comatose; and MRI in patients who are comatose at 72 to 120 hours after a cause of ongoing
resuscitation.12,15–17,65 If this testing is not done, critical information will not secondary injury.
be available to inform family decisions at 5 to 14 days after resuscitation, and
● Patients who are
provider-family discussions around the goals of care must rely on less accurate comatose after cardiac
qualitative assessments of prognosis alone. arrest should be monitored
with continuous EEG, at
least through the process of
BE PATIENT. Unless the family feels strongly that the patient would not rewarming to normal body
want even temporary life-sustaining measures, wait for sedation to clear and temperature and the
prognostic test results to become available. Many experts believe formal resolution or waning of
assessment of prognosis should be delayed for at least 72 hours after rewarming epileptiform activity.
from targeted temperature management and potentially longer.51 Informal
● Patients with hypoxic-
conjecture about prognosis before the availability of appropriate testing is ischemic encephalopathy–
unprofessional and unhelpful and may negatively bias care. Neurologists should related seizures who do not
be familiar with factors associated with delayed awakening, including renal have very high neuron-
insufficiency, older age, and postresuscitation shock.69 specific enolase levels and
extensive brain injury on MRI
may make a good recovery,
CONSIDER THE AGREEMENT OR DISAGREEMENT OF DIAGNOSTIC TESTS. When and ongoing supportive
prognostic tests that were appropriately performed and interpreted line up, or therapy is appropriate
unless circulatory collapse
agree on either a favorable or unfavorable prognosis, neurologists can feel
or multiple organ failure
confident in guiding families through the difficult process of decision making preclude ongoing
regarding continuation of life support in patients who have not awoken aggressive care.
after resuscitation. Conversely, when these tests seem to disagree, caution
is warranted. Almost every variation of prognostic failure after cardiac arrest ● When performed too
soon or without adequate
has been described, including patients who nonetheless made a good recovery diagnostic testing,
who had very high neuron-specific enolase levels,70 absent N20 signals on prognostication after
somatosensory evoked potentials,71 early severe myoclonic status resuscitation from cardiac
epilepticus,2–4,72 and abnormal MRI examination; however, not infrequently a arrest can cost some
patients their lives.
patient who clinicians feel “should” do well fails to awaken. Such errors are
minimized when the conservative approach described above is taken, and ● Multimodal
frank discrepancies in testing modalities appropriately result in delivery of an prognostication is standard
“uncertain” prognosis to the patient’s surrogates. after cardiac arrest.
● When the prognosis after

Shared Decision Making cardiac arrest is uncertain,
Shared decision making is described as “a collaborative process that allows the principles of shared
patients, or their surrogates, and clinicians to make health care decisions decision making are
together, taking into account the best scientific evidence available, as well as the employed to help the family
and treatment team
patient’s values, goals, and preferences.”20 When applied to postresuscitation determine a patient- and
cardiac arrest care, shared decision making means that when prognostication has family-centered pathway of
been performed, a multidisciplinary meeting that includes the patient’s care that offers comfort,
surrogates and health care stakeholders should be held to exchange information, support, hope, and
direction.
deliberate, and make a decision about the goals of care. This process should
include acknowledgment of any uncertainties that exist, and the health care team
should help the patient’s surrogates come to a decision that is consistent with the
patient’s perceived wishes. Therefore, the best treatment pathway is not solely
the result of neuroprognostic test results but also of the interplay between those

test results, other medical and surgical concerns, residual prognostic uncertainty,
the patient’s values and preferences, and the skill of the clinicians at resolving
these overlapping and sometimes conflicting concerns.
THE NEED FOR NEUROCOGNITIVE EVALUATION AND COUNSELING

OF SURVIVORS
When assessed before hospital discharge, more than half of survivors of cardiac
arrest have measurable neurocognitive dysfunction, and 20% to 50% have
persistent deficits of memory, learning, or executive function even after
3 months (CASE 7-4).21,73 Although symptoms often improve over time, these
deficits are upsetting to patients and their families who need predischarge
counseling about the potential for cognitive deficits, the likelihood of recovery,
and the availability of cognitive therapy in the community to address
persistent deficits.
CONCLUSION
Hypoxic-ischemic encephalopathy is a treatable condition, and the overall
outcomes of patients who survive cardiopulmonary resuscitation and are
subsequently hospitalized have steadily improved over the past few decades.
Because of the high complexity, modern postresuscitation cardiac arrest care is
based on a team approach, with neurologists playing a fundamental and crucial
role in early triage as advocates for neuroprotective measures, in identifying
and treating seizures, in prognosticating if the patient fails to awaken, and in
evaluating survivors for neurocognitive dysfunction and counseling them. A
standardized approach to cardiac arrest care is recommended to improve the
quality of care.
CASE 7-4 A 19-year-old woman was brought to the emergency department after a
sudden cardiopulmonary arrest witnessed by her mother. Her mother
(a registered nurse) and emergency medical services providers had
resuscitated her after 24 minutes of cardiopulmonary resuscitation,
which had been started immediately after onset. She was admitted to the
intensive care unit. After targeted temperature management and routine
care, she awoke, underwent cardiac catheterization and an
electrophysiologic study, and was discharged home in good condition.
Upon returning to college, she found she could not concentrate in
class or when studying, so she dropped out. She was referred to a
neuropsychologist, who performed a cognitive evaluation and then
worked with her for 3 months to develop alternative learning strategies to
compensate for her (partially resolved) deficits. The following semester,
she returned to college and later graduated with honors.
COMMENT This case illustrates the importance of appropriate neuropsychological

testing and management of cognitive deficits in patients who survive a
cardiac arrest and are discharged home.
1748 DECEMBER 2018

KEY POINTS
REFERENCES
● Many experts believe
1 Nielsen N, Wetterslev J, Cronberg T, et al. Targeted 13 Seder DB, Fraser G, Robbins T, et al. The formal assessment of
temperature management at 33°C versus 36°C bispectral index and suppression ratio are very prognosis after cardiac
after cardiac arrest. N Engl J Med 2013;369(23): early predictors of neurological outcome during arrest should be delayed for
2197–2206. doi:10.1056/NEJMoa1310519. therapeutic hypothermia after cardiac arrest.
at least 72 hours after
Intensive Care Med 2010;36(2):281–88.
2 Seder DB, Sunde K, Rubertsson S, et al. rewarming from targeted
doi:10.1007/s00134-009-1691-1.
Neurologic outcomes and postresuscitation care temperature management
of patients with myoclonus following cardiac 14 Burjek NE, Wagner CE, Hollenbeck RD, et al. Early and potentially longer.
arrest. Crit Care Med 2015;43(5):965–972. bispectral index and sedation requirements
doi:10.1097/CCM.0000000000000880. during therapeutic hypothermia predict
● Neurologists should be
neurologic recovery following cardiac arrest. Crit
3 Aicua Rapun I, Novy J, Solari D, et al. Early
Care Med 2014;42(5):1204–1212. doi:10.1097/
familiar with factors
Lance-Adams syndrome after cardiac arrest: associated with delayed
CCM.0000000000000126.
prevalence, time to return to awareness, and awakening after cardiac
outcome in a large cohort. Resuscitation 2017;115: 15 Stammet P, Collignon O, Hassager C, et al. arrest, including renal
169–172. doi:10.1016/j.resuscitation.2017.03.020. Neuron-specific enolase as a predictor of
insufficiency, older age, and
death or poor neurological outcome after
4 Elmer J, Rittenberger JC, Faro J, et al. Clinically
out-of-hospital cardiac arrest and targeted
postresuscitation shock.
distinct electroencephalographic phenotypes of
temperature management at 33°C and 36°C.
early myoclonus after cardiac arrest. Ann Neurol ● Shared decision making is
J Am Coll Cardiol 2015;65(19):2104–2114.
2016;80(2):175–184. doi:10.1002/ana.24697.
doi:10.1016/j.jacc.2015.03.538. considered the standard
5 Rossetti AO, Carrera E, Oddo M. Early EEG approach to conversations
16 Greer DM, Wu O. Neuroimaging in cardiac arrest
correlates of neuronal injury after brain anoxia. about end-of-life care.
prognostication. Semin Neurol 2017;37(1):66–74.
Neurology 2012;78(11):796–802. doi:10.1212/ When prognostication has
doi:10.1055/s-0036-1594253.
WNL.0b013e318249f6bb. been performed, a
17 Wijman CA, Mlynash M, Caulfield AF, et al.
6 Ko SB, Ortega-Gutierrez S, Choi HA, et al. Status multidisciplinary meeting
Prognostic value of brain diffusion-weighted
epilepticus-induced hyperemia and brain tissue that includes the patient’s
imaging after cardiac arrest. Ann Neurol 2009;
hypoxia after cardiac arrest. Arch Neurol 2011; surrogates and health care
65(4):394–402. doi:10.1002/ana.21632.
68(10):1323–1326. doi:10.1001/archneurol.2011.240. stakeholders should be held
18 Sandroni C, Cavallaro F, Callaway CW, et al. to exchange information,
7 Reynolds AS, Claassen J. Treatment of seizures
Predictors of poor neurological outcome in
and postanoxic status epilepticus. Semin Neurol deliberate, and make a
adult comatose survivors of cardiac arrest: a
2017;37(1):33–39. doi:10.1055/s-0036-1593862. decision about the goals
systematic review and meta-analysis. Part 2:
8 Samaniego EA, Mlynash M, Caulfield AF, et al. patients treated with therapeutic hypothermia. of care.
Sedation confounds outcome prediction in Resuscitation 2013;84(10):1324–1338. doi:10.1016/j.
cardiac arrest survivors treated with resuscitation.2013.06.020. ● More than half of
hypothermia. Neurocrit Care 2011;15(1):113–119. survivors of cardiac arrest
19 Bauer TM, Ritz R, Haberthür C, et al. Prolonged
doi:10.1007/s12028-010-9412-8.
sedation due to accumulation of conjugated
have measurable
9 Šunjić KM, Webb AC, Šunjić I, et al. metabolites of midazolam. Lancet 1995;346(8968): neurocognitive dysfunction,
Pharmacokinetic and other considerations for 145–147. doi:10.1016/S0140-6736(95)91209-6. and 20% to 50% have
drug therapy during targeted temperature persistent deficits of
20 Kon AA, Davidson JE, Morrison W, et al.
management. Crit Care Med 2015;43(10): memory, learning, or
Shared decision-making in intensive care units.
2228–2238. doi:10.1097/CCM.
Executive summary of the American College of
executive function even
0000000000001223. after 3 months.
Critical Care Medicine and American Thoracic
10 Greer DM, Rosenthal ES, Wu O. Society policy statement. Am J Respir Crit Care
Neuroprognostication of hypoxic-ischaemic Med 2016;193(12):1334–1336. doi:10.1164/
coma in the therapeutic hypothermia era. rccm.201602-0269ED.
Nat Rev Neurol 2014;10(4):190–203. doi:10.1038/
21 Lilja G, Nielsen N, Friberg H, et al. Cognitive
nrneurol.2014.36.
function in survivors of out-of-hospital cardiac
11 Elmer J, Torres C, Aufderheide TP, et al. arrest after target temperature management
Association of early withdrawal of life-sustaining at 33°C versus 36°C. Circulation 2015;131(15):
therapy for perceived neurological prognosis 1340–1349. doi:10.1161/CIRCULATIONAHA.114.014414.
with mortality after cardiac arrest. Resuscitation
22 Fugate JE, Moore SA, Knopman DS, et al. Cognitive
2016;102:127–135. doi:10.1016/j.resuscitation.
outcomes of patients undergoing therapeutic
2016.01.016.
hypothermia after cardiac arrest. Neurology 2013;
12 Sandroni C, Cariou A, Cavallaro F, et al. 81(1):40–45. doi:10.1212/WNL.0b013e318297ee7e.
Prognostication in comatose survivors of cardiac
23 Youn CS, Callaway CW, Rittenberger JC, Post
arrest: an advisory statement from the European
Cardiac Arrest Service. Combination of initial
Resuscitation Council and the European Society of
neurologic examination and continuous EEG
Intensive Care Medicine. Intensive Care Med 2014;
to predict survival after cardiac arrest.
40(12):1816–1831. doi:10.1007/s00134-014-3470-x.
Resuscitation 2015;94:73–79. doi:10.1016/
j.resuscitation.2015.06.016.

24 Rabinstein AA. Coma and brain death. Continuum 36 Safar P, Behringer W, Böttiger BW, Sterz F.
(Minneap Minn) 2018;24(6 Neurocritical Care): Cerebral resuscitation potentials for cardiac
1708–1731. arrest. Crit Care Med 2002;30(4 suppl):
S140–S144. doi:10.1097/00003246-200204001-
25 Bougouin W, Aissaoui N, Combes A, et al.
00004.
Post-cardiac arrest shock treated with
veno-arterial extracorporeal membrane 37 Krep H, Böttiger BW, Bock C, et al. Time course
oxygenation: an observational study and of circulatory and metabolic recovery of
propensity-score analysis. Resuscitation 2017;110: cat brain after cardiac arrest assessed by
126–132. doi:10.1016/j.resuscitation.2016.11.005. perfusion- and diffusion-weighted imaging and
MR-spectroscopy. Resuscitation 2003;58(3):
26 Metter RB, Rittenberger JC, Guyette FX, Callaway
337–348. doi:10.1016/S0300-9572(03)00151-5.
CW. Association between a quantitative CT scan
measure of brain edema and outcome after 38 Bro-Jeppesen J, Johansson PI, Hassager C, et al.
cardiac arrest. Resuscitation 2011;82(9):1180–1185. Endothelial activation/injury and associations
doi:10.1016/j.resuscitation.2011.04.001. with severity of post-cardiac arrest syndrome
and mortality after out-of-hospital cardiac
27 Pitcher J, Seder DB. Cardiac arrest. In: Reich DL,
arrest. Resuscitation 2016;107:71–79. doi:10.1016/
Mayer SA, Uysal S, editors. Neuroprotection in
critical care and perioperative medicine. 1st ed.
New York, NY: Oxford University Press, 2018:160–161. 39 Leonov Y, Sterz F, Safar P, et al. Hypertension
with hemodilution prevents multifocal cerebral
28 Adrie C, Cariou A, Mourvillier B, et al. Predicting
hypoperfusion after cardiac arrest in dogs.
survival with good neurological recovery at
Stroke 1992;23(1):45–53. doi:10.1161/01.STR.23.1.45.
hospital admission after successful resuscitation
of out-of-hospital cardiac arrest: the OHCA 40 Sundgreen C, Larsen FS, Herzog TM, et al.
score. Eur Heart J 2006;27(23):2840–2845. Autoregulation of cerebral blood flow in patients
doi:10.1093/eurheartj/ehl335. resuscitated from cardiac arrest. Stroke 2001;
32(1):128–132. doi:10.1161/01.STR.32.1.128.
29 Oddo M, Schaller MD, Feihl F, et al. From
evidence to clinical practice: effective 41 Spaulding CM, Joly LM, Rosenberg A, et al.
implementation of therapeutic hypothermia to Immediate coronary angiography in survivors
improve patient outcome after cardiac arrest. of out-of-hospital cardiac arrest. N Engl J Med
Crit Care Med 2006;34(7):1865–1873. doi: 1997;336(23):1629–1633. doi:10.1056/
10.1097/01.CCM.0000221922.08878.49. NEJM199706053362302.
30 Natsukawa T, Sawano H, Natsukawa M, et al. 42 Laurent I, Monchi M, Chiche JD, et al. Reversible
At what level of unconsciousness is mild myocardial dysfunction in survivors of out-of-
therapeutic hypothermia indicated for hospital cardiac arrest. J Am Coll Cardiol 2002;
out-of-hospital cardiac arrest: a retrospective, 40(12):2110–2116. doi:10.1016/S0735-1097(02)02594-9.
historical cohort study. J Intensive Care
43 Bisschops LL, Hoedemaekers CW, Mollnes TE,
2015;3(1):38. doi:10.1186/s40560-015-0104-5.
van der Hoeven JG. Rewarming after
31 Kim SH, Choi SP, Park KN, et al. Early brain hypothermia after cardiac arrest shifts the
computed tomography findings are associated inflammatory balance. Crit Care Med 2012;40:
with outcome in patients treated with 1136–1142. doi:10.1097/CCM.0b013e3182377050.
therapeutic hypothermia after out-of-hospital
44 Bouzat P, Suys T, Sala N, Oddo M. Effect of
cardiac arrest. Scand J Trauma Resusc Emerg
moderate hyperventilation and induced
Med 2013;21:57. doi:10.1186/1757-7241-21-57.
hypertension on cerebral tissue oxygenation
32 Rossetti AO, Urbano LA, Delodder F, et al. after cardiac arrest and therapeutic
Prognostic value of continuous EEG monitoring hypothermia. Resuscitation 2013;84(11):1540–1545.
during therapeutic hypothermia after cardiac doi:10.1016/j.resuscitation.2013.05.014.
arrest. Crit Care 2010;14(5):R173. doi:10.1186/cc9276.
45 Schneider AG, Eastwood GM, Bellomo R, et al.
33 Cloostermans MC, van Meulen FB, Eertman CJ, Arterial carbon dioxide tension and outcome in
et al. Continuous electroencephalography patients admitted to the intensive care unit after
monitoring for early prediction of neurological cardiac arrest. Resuscitation 2013;84(7):927–934.
outcome in postanoxic patients after cardiac doi:10.1016/j.resuscitation.2013.02.014.
arrest: a prospective cohort study. Crit Care
46 Seder DB, Jagoda A, Riggs B. Emergency
Med 2012;40(10):2867–2875. doi:10.1097/
neurological life support: airway, ventilation, and
CCM.0b013e31825b94f0.
sedation. Neurocrit Care 2015;23(suppl 2):
34 Rossetti AO, Urbano LA, Delodder F, et al. S5–S22. doi:10.1007/s12028-015-0164-3.
Prognostic value of continuous EEG monitoring
47 Callaway CW, Donnino MW, Fink EL, et al. Part 8:
during therapeutic hypothermia after cardiac
post-cardiac arrest care: 2015 American Heart
arrest. Crit Care 2010;14(5):R173. doi:10.1186/cc9276.
Association guidelines update for
35 Seder DB, Dziodzio J, Smith KA, et al. Feasibility cardiopulmonary resuscitation and emergency
of bispectral index monitoring to guide early cardiovascular care. Circulation 2015;
post-resuscitation cardiac arrest triage. 132(18 suppl 2):S465–S482. doi:10.1161/
Resuscitation 2014;85(8):1030–1036. doi:10.1016/ CIR.0000000000000262.
1750 DECEMBER 2018

48 Trzeciak S, Jones AE, Kilgannon JH, et al. 59 Helmerhorst HJ, Roos-Blom MJ, van Westerloo
Significance of arterial hypotension after DJ, de Jonge E. Association between arterial
resuscitation from cardiac arrest. Crit Care hyperoxia and outcome in subsets of critical
Med 2009;37(11):2895–2903. doi:10.1097/ illness: a systematic review, meta-analysis, and
CCM.0b013e3181b01d8c. meta-regression of cohort studies. Crit Care
Med 2015;43(7):1508–1519. doi:10.1097/
49 Hypothermia after Cardiac Arrest Study Group.
CCM.0000000000000998.
Mild therapeutic hypothermia to improve the
neurologic outcome after cardiac arrest. N Engl J 60 Rittenberger JC, Popescu A, Brenner RP, et al.
Med 2002;346(8):549–556. doi:10.1056/ Frequency and timing of nonconvulsive status
NEJMoa012689. epilepticus in comatose post-cardiac
arrest subjects treated with hypothermia.
50 Nielsen N, Wetterslev J, Cronberg T, et al.
Targeted temperature management at 33°C
s12028-011-9565-0.
versus 36°C after cardiac arrest. N Engl J Med 2013;
369(23):2197–2206. doi:10.1056/NEJMoa1310519. 61 Sandroni C, Nolan JP. Neurologically favorable
outcome is still possible despite myoclonus in
51 Nolan JP, Soar J, Cariou A, et al. European
comatose survivors of cardiac arrest. Crit Care
Resuscitation Council and European Society
Med 2015;43(9):e396–e397. doi:10.1097/
of Intensive Care Medicine 2015 guidelines
CCM.0000000000001070.
for post-resuscitation care. Intensive
Care Med 2015;41(12):2039–2056. doi:10.1007/ 62 Dhakar MB, Sivaraju A, Maciel CB, et al. Electro-
s00134-015-4051-3. clinical characteristics and prognostic
significance of post anoxic myoclonus
52 Madden LK, Hill M, May TL, et al. The
[published online ahead of print June 28,
implementation of targeted temperature
2018]. Resuscitation. doi:10.1016/j.resuscitation.
management: an evidence-based guideline
2018.06.030.
from the Neurocritical Care Society. Neurocrit
Care 2017;27(3):468–487. doi:10.1007/s12028- 63 Mlynash M, Buckwalter MS, Okada A, et al.
017-0469-5. Serum neuron-specific enolase levels from the
same patients differ between laboratories:
53 Annborn M, Bro-Jeppesen J, Nielsen N, et al.
assessment of a prospective post-cardiac arrest
The association of targeted temperature
cohort. Neurocrit Care 2013;19(2):161–166.
management at 33 and 36 °C with outcome in
doi:10.1007/s12028-013-9867-5.
patients with moderate shock on admission
after out-of-hospital cardiac arrest: a post hoc 64 Reynolds JC, Rittenberger JC, Callaway CW
analysis of the Target Temperature Management Post Cardiac Arrest Service. Patterns of organ
trial. Intensive Care Med 2014;40(9):1210–1219. donation among resuscitated patients at a
doi:10.1007/s00134-014-3375-8. regional cardiac arrest center. Resuscitation
2014;85(2):248–252. doi:10.1016/j.resuscitation.
54 Kilgannon JH, Jones AE, Parrillo JE, et al.
2013.11.001.
Relationship between supranormal oxygen
tension and outcome after resuscitation from 65 Taccone F, Cronberg T, Friberg H, et al.
cardiac arrest. Circulation 2011;123(23):2717–2722. How to assess prognosis after cardiac arrest and
doi:10.1161/CIRCULATIONAHA.110.001016. therapeutic hypothermia. Crit Care 2014;18(1):202.
doi:10.1186/cc13696.
55 Nielsen N, Sunde K, Hovdenes J, et al. Adverse
events and their relation to mortality in out-of- 66 Seder DB, Patel N, McPherson J, et al. Geriatric
hospital cardiac arrest patients treated with experience following cardiac arrest at six
therapeutic hypothermia. Crit Care Med 2011;39(1): interventional cardiology centers in the United
57–64. doi:10.1097/CCM.0b013e3181fa4301. States 2006-2011: interplay of age, do-not-
resuscitate order, and outcomes. Crit Care Med
56 Pilcher J, Weatherall M, Shirtcliffe P, et al. The
2014;42(2):289–295. doi:10.1097/
effect of hyperoxia following cardiac arrest—a
CCM.0b013e3182a26ec6.
systematic review and meta-analysis of animal
trials. Resuscitation 2012;83(4):417–422. 67 Barry MJ, Edgman-Levitan S. Shared decision
doi:10.1016/j.resuscitation.2011.12.021. making—pinnacle of patient-centered care.
N Engl J Med 2012;366(9):780–781. doi:10.1056/
57 Vereczki V, Martin E, Rosenthal RE, et al.
NEJMp1109283.
Normoxic resuscitation after cardiac arrest
protects against hippocampal oxidative stress, 68 May TL, Seder DB, Fraser GL, et al. Moderate-
metabolic dysfunction, and neuronal death. dose sedation and analgesia during targeted
J Cereb Blood Flow Metab 2006;26(6):821–835. temperature management after
doi:10.1038/sj.jcbfm.9600234. cardiac arrest. Neurocrit Care 2015;22:105–111.
doi:10.1007/s12028-014-9998-3.
58 Kilgannon JH, Jones AE, Shapiro NI, et al.
Emergency Medicine Shock Research Network 69 Paul M, Bougouin W, Geri G, et al. Delayed
(EMShockNet) Investigators. Association between awakening after cardiac arrest: prevalence
arterial hyperoxia following resuscitation from and risk factors in the Parisian registry. Intensive
cardiac arrest and in-hospital mortality. JAMA Care Med 2016;42(7):1128–1136. doi:10.1007/
2010;303(21):2165–2171. doi:10.1001/jama. s00134-016-4349-9.
2010.707.

70 Hofer B, Dunzendorfer S, Beer R, Joannidis M. 72 Lucas JM, Cocchi MN, Salciccioli J, et al.
“Cardiac arrest—favorable functional outcome Neurologic recovery after therapeutic
despite high NSE levels and early brain hypothermia in patients with post-cardiac arrest
swelling”. Resuscitation 2017;116:e3. doi:10.1016/ myoclonus. Resuscitation 2012;83(2):265–269.
j.resuscitation.2017.04.031. doi:10.1016/j.resuscitation.2011.09.017.
71 Bender A, Howell K, Frey M, et al. Bilateral loss 73 Tiainen M, Poutiainen E, Kovala T, et al. Cognitive
of cortical SSEP responses is compatible and neuro physiologic outcome of cardiac arrest
with good outcome after cardiac arrest. survivors treated with therapeutic hypothermia.
J Neurol 2012;259(11):2481–2483. doi:10.1007/ Stroke 2007;38:2303–2308. doi:10.1161/
s00415-012-6573-8. STROKEAHA.107.483867.
1752 DECEMBER 2018

Critical Care of REVIEW ARTICLE

Neuromuscular Disorders C O N T I N U UM A U D I O
ONLINE
By Diana Greene-Chandos, MD, FNCS;
Michel Torbey, MD, MPH, FCCM, FAHA, FNCS, FAAN
ABSTRACT
PURPOSE OF REVIEW: Weakness is a common reason patients are seen in
neurologic consultation. This article reviews the differential diagnosis of
neuromuscular disorders in the intensive care unit (ICU), discusses the
intensive care needs and evaluation of respiratory failure in patients with CITE AS:
neuromuscular disorders, and provides a practical guide for management.
1753–1775.
RECENT FINDINGS: Although primary neuromuscular disorders used to be the
most common cause for weakness from peripheral nervous system disease Address correspondence to
Dr Michel Torbey, University of
in the ICU, a shift toward ICU-acquired weakness is observed in today’s New Mexico, Department of
clinical practice. Therefore, determining the cause of weakness is important Neurology, 2211 Lomas Blvd NE,
and may have significant prognostic implications. Guillain-Barré syndrome Albuquerque, NM 87106,
mtorbey@salud.unm.edu.
and myasthenia gravis remain the most common primary neuromuscular
disorders in the ICU. In patients with myasthenia gravis, it is important to be RELATIONSHIP DISCLOSURE:
Dr Greene-Chandos has
vigilant with the airway and institute noninvasive ventilation early in the
received personal
course of the disease to attempt to avoid the need for intubation. On the compensation as a speaker for
other hand, patients with Guillain-Barré syndrome should be intubated the Ecuador Neurocritical Care
Society Annual Meeting and the
without delay if the airway is at risk to avoid further complications. In Neurocritical Care Society
patients with ICU-acquired weakness, failure to wean from the ventilator is Annual Meeting and has
usually the challenge. Early mobility, glucose control, minimizing sedation, received research/grant
support from Biogen, the
and avoiding neuromuscular blocking agents remain the only therapeutic Neuroemergency Treatment
regimen available for ICU-acquired weakness. Trials Grant from the National
Institutes of Health, and the
Stroke Network Grant from the
SUMMARY: Critical care management of neuromuscular disorders requires a National Institute of
multidisciplinary approach engaging members of the ICU and consultative Neurological Disorders and
Stroke. Dr Greene-Chandos has
teams. Developing an airway management protocol could have implications worked as a consultant for court
on outcome and length of stay for patients with neuromuscular disorders cases. Dr Torbey has received
in the ICU. Tending to the appropriate nuances of each patient who is personal compensation as the
past president of the
critically ill with a neuromuscular disorder through evidence-based Neurocritical Care Society and
medicine can also have implications on length of stay and outcome. has received grant support from
the National Institute of
Neurological Disorders and
Stroke.
INTRODUCTION UNLABELED USE OF
M
anagement of neuromuscular disorders in the intensive care PRODUCTS/INVESTIGATIONAL
USE DISCLOSURE:
unit (ICU) dates back to the 1950s when the iron lung was used Drs Greene-Chandos and
during the poliomyelitis epidemic. This was the impetus for Torbey report no disclosures.
early critical care neurology. The Batten Respiratory Care Unit
and National Hospital for Neurology and Neurosurgery in © 2018 American Academy
London, United Kingdom, started admitting patients with poliomyelitis, of Neurology.

CRITICAL CARE OF NEUROMUSCULAR DISORDERS
KEY POINTS myasthenia gravis, and tetanus in the 1950s.1 Acute or progressive respiratory
failure remains the most common cause for admission to the neurocritical care
● Intensive care
unit–acquired weakness
unit for patients with neuromuscular disorders. This article focuses on the critical
is a clinical diagnosis. care management of neuromuscular disorders and is not intended to be an
exhaustive review about each disorder.
● Early clinical signs of
respiratory failure include
CHALLENGES OF EVALUATING WEAKNESS IN THE INTENSIVE
tachycardia, tachypnea,
increased sweating, and use CARE UNIT
of accessory muscles. Clinical evaluation of weakness in the ICU can be challenging. Often, patients are
encephalopathic, sedated, mechanically ventilated, or uncooperative. Hence, a
● Hypercapnia, as indicated systematic approach is key to identifying the etiology of the weakness.
on arterial blood gases, is
often a late sign of intensive
Management of the patient’s airway, breathing, and circulation should always
care unit–acquired take precedence over obtaining a detailed history and neurologic examination.2 A
weakness; therefore, differential diagnosis should be created based on the presenting clinical signs and
arterial blood gases are anatomic location of the lesion (TABLE 8-1 and TABLE 8-2).
not very useful in acute
triage decisions.
Although, historically, most patients admitted to the ICU with peripheral
nervous system weakness have an established etiology, the spectrum is shifting
today toward disorders acquired in the ICU. In one study, 28% of patients
admitted to the ICU with weakness had an established neuromuscular disorder.3
The mnemonic MUSCLES (medications, undiagnosed neuromuscular disorder,
spinal cord disease, critical illness related, loss of muscle mass, electrolyte
disorders, systemic illness) is an easy way to remember the different disorders
associated with weakness in the ICU.4 TABLE 8-3 summarizes a more thorough
list of the differential diagnosis options of weakness in the ICU.
RESPIRATORY FAILURE IN NEUROMUSCULAR DISORDERS

Risk of respiratory failure is high in acute neuromuscular disorders, especially
when respiratory mechanics are affected. The early signs of respiratory failure
can be as subtle as restlessness. Other signs of impending respiratory failure
include tachycardia, tachypnea, use of accessory respiratory muscles, staccato
speech, paradoxical breathing, weak cough, and difficulty counting aloud from 1
to 20 in a single breath. Any sign of respiratory distress should trigger an ICU
admission. Oxygen should be administered judiciously to keep SaO2 between
90% and 95% to avoid carbon dioxide retention. Although noninvasive
ventilation could be a temporizing measure, it should not be used to avoid
intubation if the patient is at high risk of respiratory failure. Arterial blood gases
are not very useful in acute triage decisions. A normal PaCO2 in a patient with
tachypnea may be a sign of impending fatigue and respiratory failure. An
elevated PaCO2 may be too late of a finding.
CAUSES OF WEAKNESS IN THE INTENSIVE CARE UNIT

The following section reviews the most common neuromuscular causes of
weakness in patients in the ICU.
Critical Illness Myopathies and Neuropathies

New weakness in a patient who is critically ill is referred to as ICU-acquired
weakness, which is defined as “clinically detected weakness in critically ill
patients in whom there is no plausible etiology other than critical illness.”5 The
incidence of ICU-acquired weakness ranges from 50% to 100% in patients in
the ICU.6 Depending on the pathology, different terms are used to describe
1754 DECEMBER 2018

Clinical Findings Associated With Weakness in the Intensive Care Unita TABLE 8-1
Clinical Finding Some Possible Diagnoses
Acute tetraplegia with sparing of eye movements Locked-in syndrome, residual neuromuscular blockade
Descending paralysis Envenomation, botulism
Ascending paralysis Tick paralysis, Guillain-Barré syndrome
Gastrointestinal symptoms, multiorgan failure Heavy metal toxicity
Abdominal pain, psychiatric symptoms Acute intermittent porphyria
Heliotrope rash Dermatomyositis
Episodic weakness with family history Periodic paralysis
a
Data from Caulfield AF, et al, Neurocrit Care.2
Anatomic Localization of Weakness in Patients in the Intensive Care Unita TABLE 8-2
Localization Weakness Pattern Sensory Loss Reflexes
Cerebral cortex, brainstem, Distal more than proximal, extensors Present if sensory Increased
spinal cord more than flexors tracts involved
Anterior horn cell Proximal and distal Absent Increased in amyotrophic

lateral sclerosis
Peripheral nerve Specific distribution, stocking and Present Decreased

glove pattern
Neuromuscular junction Generalized weakness Absent Normal
Muscle Proximal Absent Normal
a
Data from Caulfield AF, et al, Neurocrit Care.2
Differential Diagnosis of Weakness Requiring Intensive Care Unit Admission TABLE 8-3
Anatomic Structure Diseases
Motor neuron Amyotrophic lateral sclerosis, poliomyelitis, West Nile virus, paraneoplastic motor neuron disease,
Kennedy disease
Peripheral nerve Guillain-Barré syndrome, polyneuropathies, entrapment or compression syndrome, trauma, critical
illness neuropathy, vasculitic neuropathy, porphyria, toxins
Muscle Critical illness myopathy, polymyositis, periodic paralysis, metabolic myopathy, mitochondrial
myopathy, acid maltase deficiency, muscular dystrophy
Neuromuscular Myasthenia gravis, Lambert-Eaton myasthenic syndrome, botulism, snake bite, organophosphate
junction toxicity, hypermagnesemia, prolonged neuromuscular blockade, tick paralysis

this syndrome, including critical illness polyneuropathy, critical illness

myopathy, critical care illness neuromyopathies, and acute necrotizing
myopathy of intensive care. Risk factors for ICU-acquired weakness include bed
rest and immobilization, sepsis, multiorgan failure, use of vasopressors
or aminoglycosides, septic encephalopathy, hyperglycemia, corticosteroid use,
older age, hyperosmolality, hypoalbuminemia, and renal replacement therapy.6
The pathophysiology of ICU-acquired weakness is multifactorial, ranging
from bioenergetic failure to microvascular alterations, ultimately resulting in
functional and structural alterations to the nerves, muscles, or both.
Patients with ICU-acquired weakness (CASE 8-1) typically present with
generalized weakness more pronounced proximally and failure to wean from
mechanical ventilation. Facial, ocular, and bulbar muscles are usually spared.7
The American Thoracic Society guidelines emphasize the importance of using
the Medical Research Council (MRC) scale as a standard to assess muscle
weakness. Six different bilateral muscle groups should be tested and summed to
generate an MRC summed score5: (1) wrist extension, (2) elbow flexion,
CASE 8-1 A 41-year-old man presented to the emergency department because of

shortness of breath and was diagnosed with pneumonia. He had a
history of sarcoidosis involving both lung and brain. He had been on
intermittent corticosteroids for 5 years, and prior to admission he was
taking prednisone 20 mg/d.
He was transferred to the medical intensive care unit (ICU), where he
was intubated, heavily sedated, and placed on bilevel ventilator
support. He met criteria for systemic inflammatory response syndrome
and required 12 hours of vasopressor support because of hypotension,
in part, from sedative drips.
Over the week, he was weaned from bilevel to synchronized
intermittent mandatory ventilation but never to spontaneous modes,
despite improvement in his pneumonia and acute respiratory distress
syndrome, which responded to antibiotics. As sedation was weaned, it
was discovered he was weak, and neurology was consulted for the
diffuse weakness and failure to wean from the ventilator.
Neurologic examination showed 1/5 strength proximally and 2/5
strength distally symmetrically in all four extremities. Facial
movements were strong and symmetric. Reflexes were 1+ throughout,
sensation was intact, and toe signs were equivocal.
Neuroimaging of brain and spine were unchanged from images 6 months
earlier. CSF analysis was normal. A concern for a neurosarcoid flare
remained, and stress dose steroids of 125 mg/d IV methylprednisolone for
systemic inflammatory response syndrome were increased to 1 g of IV
methylprednisolone for 3 days followed by 1 mg/kg/d of prednisone orally.
His examination worsened to 0/5 strength, and his reflexes became
absent. Sensation remained intact, and his mental status showed
intermittent agitation. His creatine kinase level was 83 mg/dL. EMG showed
rare fibrillation potentials and positive sharp waves. Nerve conduction
1756 DECEMBER 2018

(3) shoulder abduction, (4) foot dorsiflexion, (5) knee extension, and (6) hip
flexion. An MRC summed score of less than 48 is diagnostic of ICU-acquired
weakness. In patients who are not able to participate in the examination,
handgrip strength is an alternative test to use.8 Deep tendon reflexes are reduced
or absent in severe weakness or neuropathic cases. Sensory impairment, although
present, is not as severe as the weakness.
Although the diagnosis of ICU-acquired weakness can be made based on the
neurologic examination findings, neurodiagnostic tests can be useful in patients
who are uncooperative.9 Nerve conduction studies usually show a reduction in
the amplitude of the compound muscle action potential (CMAP). A normal
CMAP 8 days after ICU admission has a high predictive value of excluding
ICU-acquired weakness.10 A nerve to muscle CMAP ratio can help differentiate
between a muscle and nerve injury. The first CMAP is obtained by stimulating
the nerve. The second CMAP is obtained by direct stimulation of the muscle
following needle insertion. A nerve-to-muscle CMAP ratio of less than 0.5 is
diagnostic of an isolated neuropathy, whereas a ratio of more than 0.5 is
studies showed low-normal amplitudes of sensory nerve action potentials

(SNAPs) and decreased amplitudes of compound muscle action potentials
(CMAPs). The patient underwent a tracheostomy. The diagnosis of critical
illness myopathy was finally confirmed by muscle biopsy.
This is a complex case, given the neurosarcoidosis component, but the COMMENT
diagnosis ultimately was critical illness myopathy. This condition typically
occurs in the setting of a prolonged ICU course with multiorgan system failure,
particularly in the setting of acute respiratory distress syndrome. The severity
of the disorder was enhanced by using high-dose steroids and neuromuscular
blockade. Examination typically shows low tone with proximal greater than
distal weakness (but also may be equally diffuse) and minimal to absent
reflexes. Sensation is normal. Creatine kinase can be normal to mildly
elevated. EMG may show muscle irritability and decreased recruitment.
Nerve conduction studies show low to normal amplitude SNAPs and
low-amplitude CMAPs. CSF is unremarkable and is not part of the diagnosis
for critical illness myopathy but was evaluated in this case because of
concerns of neurosarcoidosis being responsible for neurologic decline.
The treatment is to reduce steroids as quickly and safely as possible (here,
the high doses of methylprednisolone that were given due to concerns of
neurosarcoid exacerbation worsened the situation), optimize nutrition, and
control blood glucose. Critical illness polyneuropathy was also a possibility in
this case given the need for vasopressors and systemic inflammatory
response syndrome. (Vasopressor use is a well-known risk factor for the
development of polyneuropathy, especially in the context of systemic
inflammatory response syndrome and neuromuscular blockage.) Therefore, it
is recommended to do both a nerve and muscle biopsy in most settings.

KEY POINTS diagnostic of either a myopathy or combined etiology.6 To further differentiate

between a muscle and nerve etiology, measurement of the motor unit action
● Early mobilization and
glucose control are key in
potentials (MUAPs) can be helpful. In a myopathy, MUAP duration is short and
the prevention of intensive amplitude is low. The limiting step for measuring MUAPs is patient cooperation.
care unit–acquired Sedated patients will not be able to actively participate in recruitment efforts.
weakness.
Therapeutic Options
● Rhabdomyolysis occurs
with a wide spectrum of Early mobilization is very important. Early therapy combined with minimizing
signs and symptoms, ranging sedation is associated with a favorable outcome and shortened duration of
from severe muscle pain and mechanical ventilation.11
varying degrees of Hyperglycemia is a significant risk factor for ICU-acquired weakness.12
generalized weakness.
Markedly elevated plasma Intensive insulin therapy significantly reduces the incidence of critical illness
creatine kinase may lead to polyneuropathy/critical illness myopathy as well as the duration of mechanical
myoglobinuria and acute ventilation and length of stay in the ICU.13–15 In a prospective clinical trial, rates
kidney injury. of critical illness polyneuropathy were significantly lower in the intensive insulin
therapy group (80 mg/dL to 140 mg/dL) compared to the less restrictive
(180 mg/dL to 200 mg/dL) group (10% versus 45%, P=.01).16 The duration of
mechanical ventilation was also reduced in the intensive insulin therapy group
(9.72 days versus 14.05 days; P=.04).
Although immunoglobulin infusion may theoretically have a possible
therapeutic application in preventing critical illness polyneuropathy and critical
illness myopathy, a randomized clinical trial using either 0.25 g/kg/d or 1%
human albumin solution over 3 days found no treatment effect.17
Rhabdomyolysis
Almost 26,000 patients with rhabdomyolysis are hospitalized every year in
the United States.18 Rhabdomyolysis occurs with a wide spectrum of signs and
symptoms, ranging from severe muscle pain and varying degrees of generalized
weakness. Markedly elevated plasma creatine kinase (CK) may lead to
myoglobinuria and acute kidney injury.19 The laboratory diagnosis of
rhabdomyolysis is based on the measurement of serum or plasma CK. This is still
considered the most sensitive test. A fivefold to 10-fold increase in concentration
is considered diagnostic. In the case of a single event (ie, mostly trauma), CK levels
tend to rise in the first 12 hours, peak on the second or third day, and return to
baseline 3 to 5 days later. Although the CK measurement is still considered
the biochemical gold standard for diagnosing rhabdomyolysis, serum myoglobin
is considered the gold standard in prognostication, especially for the extent of
acute kidney injury.
Once a definitive diagnosis of rhabdomyolysis is established or suspected,
fluid infusion should be immediately initiated. The goal is to maintain a urinary
output of 200 mL/h to 300 mL/h, urine pH of more than 6.5, and plasma pH of
less than 7.50. Hemodynamic parameters and urine output should be monitored
closely. It is still important to recognize that the treatment hallmark is
“aggressive hydration by means of sterile saline and glucose solutions, since
studies show that alkalinization adds little to the beneficial effect of hydration.”20
Fluid should be tapered once myoglobinuria is cleared. Precipitating conditions
must be promptly managed.
One rare cause of rhabdomyolysis is propofol infusion syndrome. High doses
of propofol (>5 mg/kg/h) given for a prolonged period of time (>48 hours) is a
contributing factor. Propofol infusion syndrome can be fatal and is characterized
1758 DECEMBER 2018

by metabolic acidosis, cardiac and renal failure, rhabdomyolysis, myoglobinuria,
and hyperlipidemia. Prompt recognition and interruption of the propofol infusion
can reverse the syndrome.
Myasthenia Gravis
Myasthenia gravis is an autoimmune disorder characterized by defective
neuromuscular junction transmission most commonly due to antibodies directed
against the postsynaptic acetylcholine receptor (AChR).21,22 Autoimmune
myasthenia gravis has a worldwide prevalence of 40 to 180 per million people
and an annual incidence of 4 to 12 cases per million.23,24 Myasthenia gravis has a
bimodal age distribution, with a higher incidence in young women and older men,
but it can occur at any age.
PATHOPHYSIOLOGY. Myasthenia gravis is now being categorized by individual

antibodies targeting different components of the postsynaptic membrane including
AChR, muscle-specific tyrosine kinase (MuSK), and lipoprotein receptor–related
protein 4 (LRP4). About 85% of antibodies are directed against AChR,25 and 50%
of patients with myasthenia gravis who are AChR seronegative have antibodies
against MuSK. The latter are usually associated with bulbar and respiratory muscle
involvement. TABLE 8-4 summarizes the different types of myasthenia gravis.
CLINICAL FEATURES. Fatigable weakness is the hallmark of the disease. All muscle
groups can be affected. Patients may have ptosis triggered by sustained upgaze.
Ophthalmoparesis is common. Weakness in extraocular muscles is rarely symmetric,
whereas limb weakness is symmetric with proximal preponderance.26 Gag reflex
could be depressed. Head drop can be seen in patients with MuSK antibodies.
Myasthenic crisis is defined as “myasthenia gravis complicated by respiratory
failure requiring mechanical ventilation or delayed extubation for more than
24 hours in an already intubated patient with myasthenia gravis.”24 Myasthenic
crisis is the most common reason for admission of a patient with myasthenia
gravis to the ICU. About 20% of patients will have a crisis within the first year
of diagnosis, usually triggered by a precipitating factor such as preceding respiratory
Myasthenia Gravis Classification TABLE 8-4
Myasthenia Gravis Subtype Antibody Age at Onset Respond to Thymectomy
Early onset Acetylcholine receptor <50 Yes
Late onset Acetylcholine receptor >50 No
Thymoma associated Acetylcholine receptor Variable Yes
Muscle-specific tyrosine Muscle-specific tyrosine kinase Variable No

kinase–myasthenia gravisa
Lipoprotein receptor–related protein Lipoprotein receptor–related protein 4 Variable No data

4–myasthenia gravisb
a
Predominant symptoms of muscle-specific tyrosine kinase–myasthenia gravis include cranial and bulbar muscle symptoms.
b
Predominant symptoms of lipoprotein receptor–related protein 4–myasthenia gravis include ocular symptoms of generalized mild myasthenia
gravis; respiratory failure is rare.

CASE 8-2 A 58-year-old woman with a history of acetylcholine receptor antibody–

positive myasthenia gravis, hypertension, diabetes mellitus, and
diverticulitis who underwent colectomy 5 years ago was hospitalized for
small bowel obstruction. She was managed with a status of nothing by
mouth and nasogastric tube to low wall intermittent suctioning. She had
not received her usual medications, pyridostigmine and prednisone, for
2 days.
She decompensated from a bowel standpoint and was taken to the
operating room for lysis of adhesions. She was extubated in the
postanesthesia recovery unit upon the first sign of awakening. She had
been given IV methylprednisolone by the anesthesiologist as a stress
dose intraoperatively.
She immediately failed extubation and required urgent reintubation.
She was taken to the surgical intensive care unit, but within a few hours it
was clear that she had an exacerbation of her myasthenia gravis. She had
bilateral ptosis, facial and neck weakness, as well as 3/5 proximal
strength diffusely with 4/5 distally. Her negative inspiratory force on the
ventilator was –10 cm H2O.
She was transferred to the neurocritical care unit where plasma
exchange was started. Once extubated after two rounds of plasma
exchange, her usual total daily dose of pyridostigmine was given in a
continuous IV infusion (as 1/60 of the daily dose) while waiting for return
of bowel sounds and ability to take food and water by mouth. Of note, it
was discovered that the patient had also received magnesium sulfate 4 g
IV in the surgical intensive care unit as part of an order set for electrolyte
replacement prior to transfer.
COMMENT Patients with myasthenia gravis who have gastrointestinal illnesses with
vomiting and are unable to take their medications, such as prednisone and
pyridostigmine, are not only at risk for decompensation because of the
missed medications but also from the underlying illness. If patients are
hospitalized, these medications should be given in an IV formulation of an
equivalent dose. Giving pyridostigmine in an IV infusion is preferred to
prolong the effects of the formulation and decrease the cardiac effects of
a bolus. If a patient is likely going to be sent to the operating room for
any reason (especially in the setting of emergent surgery), a risk exists for
exacerbation from the stress of the operation as well as from the effects
of the pharmacologic neuromuscular blockade. Thus, plasma exchange
or IV immunoglobulin (IVIg) prior to the planned operation can be provided
as a preventative measure. Care must be taken by the anesthesiologist
to know the respiratory parameters when considering extubation in a
patient with myasthenia gravis requiring an urgent operation and to not
extubate based on recovery of consciousness and train of four response
alone. In addition, care must be taken so that patients with myasthenia
gravis do not receive any potentially exacerbating medications that may be
routinely ordered in hospital units.
1760 DECEMBER 2018

illness, surgery, and medications.27 Some of the common medications implicated KEY POINT
in myasthenic crisis seen in the ICU include aminoglycosides, quinolones, lithium,
● Noninvasive ventilation
calcium channel blockers, chlorpromazine, cephalosporins, and penicillins. should be considered
Although cholinergic crisis is rare, it certainly can result in an ICU admission and early in a patient with
mimic myasthenic crisis. The presenting signs are consistent with cholinergic myasthenia gravis.
toxicity. The acronym SLUDGE (salivation, lacrimation, urination,
defecation, gastrointestinal upset, and emesis) is an excellent way to
remember the symptoms of a cholinergic crisis.
DIAGNOSIS. Two useful bedside tests that can help differentiate myasthenia
gravis from other conditions are the edrophonium test and the ice pack test. The
edrophonium test should only be performed in the inpatient setting because of
its side effects of hypotension, increased secretions, and bradycardia. The ice
pack test is usually performed quickly at the bedside. To perform the ice pack
test, the examiner applies ice to the eye for about 2 minutes. The ice should be
covered to prevent any injury secondary to ice burns. The test is considered positive
with improvement of the patients’ double vision or ptosis. However,
electrodiagnostic tests remain the best standard bedside tests. A CMAP
decrement of at least 10% with repetitive stimulation is consistent with a
diagnosis of myasthenia gravis. Single fiber EMG findings include jitter and
blocking.28 These tests are usually difficult to perform and interpret in the ICU.
Neurophysiologic tests are not necessary in patients with seropositive myasthenia
gravis but are key to making the diagnosis in seronegative myasthenia gravis.
Antibody testing plays an important role. A positive AChR antibody titer with
muscle weakness is confirmatory. If AChR antibodies are negative, MuSK
antibodies should be tested, especially in bulbar myasthenia gravis. LRP4
antibodies are reported in 19% of patients with myasthenia gravis who are
seronegative.29 Antistriational muscle antibodies are identified in 90% of
patients with myasthenia gravis with thymoma.
TREATMENT. Patients with myasthenic crisis (CASE 8-2) should be admitted to

the neurocritical care unit if the airway appears at risk. Inciting drugs should be
immediately stopped if possible. Over the last decade, ICU treatment has
decreased mortality from myasthenia gravis crisis from 50% to 5%.30
MECHANICAL VENTILATION. Noninvasive ventilation should be started promptly.

Immediate bilevel positive airway pressure (BiPAP) support in patients with
myasthenia gravis crisis was found to decrease the need for intubation and decrease
length of stay in the ICU.31 Elective intubation should be considered in patients
with significant bulbar disease and respiratory weakness. When intubation is
necessary, it is important to administer lower doses of nondepolarizing blocking
agents such as 0.5 mg/kg of rocuronium bromide instead of 1 mg/kg.32 Most
patients will require intubation for more than 1 week.27,33 Indications for
intubation include severe hypercapnia, profound hypoxia, lobar atelectasis, and
contraindications for BiPAP. Extubation failure is not uncommon.34
PHARMACOLOGIC TREATMENT AND CHOLINESTERASE INHIBITORS. Oral

pyridostigmine is most commonly used. This is usually a symptomatic treatment
and can be stopped if the patient is intubated. Patients with positive MuSK
antibodies have a poor response or intolerance to acetylcholinesterase inhibitors.35

PHARMACOLOGIC TREATMENT AND IMMUNOTHERAPY. The administration of either

IV immunoglobulin (IVIg) or plasma exchange is considered standard care.36
Both therapies are equally effective in patients with moderate or severe
myasthenia gravis. Combining the two therapies does not add any benefits.37,38
Some evidence suggests that IVIg may be safer in elderly patients with
myasthenia gravis and complex medical comorbidities.30 No difference has been
reported between total doses of 1 g/kg and 2 g/kg IVIg in treatment of myasthenia
gravis exacerbation.39
All patients with myasthenia gravis in crisis should receive steroids.40 When
steroids are first initiated, daily prednisone with 0.75 mg/kg to 1 mg/kg ideal
body weight is most commonly used while receiving other immunomodulatory
therapy. Airway watch should be considered in patients who are nonintubated
since 30% to 50% of patients may experience worsening of their myasthenia
gravis symptoms within days of starting steroid therapy. Although alternate-day
steroid dosing could minimize side effects, it should be avoided in patients with
diabetes mellitus. The beneficial effect of steroids usually manifests after 2 to
6 weeks.
Azathioprine (total daily dose of 2 mg/kg to 3 mg/kg) in combination with
steroids is a first-line immunosuppressive therapy. The combination therapy
increased remission time and decreased relapses compared to steroids alone.41
Second-line agents to consider include methotrexate, cyclophosphamide,
mycophenolate mofetil, tacrolimus, and cyclosporine. A steroid-sparing strategy
is also reasonable with azathioprine or mycophenolate mofetil during the acute
or subacute phases.
Rituximab is another alternative as monotherapy and in refractory
myasthenia gravis.42 Rituximab does appear to have a more profound and
long-lasting effect in patients with MuSK antibodies. It should be considered in
refractory patients as steroid-sparing therapy and in patients with MuSK
antibodies.43,44
THYMECTOMY. The benefits of thymectomy in patients with myasthenia gravis

with thymoma are clear but remain controversial for patients without thymoma.
An increased rate of remission in patients with generalized and severe
myasthenia gravis was observed after thymectomy. Hence, it is recommended
as a therapeutic option in this group.45 Thymectomy is not recommended for
patients with myasthenia gravis who are positive for MuSK antibodies because
efficacy has not been found, and minimal to nonexistent thymic abnormalities
occur in these individuals.46 Thymectomy, if considered, should be performed
early but never as an emergency. IVIg or plasma exchange before surgery will
improve symptoms of myasthenia gravis, reduce the risk of complications, and
contribute to faster recovery.47
Lambert-Eaton Myasthenic Syndrome

Lambert-Eaton myasthenic syndrome (LEMS) is an autoimmune disorder of the
neuromuscular junction with antibodies against the presynaptic P/Q-type
voltage-gated calcium channels (VGCCs).48 LEMS is a rare disorder, with a
worldwide prevalence of 3 to 4 per million population.49 The syndrome may
occur as a paraneoplastic disorder or as a nontumor-related autoimmune
disorder.50 Paraneoplastic LEMS is most often associated with small cell lung
cancer and accounts for 47% to 62% of cases.49
1762 DECEMBER 2018

Diagnosis of LEMS is usually made based on a thorough history and physical KEY POINTS
examination, EMG, and serology. Leg weakness is most often the first presenting
● A hallmark of
symptom in patients with LEMS. The weakness slowly progresses to involve the Lambert-Eaton myasthenic
arms, distal muscles, and finally the cranial muscles. General fatigue, dry mouth, syndrome is postexercise
impotence, dry eyes, constipation, and difficulty swallowing are also reported. facilitation. Tendon reflexes
Cerebellar ataxia, although uncommon, is associated with paraneoplastic and muscle strength
improve immediately after
LEMS.49 Around 10% of patients with LEMS are seronegative for anti-P/Q-type
a brief maximal contraction.
VGCC antibodies.51 Although antibody detection is diagnostic, the absence of
VGCC antibodies does not exclude LEMS. ● Lambert-Eaton
Electrodiagnostic studies are very helpful in confirming the diagnosis and in myasthenic syndrome
helping differentiate between LEMS and myasthenia gravis. Repetitive nerve should be considered in
patients with any
stimulation should be done on at least two distal muscles.52 The test is considered unexplained weakness
positive if: (1) low-amplitude CMAP is present at rest, (2) further decrease in CMAP after pharmacologic
(at least 10%) occurs during low-rate (2 Hz to 5 Hz) repetitive nerve stimulation, neuromuscular blockade.
and (3) an increase in CMAP by 100% occurs after a brief maximal contraction.
● The potassium channel
Treatment centers around symptomatic relief and disease management. The blocker 3,4-diaminopyridine
potassium channel blocker 3,4-diaminopyridine is an effective symptomatic is an effective symptomatic
treatment. Patients with persistent symptoms may also need additional treatment for Lambert-Eaton
immunomodulatory therapy with steroids, IVIg, plasma exchange, or, in some myasthenic syndrome.
cases, azathioprine. Neurocritical care unit needs are usually centered around
respiratory failure or infectious complications.
Botulism
Botulism is associated with gram-positive, spore-forming Clostridium botulinum.
Although several forms of botulism exist, gastrointestinal and inhalational forms
would be anticipated in a deliberate terrorist attack. Botulinum toxin blocks the
release of acetylcholine at the peripheral cholinergic nerve terminals. If exposure
was through inhalation of the botulinum toxin, symptoms are expected within
6 hours. Descending paralysis is usually the first presenting symptom. Cranial
nerve dysfunction with double vision, swallowing difficulties, and ptosis usually
follow soon after. Respiratory failure is not uncommon.53,54 The diagnosis of
botulism is usually clinical, although nerve conduction studies and cultures can
also confirm the diagnosis. The treatment centers around administration of
the antitoxin and supportive care. Mechanical ventilation may be needed in
some cases.
Guillain-Barré Syndrome
Guillain-Barré syndrome (GBS) is an autoimmune-mediated nerve injury
characterized by an acute onset of rapidly progressive and symmetric ascending
weakness. Currently, GBS is a syndrome of polyradiculoneuropathies that
include an acute inflammatory demyelinating polyradiculoneuropathy (AIDP),
an acute motor axonal neuropathy (AMAN), an acute motor-sensory axonal
neuropathy (AMSAN), and Miller Fisher syndrome.
The incidence of GBS varies between 0.4 and 3.25 cases per 100,000 per year.55
The disease is most commonly precipitated by an infection. A variety of
infectious agents have been implicated including cytomegalovirus, Epstein-Barr
virus, herpes simplex viruses, mycoplasma, influenza, and Campylobacter
jejuni.56 Zika virus has recently been associated with increased incidence of GBS
in several countries.57,58 Immunization, recent surgery, and renal transplantation
have been associated with GBS.56

CLINICAL PRESENTATION. The main clinical features of GBS are weakness and
areflexia. Weakness evolves rapidly over days with an ascending pattern.
Dysesthesia of the feet and hands often presents as the initial symptom.
Neurologic examination demonstrates a symmetric limb weakness pattern,
decrease or loss of reflexes, and objective sensory findings. Cranial nerves can be
affected, with weakness mainly affecting facial and oropharyngeal muscles.
Ophthalmoparesis is a rare presentation usually associated with the Miller Fisher
syndrome variant.
LABORATORY TESTS. Elevated CSF protein without pleocytosis (albuminocytologic

dissociation) is characteristic of GBS. CSF protein is usually normal early in the
course of the disease (24 to 48 hours) but is expected to increase later in the
course of the illness. Routine laboratory studies are usually normal. Early in the
disease, the only electrophysiologic abnormality observed is delayed or absent
F waves.59
NEUROLOGIC INTENSIVE CARE MANAGEMENT. Management of patients with GBS

is complex and often requires admission to the ICU. The progression and
severity of the disease are variable, and 33% of patients will require admission
to an ICU for ventilatory support.60 The following are the criteria used at the
author’s institution for admitting patients with GBS to the neurocritical care unit:
u Rapid progression of motor weakness including respiratory muscles

u Presence of bulbar dysfunction and bilateral facial palsy
u Autonomic dysfunction
u Arrhythmia and bradycardia
u Medical complications (eg, deep vein thrombosis, pulmonary embolus, myocardial
infarction, sepsis)
RESPIRATORY FAILURE. Although the clinical manifestations of respiratory failure

are helpful in the initial triage of patients with GBS, pulmonary function
tests measuring forced maximal inspiratory pressure, maximal expiratory
pressure, and vital capacity are very useful in tracking disease progression.
TABLE 8-5 Accuracy of Postintubation Respiratory Parameters in Patients With

Neuromuscular Respiratory Failurea
Positive Pressure
Respiratory Parameter Sensitivity Specificity Ventilation
Vital capacity 67% 70% 84%
Maximal inspiratory pressure 60% 90% 94%
Maximal expiratory pressure 75% 100% 100%

b
Pulmonary function ratio <1 70% 100% 100%
a
Data from Lawn ND, Wijdicks EF, Muscle Nerve.64
b
The pulmonary function ratio is the measurement on day 12 divided by day 1.
1764 DECEMBER 2018

The 20-30-40 rule is often followed to decide on the need for intubation56: KEY POINTS
(1) vital capacity decreased to 20 mL/kg, (2) maximal inspiratory pressure
● Plasma exchange and IV
decreased to –30 cm H20, and (3) maximal expiratory pressure decreased to immunoglobulin are equally
40 cm H20. effective for the treatment
Mechanical ventilation should not be delayed if patients have not met all of Guillain-Barré syndrome.
these criteria. Earlier intubation should be considered to minimize pulmonary The decision to use either
should depend on the
complications and avoid risk of emergent intubation. Ventilatory weaning availability of these
should be considered as a patient’s condition improves. Several clinical criteria treatments at the specific
should be met prior to extubation: (1) vital capacity of greater than 15 mL/kg, treatment center.
(2) maximal inspiratory pressure of better than –30 cm H2O, and (3) satisfactory
● No evidence exists to
oxygenation.56 Extubation with direct laryngoscopic visualization or after a cuff
recommend any specific
test should be considered in cases of prolonged intubation. Most patients with ventilatory mode in patients
GBS will need a tracheostomy since mechanical ventilation duration is with Guillain-Barré
prolonged.61 Absent motor responses and widespread fibrillations early in the syndrome.
course of the disease are predictive of prolonged intubation.62 Lack of foot
flexion ability at ICU admission and at the end of therapy or inability to lift the
arm 1 week after intubation also are highly predictive of prolonged mechanical
ventilation.61,63 The pulmonary function score also has been found to be
predictive of prolonged intubation and tracheostomy.64 The pulmonary function
score is the sum of vital capacity, maximal inspiratory pressure, and maximal
expiratory pressure. The predictive performance of each of the respiratory
parameters is shown in TABLE 8-5. It may be reasonable to perform early
tracheostomy for patients expected to require prolonged intubation.
AUTONOMIC DYSFUNCTION. Dysautonomia is associated with higher mortality in

cases of severe GBS.65 Frequent manifestations often include sinus tachycardia or
bradycardia, hypertension, and labile blood pressure. Patients with labile blood
pressure should be admitted to the ICU for monitoring. The fluctuations in blood
pressure are often limited and do not require specific treatment. In severe cases of
bradycardia, placement of a temporary pacemaker may be needed. In patients
who are hypotensive, IV fluids should be initiated early. If hypotension is
refractory to fluids. a vasopressor may be needed. Patients should be assessed for
postural hypotension. Concerns of cardiac arrhythmias secondary to endotracheal
suctioning should be discussed with nurses and respiratory therapists.
NUTRITION. Patients with GBS admitted to the ICU are often hypercatabolic.
Enteral feeding should be instituted as soon as possible to avoid an increased risk
of infectious complications and failure to wean from mechanical ventilation. An
initial replacement at 30 kcal/kg to 40 kcal/kg nonprotein calories and 2.0 g
protein/kg to 2.5 g protein/kg is recommended.66 Indirect calorimetry provides a
more accurate estimate of caloric expenditure. Weekly 24-hour urine samples
can help calculate nitrogen balance and protein needs. Weekly serum transferrin
should be collected instead of albumin, given the potential effect of albumin
infusion during the course of plasma exchange.
PSYCHOLOGICAL SUPPORT. Patients with GBS may experience hopelessness and

depression, which can significantly impact their medical therapy. It is important
for the ICU team to consider engaging the patient. A reader board should be
provided for patients unable to speak to help them avoid a possible intense sense
of isolation.

PLASMA EXCHANGE AND INTRAVENOUS IMMUNOGLOBULIN. Large multicenter

controlled trials have shown benefit from plasma exchange in the treatment of
patients with GBS.67 Treatment is most effective within 1 week of symptom
onset. A total of five plasma exchange sessions are currently standard treatment.
The amount of total plasma removed is generally around 200 mL to 250 mL of
plasma per kg body weight.
IVIg is an alternative treatment for GBS. A total of 400 mg/kg/d IVIg for
5 days or 1 g/kg/d for 2 days is the current therapeutic regimen. This treatment
offers the additional advantage of simple administration and does not require
specialized equipment or personnel. Two large clinical trials have found IVIg
to be equivalent to plasma exchange but with lower complication rates.68,69 In
patients who fail their first course of IVIg treatment, a second course may be
beneficial. Combining IVIg and plasma exchange does not provide any additional
benefit.69
The decision to treat with either plasma exchange or IVIg may be center specific
depending on the availability of personnel for central access placement or
availability of plasma exchange equipment. The side effect profile may also play a
role in some patients. TABLE 8-6 summarizes the advantages/disadvantages and
side effect profile of each therapy.
CORTICOSTEROIDS. Although corticosteroids alone have no benefit in treatment

of patients with GBS, the combination of steroids and IVIg could have some
short-term benefit.70
LONG-TERM OUTCOME. Most patients with GBS are expected to make a good
recovery over weeks to months. The majority of patients who are mechanically
ventilated will regain the ability to walk independently up to 2 years after GBS
TABLE 8-6 Comparison of Intravenous Immunoglobulin and Plasma Exchange
Treatment Side Effects Advantages Disadvantages
IV Immunoglobulin Reversible encephalopathy; Convenience, could be infused in Expensive, contraindicated in

(IVIg) chest pain at initiation; peripheral veins, patient comfort, patients with congestive heart
infusion-related side effects easy to initiate failure or renal insufficiency
(up to 24 hours) including
fatigue, fever, nausea;
thromboembolism; aseptic
meningitis; acute kidney injury;
migraine; allergic reactions
(IgA deficiency)
Plasma exchange Paresthesia (hypocalcemia Allows for IVIg use as second Requires central line access;
secondary to citrate in course of treatment in patients may not be available in facilities
exchange solution), transfusion who are refractory to treatment with poor resources;
reaction, hypotension, central contraindicated in patients with
access complications, septic shock, myocardial
infection, pneumothorax, infarction within 6 months;
hematoma marked dysautonomia; active
bleeding
IgA = immunoglobulin A; IV = intravenous.
1766 DECEMBER 2018

onset.71 Factors associated with poor outcome include: (1) age older than 60 years, KEY POINTS
(2) rapid progression to severe weakness (7 days or less), (3) need for ventilatory
● Noninvasive ventilation
support, and (4) a mean distal CMAP amplitude of 20% of normal or less.72 has a limited role in patients
The overall mortality rate varies between 2.4% and 6.4%.73 Patients who are with severe Guillain-Barré
mechanically ventilated continue to have a higher mortality rate, ranging from syndrome.
15% to 30%.56 Despite close monitoring in the ICU, acute mortality is mostly
● Appropriate
caused by dysautonomia, infection, acute respiratory distress syndrome, or
psychological support is
pulmonary emboli. recommended for patients
with Guillain-Barré
Vasculitic Neuropathy syndrome in the intensive
Although the clinical presentation of the vasculitides is diverse, they all share care unit.
a common pathology of blood vessel inflammatory disease resulting in vascular ● The types of amyotrophic
damage and ischemic injury. Vasculitic neuropathy is often a complication lateral sclerosis include
of vasculitic disorders affecting small- to medium-size vessels but can also limb-onset amyotrophic
occur without any evidence of systemic vasculitis. This single-organ vasculitis lateral sclerosis, which has
both upper motor neuron
of the peripheral nervous system is referred to as nonsystemic vasculitic and lower motor neuron
neuropathy. Although nonsystemic vasculitic neuropathy presents as an acute, signs; bulbar-onset
relapsing, multifocal neuropathy, it can present as a slowly progressive amyotrophic lateral
neuropathy without any distinct asymmetries. The diagnosis requires a nerve sclerosis, which features
dysphagia and dysarthria
biopsy with or without a concomitant muscle or skin biopsy.74 All patients
and upper motor neuron and
with progressive nonsystemic vasculitic neuropathy should be treated. Initial lower motor neuron signs
treatment includes corticosteroid monotherapy unless the neuropathy is rapidly that occur after the disease
progressive. Combination therapy of steroids and immunosuppressive therapy progresses; primary lateral
sclerosis, which has upper
should be considered for patients with rapidly progressive nonsystemic
motor neuron involvement
vasculitic neuropathy or in patients whose monotherapy has failed.74 only; and progressive
muscular atrophy, which has
Amyotrophic Lateral Sclerosis lower motor neuron
Amyotrophic lateral sclerosis (ALS) is a progressive disease characterized by involvement only.
upper and lower motor neuron signs. Progressive weakness is observed in the
bulbar, limb, thoracic, and abdominal muscles. Oculomotor and sphincter
function are usually not affected.75 The diagnosis is based primarily on clinical
examination in conjunction with EMG.
In the United States and Europe, the cumulative lifetime risk of ALS is
1 in 400.76 Typically, death due to respiratory paralysis occurs in 3 to 5 years.
About 5% to 10% of patients may survive for a decade or more.77 Factors
associated with shorter survival time are older age at symptom onset, early
respiratory muscle dysfunction, and bulbar-onset disease.78 The mean age of
onset is 43 to 52 years in familial cases and 58 to 63 years in sporadic cases
of ALS.79
The types of ALS include limb-onset ALS, which has both upper motor neuron
and lower motor neuron signs; bulbar-onset ALS, which features dysphagia
and dysarthria and upper motor neuron and lower motor neuron signs that occur
after the disease progresses; primary lateral sclerosis, which has upper motor
neuron involvement only; and progressive muscular atrophy, which has lower
motor neuron involvement only. Frontotemporal dementia and frontal lobe
cognitive impairment can occur in patients with ALS. Isolated diaphragmatic
weakness with progressive ventilator failure is an uncommon variant of ALS.3,80
These patients develop progressive shortness of breath with exertion. They are
usually intubated in the emergency department, and a neurology consult is
requested for failure to wean days or weeks later. Although mild oropharyngeal

and limb weakness may be present, respiratory weakness is the dominating

presenting symptom.
Riluzole was the first drug approved for treatment of ALS by the US Food and
Drug Administration (FDA).81 In 2017, the FDA also approved edaravone for
the treatment of ALS.82 Patients and their families should be involved in all
treatment decisions based on realistic expectations of the outcome.
INTENSIVE CARE UNIT MANAGEMENT. Symptomatic treatment is the mainstay

of the disease. TABLE 8-7 summarizes the approach to the different prominent
and incapacitating symptoms.83,84
VENOUS THROMBOSIS. The annual incidence of deep vein thrombosis in patients

85
with ALS is 2.7%. The increased risk is mostly due to immobility. Although
no studies have been conducted on the management of deep vein thrombosis
in patients with ALS, deep vein thrombosis should be treated with
anticoagulants.84
RESPIRATORY FAILURE. Respiratory complications are the main cause of death in

patients with ALS due to diaphragmatic weakness, aspiration, or pneumonia.
Evaluating Respiratory Function in Patients With Amyotrophic

Lateral Sclerosis
Although forced vital capacity is the most commonly used respiratory function
test in patients with ALS, it is not a very sensitive test.86 Supine forced vital
capacity may be a better predictor of diaphragmatic weakness than erect forced
vital capacity.87 Sniff nasal pressure showed greater predictive power than
forced vital capacity. Sniff nasal pressure detected hypercapnia with a sensitivity
of 90% and a specificity of 87%.81 Nocturnal desaturation of less than 90% for
1 cumulative minute is another sensitive indicator of respiratory failure.86
A mean SaO2 of less than 93% was associated with poor outcome.88 No test
has predictive power in patients with bulbar weakness.
Noninvasive positive pressure ventilation typically consists of BiPAP. Patients
with ALS without severe bulbar dysfunction treated with noninvasive positive
pressure ventilation had an improved quality of life and survived longer
TABLE 8-7 Approach to Symptomatic Management of Amyotrophic Lateral Sclerosis
Symptoms Treatment
Sialorrhea Amitriptyline 10 mg oral 3 times a day; atropine drops 0.5% to 1% 3 to 4 times a day sublingually;
transdermal scopolamine 1.5 mg every third day; botulinum toxin injection into the parotid gland
may be considered in patients refractory to treatment
Pseudobulbar emotional Combination of dextromethorphan and quinidine; antidepressants such as amitriptyline,

lability fluvoxamine, and citalopram
Cramps Levetiracetam 500 mg twice daily for 7 days, then 1000 mg twice daily for 7 days, then 1500 mg
twice daily83
Spasticity Regular physical therapy can help, antispasticity drug can be tried
1768 DECEMBER 2018

compared with those who underwent standard care.89,90 The median survival KEY POINTS
in the noninvasive positive pressure ventilation group was 48 days longer than
● Respiratory complications
the standard care group. Noninvasive positive pressure ventilation did not are the main cause of death
appear to prolong survival in participants with poor bulbar function.89 in patients with amyotrophic
lateral scerlosis due to
MECHANICAL VENTILATION. Ventilation via tracheostomy should be offered as diaphragmatic weakness,
aspiration, or pneumonia.
an alternative to patients who do not tolerate noninvasive positive pressure
ventilation. Tracheostomy also prolongs survival of patients with ALS ● Patients with
(10.39 months versus 0.83 months for patients who refuse tracheostomy; amyotrophic lateral
P<.0001 in one prospective study).91 sclerosis without severe
bulbar dysfunction treated
with noninvasive positive
SECRETION MANAGEMENT. Ineffective coughing often leads to retained pressure ventilation had an
secretions and ultimately to pneumonia. Peak cough expiratory flow is improved quality of life and
commonly used to measure cough effectiveness. Patients with a mean peak survived longer compared
cough expiratory flow of more than 337 L/min had a significantly greater with those who underwent
standard care.
chance of being alive at 18 months.81 A peak cough expiratory flow of more than
160 L/min is needed to clear secretions. Assistive devices are recommended ● Ventilation via
once peak cough expiratory flow drops below 270 L/min.81 Mechanical tracheostomy should be
insufflation/exsufflation is very effective in increasing peak cough expiratory offered as an alternative to
flow. High-frequency chest wall oscillation is also an alternative approach in patients who do not tolerate
noninvasive positive
clearing secretions. N-acetylcysteine (200 mg to 400 mg 3 times daily), pressure ventilation.
β-receptor antagonists, or anticholinergic bronchodilators should be tried to
decrease bronchial secretions.
NUTRITION IN AMYOTROPHIC LATERAL SCLEROSIS. Oral intake is usually limited

in patients with dysphagia. The prevalence of malnutrition in patients with ALS
varies between 16% and 55%.92 Multiple groups have reported an association
between body mass index (BMI) and ALS survival, with a BMI of <18.5 kg/m2
associated with shorter survival, and moderate obesity (BMI of 30 kg/m2 to
35 kg/m2) associated with slower disease progression and longer survival. The
clinical trial High Fat/High Calorie Diet Versus Optimal Nutrition in ALS showed
some benefit for a high-calorie diet.93
According to the American Academy of Neurology (AAN), a percutaneous
endoscopic gastrostomy tube is recommended in patients with ALS with poor
oral intake to improve survival and stabilize weight loss.81 Although patients
with a forced vital capacity of <50% are at greater risk of aspiration and
postprocedural respiratory failure, a percutaneous endoscopic gastrostomy
insertion is safe with an upright forced vital capacity of ≤50% when performed
by a dedicated team of gastroenterologists and anesthesiologists using monitored
anesthesia care and deep sedation.94
Traditionally, percutaneous endoscopic gastrostomy is the procedure of
choice for patients with ALS. The ProGas (Gastronomy in Patients With
Amyotrophic Lateral Sclerosis) study compared three main methods of
gastrostomy insertion used in patients with ALS95: (1) percutaneous endoscopic
gastrostomy was used in 49% of patients, (2) radiologically inserted gastrostomy
was used in 37% of patients, and (3) perioral image-guided gastrostomy was
used in 13% of patients. Surgical gastrostomy was rarely used. Mortality did not
differ significantly between the three insertion techniques.95 A percutaneous
endoscopic gastrostomy is the optimum method of gastrostomy when
respiratory function is largely unimpaired, and perioral image-guided

gastrostomy is recommended when respiratory function is significantly

compromised in patients with ALS. TABLE 8-8 summarizes the periprocedural
and postprocedural complications of each technique.95
The timing of percutaneous endoscopic gastrostomy placement is also very
important. Patients are unlikely to recover their weight loss if use of the gastrostomy
tube was delayed. Loss of more than 10% of body weight has been associated with
shorter survival time. This may suggest a possible benefit from early gastrostomy
before substantial weight loss occurs, which may not be reversible.95 FIGURE 8-1
shows the survival functions for the different subgroups of patients in terms of
weight loss at gastrostomy compared with weight at the time of diagnosis.
Organophosphate Intoxication
Organophosphates have an increased potential for use in chemical warfare. They
irreversibly inhibit acetylcholinesterases, resulting in excessive amounts of
acetylcholine at the neuromuscular junction and prolonged endplate potential
and desensitization of the postsynaptic membrane. The desensitization is in
large part responsible for the decremental response occurring with
organophosphate intoxication.
Symptoms may begin within 3 hours of exposure and progress to death within
10 hours if the intoxication is severe. The constellation of symptoms can be
divided into muscarinic and nicotinic. Muscarinic symptoms include pupillary
constriction, conjunctival hyperemia, increased bronchial secretion, sweating,
and bradycardia. The muscarinic symptoms are secondary to the inhibition of
acetylcholinesterase at autonomic synapses. The nicotinic effects reflect
inhibition of acetylcholinesterase at the neuromuscular junction. Fasciculations
are initially observed in the eyelids, progress to the face and calves, and then
become generalized. Weakness and paralysis will follow within the next 24 hours.
Respiratory muscles are specifically affected, resulting in respiratory failure.
Severe intoxication can progress to coma and possibly tonic-clonic seizures.
Recovery is expected within 3 to 4 days, and complete recovery is expected
within 1 to 3 weeks depending on the severity of intoxication and effectiveness of
the initial therapy. Mortality is usually secondary to respiratory failure.
TABLE 8-8 Periprocedural and Postprocedural Complications Postgastrostomya
Percutaneous
Patient Endoscopic Radiologically Perioral Image-guided
Characteristics Gastrostomy Inserted Gastrostomy Gastrostomy
SaO2 desaturation 4% 2% 7%
Patient distress 16% 3% 5%
Respiratory arrest 0% 0% 0%
Pain 19% 35% 40%
Pneumonia 3% 4% 16%
a
Data from ProGas Study Group, Lancet Neurol.95
SaO2 = oxygen saturation.
1770 DECEMBER 2018

KEY POINT
● Organophosphates
irreversibly inhibit
acetylcholinesterases,
resulting in excessive
amounts of acetylcholine
at the neuromuscular
junction and prolonged
endplate potential and
desensitization of the
postsynaptic membrane.
FIGURE 8-1
Relationship between gastrostomy placement time and outcome in patients with
amyotrophic lateral sclerosis.
Reprinted with permission from ProGas Study Group, Lancet Neurol.95 © 2017 Elsevier.
Patients should be admitted to the ICU for monitoring of blood pressure, vital
capacity, and neurologic examination. Airway watch and potential need for
mechanical ventilation is one of the main reasons for ICU admissions. Frequent
washing of contaminated skin is very important. Anticonvulsants are advised
despite the low risk of seizures. In case of excessive secretions, atropine is
recommended in doses of 1 mg to 2 mg every hour. Patients should be treated
with the cholinesterase reactivator pralidoxime within a few hours to avoid
phosphorylated acetylcholinesterase becoming resistant to reactivation. The dose
can be repeated if weakness persists after 20 minutes. Potential side effects of
pralidoxime include abdominal discomfort, headaches, dizziness, diplopia,
nervousness, and malaise.96
CONCLUSION
Management of neuromuscular disorders in the neurocritical care unit centers
around prompt identification and management of respiratory failure. It is clear
that ICU management has contributed to improved mortality in the last decade.
As the spectrum has changed, with most neuromuscular disorders in the ICU
being related to ICU-acquired weakness, it is essential that we attempt to
better understand the pathophysiology of these disorders and develop new
therapeutic approaches.
REFERENCES
1 Wijdicks EF. The history of neurocritical care. 2 Caulfield AF, Flower O, Pineda JA, Uddin S.
Handb Clin Neurol 2017;140:3–14. doi:10.1016/ Emergency neurological life support: acute
B978-0-444-63600-3.00001-5. non-traumatic weakness. Neurocrit Care 2017;
27(suppl 1):29–50. doi:10.1007/s12028-017-0450-3.

3 Lacomis D, Petrella JT, Giuliani MJ. Causes 16 Mikaeili H, Yazdchi M, Barazandeh F, Ansarin K.
of neuromuscular weakness in the intensive Euglycemic state reduces the incidence of
care unit: a study of ninety-two patients. critical illness polyneuropathy and duration of
Muscle Nerve 1998;21(5):610–617. doi:10.1002/ ventilator dependency in medical intensive care
(SICI)1097-4598(199805)21:5<610::AID-MUS7>3. unit. Bratisl Lek Listy 2012;113(10):616–619.
0.CO;2-B.
17 Brunner R, Rinner W, Haberler C, et al. Early
4 Maramattom BV, Wijdicks EF. Acute treatment with IgM-enriched intravenous
neuromuscular weakness in the intensive care immunoglobulin does not mitigate critical illness
unit. Crit Care Med 2006;34(11):2835–2841. polyneuropathy and/or myopathy in patients
doi:10.1097/01.CCM.0000239436.63452.81. with multiple organ failure and SIRS/sepsis: a
prospective, randomized, placebo-controlled,
5 Stevens RD, Marshall SA, Cornblath DR, et al.
double-blinded trial. Crit Care 2013;17(5):R213.
A framework for diagnosing and classifying
doi:10.1186/cc13028.
intensive care unit-acquired weakness. Crit Care
Med 2009;37(10 suppl):S299–S308. doi:10.1097/ 18 Sauret JM, Marinides G, Wang GK. Rhabdomyolysis.
CCM.0b013e3181b6ef67. Am Fam Physician 2002;65(5):907–912.
6 Horn J, Hermans G. Intensive care unit-acquired 19 Bosch X, Poch E, Grau JM. Rhabdomyolysis and
weakness. Handb Clin Neurol 2017;141:531–543. acute kidney injury. N Engl J Med 2009;361(1):
doi:10.1016/B978-0-444-63599-0.00029-6. 62–72. doi:10.1056/NEJMra0801327.
7 Latronico N, Bolton CF. Critical illness 20 Greaves I, Porter K, Smith JE, et al. Consensus
polyneuropathy and myopathy: a major cause statement on the early management of crush
of muscle weakness and paralysis. Lancet injury and prevention of crush syndrome.
Neurol 2011;10(10):931–941. doi:10.1016/S1474- J R Army Med Corps 2003;149(4):255–259.
4422(11)70178-8.
21 Sanders DB, Guptill JT. Myasthenia gravis
8 Ali NA, O'Brien JM Jr, Hoffmann SP, et al. and Lambert-Eaton myasthenic syndrome.
Acquired weakness, handgrip strength, and Continuum (Minneap Minn) 2014;20(5 Peripheral
mortality in critically ill patients. Am J Respir Crit Nervous System Disorders):1413–1425.
Care Med 2008;178(3):261–268. doi:10.1164/ doi:10.1212/01.CON.0000455873.30438.9b.
rccm.200712-1829OC.
22 Silvestri NJ, Wolfe GI. Myasthenia gravis. Semin
9 Lacomis D. Electrophysiology of neuromuscular Neurol 2012;32(3):215–226. doi:10.1055/s-0032-
disorders in critical illness. Muscle Nerve 2013; 1329200.
47(3):452–463. doi:10.1002/mus.23615.
23 Carr AS, Cardwell CR, McCarron PO, McConville
10 Hermans G, Van Mechelen H, Bruyninckx F, et al. J. A systematic review of population based
Predictive value for weakness and 1-year epidemiological studies in Myasthenia Gravis.
mortality of screening electrophysiology tests BMC Neurol 2010;10:46. doi:10.1186/1471-2377-10-46.
in the ICU. Intensive Care Med 2015;41(12):
2138–2148. doi:10.1007/s00134-015-3979-7. 24 Wijdicks EF. Management of acute neuromuscular
disorders. Handb Clin Neurol 2017;140:229–237.
11 Schweickert WD, Pohlman MC, Pohlman AS, et al. doi:10.1016/B978-0-444-63600-3.00013-1.
Early physical and occupational therapy in
mechanically ventilated, critically ill patients: 25 Meriggioli MN. Myasthenia gravis with
a randomised controlled trial. Lancet 2009; anti-acetylcholine receptor antibodies. Front
373(9678):1874–1882. doi:10.1016/S0140- Neurol Neurosci 2009;26:94–108. doi:10.1159/
6736(09)60658-9. 000212371.
12 Callahan LA, Supinski GS. Hyperglycemia and 26 Rodolico C, Toscano A, Autunno M, et al. Limb-
acquired weakness in critically ill patients: girdle myasthenia: clinical, electrophysiological
potential mechanisms. Crit Care 2009;13(2):125. and morphological features in familial and
doi:10.1186/cc7728. autoimmune cases. Neuromuscul Disord
2002;12(10):964–969. doi:10.1016/S0960-
13 Hermans G, Schrooten M, Van Damme P, et al.
8966(02)00137-2.
Benefits of intensive insulin therapy on
neuromuscular complications in routine daily 27 Thomas CE, Mayer SA, Gungor Y, et al.
critical care practice: a retrospective study. Myasthenic crisis: clinical features, mortality,
Crit Care 2009;13(1):R5. doi:10.1186/cc7694. complications, and risk factors for prolonged
14 van den Berghe G, Wouters P, Weekers F, et al. intubation. Neurology 1997;48(5):1253–1260.
Intensive insulin therapy in critically ill patients. doi:10.1212/WNL.48.5.1253.
N Engl J Med 2001;345(19):1359–1367. doi:10.1056/ 28 Gilchrist JM, Massey JM, Sanders DB. Single fiber
NEJMoa011300. EMG and repetitive stimulation of the same
15 Van den Berghe G, Wilmer A, Hermans G, et al. muscle in myasthenia gravis. Muscle Nerve 1994;
Intensive insulin therapy in the medical ICU. 17(2):171–175. doi:10.1002/mus.880170207.
N Engl J Med 2006;354(5):449–461. doi:10.1056/ 29 Jacob S, Viegas S, Leite MI, et al. Presence and
NEJMoa052521. pathogenic relevance of antibodies to clustered
acetylcholine receptor in ocular and generalized
myasthenia gravis. Arch Neurol 2012;69(8):
994–1001. doi:10.1001/archneurol.2012.437.
1772 DECEMBER 2018

30 Mandawat A, Kaminski HJ, Cutter G, et al. 44 Díaz-Manera J, Rojas García R, Illa I. Treatment
Comparative analysis of therapeutic options strategies for myasthenia gravis: an update.
used for myasthenia gravis. Ann Neurol 2010; Expert Opin Pharmacother 2012;13(13):
68(6):797–805. doi:10.1002/ana.22139. 1873–1883. doi:10.1517/14656566.2012.
705831.
31 Rabinstein A, Wijdicks EF. BiPAP in acute
respiratory failure due to myasthenic crisis 45 Gronseth GS, Barohn RJ. Practice parameter:
may prevent intubation. Neurology 2002; thymectomy for autoimmune myasthenia gravis
59(10):1647–1649. doi:10.1212/01.WNL. (an evidence-based review): report of the
0000033797.79530.16. Quality Standards Subcommittee of the
American Academy of Neurology. Neurology
32 Flower O, Bowles C, Wijdicks E, et al. Emergency
2000;55(1):7–15. doi:10.1212/WNL.55.1.7.
neurological life support: acute non-traumatic
weakness. Neurocrit Care 2012;17(suppl 1):S79–S95. 46 El-Salem K, Yassin A, Al-Hayk K, et al. Treatment
doi:10.1007/s12028-012-9752-7. of MuSK-associated myasthenia gravis. Curr
Treat Options Neurol 2014;16(4):283. doi:10.1007/
33 Kalita J, Kohat AK, Misra UK. Predictors of outcome
s11940-014-0283-8.
of myasthenic crisis. Neurol Sci 2014;35(7):1109–1114.
doi:10.1007/s10072-014-1659-y. 47 Gilhus NE, Verschuuren JJ. Myasthenia gravis:
subgroup classification and therapeutic
34 Seneviratne J, Mandrekar J, Wijdicks EF,
strategies. Lancet Neurol 2015;14(10):1023–1036.
Rabinstein AA. Predictors of extubation failure
doi:10.1016/S1474-4422(15)00145-3.
in myasthenic crisis. Arch Neurol 2008;65(7):
929–933. doi:10.1001/archneur.65.7.929. 48 Titulaer MJ, Lang B, Verschuuren JJ.
Lambert-Eaton myasthenic syndrome: from
35 Guptill JT, Sanders DB, Evoli A. Anti-MuSK
clinical characteristics to therapeutic strategies.
antibody myasthenia gravis: clinical findings and
Lancet Neurol 2011;10(12):1098–1107. doi:10.1016/
response to treatment in two large cohorts.
S1474-4422(11)70245-9.
Muscle Nerve 2011;44(1):36–40. doi:10.1002/
mus.22006. 49 Schoser B, Eymard B, Datt J, Mantegazza R.
Lambert-Eaton myasthenic syndrome (LEMS):
36 Skeie GO, Apostolski S, Evoli A, et al. Guidelines
a rare autoimmune presynaptic disorder
for treatment of autoimmune neuromuscular
often associated with cancer. J Neurol 2017;
transmission disorders. Eur J Neurol 2010;17(7):
264(9):1854–1863. doi:10.1007/s00415-017-8541-9.
893–902. doi:10.1111/j.1468-1331.2010.03019.x.
50 Wirtz PW, Smallegange TM, Wintzen AR,
37 Gajdos P, Chevret S, Toyka KV. Intravenous
Verschuuren JJ. Differences in clinical features
immunoglobulin for myasthenia gravis. Cochrane
between the Lambert-Eaton myasthenic
Database Syst Rev 2012;12:CD002277.
syndrome with and without cancer: an
doi:10.1002/14651858.CD002277.pub4.
analysis of 227 published cases. Clin Neurol
38 Barth D, Nabavi Nouri M, Ng E, et al. Comparison Neurosurg 2002;104(4):359–363. doi:10.1016/
of IVIg and PLEX in patients with myasthenia S0303-8467(02)00054-9.
gravis. Neurology 2011;76(23):2017–2023. doi:
51 Lennon VA, Kryzer TJ, Griesmann GE, et al.
10.1212/WNL.0b013e31821e5505.
Calcium-channel antibodies in the
39 Gajdos P, Tranchant C, Clair B, et al. Treatment Lambert-Eaton syndrome and other
of myasthenia gravis exacerbation with paraneoplastic syndromes. N Engl J Med
intravenous immunoglobulin: a randomized 1995;332(22):1467–1474. doi:10.1056/
double-blind clinical trial. Arch Neurol 2005; NEJM199506013322203.
62(11):1689–1693. doi:10.1001/archneur.62.11.1689.
52 Evoli A, Liguori R, Romani A, et al. Italian
40 Schneider-Gold C, Gajdos P, Toyka KV, recommendations for Lambert-Eaton
Hohlfeld RR. Corticosteroids for myasthenia myasthenic syndrome (LEMS) management.
gravis. Cochrane Database Syst Rev 2005;(2): Neurol Sci 2014;35(4):515–520. doi:10.1007/
CD002828. doi:10.1002/14651858.CD002828.pub2. s10072-014-1637-4.
41 Palace J, Newsom-Davis J, Lecky B. A randomized 53 McCarty CL, Angelo K, Beer KD, et al. Large
double-blind trial of prednisolone alone outbreak of botulism associated with a church
or with azathioprine in myasthenia gravis. potluck meal–Ohio, 2015. MMWR Morb Mortal
Myasthenia Gravis Study Group. Neurology 1998; Wkly Rep 2015;64(29):802–803.
50(6):1778–1783. doi:10.1212/WNL.50.6.1778.
54 Adalja AA, Toner E, Inglesby TV. Clinical
42 Collongues N, Casez O, Lacour A, et al. management of potential bioterrorism-related
Rituximab in refractory and non-refractory conditions. N Engl J Med 2015;372(10):954–962.
myasthenia: a retrospective multicenter study. doi:10.1056/NEJMra1409755.
Muscle Nerve 2012;46(5):687–691. doi:10.1002/
55 Goodfellow JA, Willison HJ. Guillain-Barré
mus.23412.
syndrome: a century of progress. Nat Rev Neurol
43 Díaz-Manera J, Martínez-Hernández E, Querol L, 2016;12(12):723–731. doi:10.1038/nrneurol.2016.172.
et al. Long-lasting treatment effect of rituximab
56 Wijdicks EF, Klein CJ. Guillain-Barré syndrome.
in MuSK myasthenia. Neurology 2012;78(3):
Mayo Clin Proc 2017;92(3):467–479. doi:10.1016/
189–193. doi:10.1212/WNL.0b013e3182407982.
j.mayocp.2016.12.002.

57 Blázquez AB, Saiz JC. Neurological 71 Fletcher DD, Lawn ND, Wolter TD, Wijdicks EF.
manifestations of Zika virus infection. World J Long-term outcome in patients with
Virol 2016;5(4):135–143. doi:10.5501/wjv.v5.i4.135. Guillain-Barré syndrome requiring mechanical
ventilation. Neurology 2000;54(12):2311–2315.
58 Saiz JC, Vázquez-Calvo Á, Blázquez AB, et al. Zika
doi:10.1212/WNL.54.12.2311.
virus: the latest newcomer. Front Microbiol 2016;
7:496. doi:10.3389/fmicb.2016.00496. 72 McKhann GM, Griffin JW, Cornblath DR, et al.
Plasmapheresis and Guillain-Barré syndrome:
59 Asbury AK, Cornblath DR. Assessment of current analysis of prognostic factors and the effect of
diagnostic criteria for Guillain-Barré syndrome. plasmapheresis. Ann Neurol 1988;23(4):347–353.
Ann Neurol 1990;27(suppl):S21–S24. doi:10.1002/ doi:10.1002/ana.410230406.
ana.410270707.
73 Plasmapheresis and acute Guillain-Barré
60 Lawn ND, Fletcher DD, Henderson RD, et al. syndrome. The Guillain-Barré syndrome study
Anticipating mechanical ventilation in group. Neurology 1985;35(8):1096–1104.
Guillain-Barré syndrome. Arch Neurol 2001;58(6):
893–898. doi:10.1001/archneur.58.6.893. 74 Collins MP, Hadden RD. The nonsystemic
vasculitic neuropathies. Nat Rev Neurol 2017;
61 Walgaard C, Lingsma HF, van Doorn PA, et al. 13(5):302–316. doi:10.1038/nrneurol.2017.42.
Tracheostomy or not: prediction of prolonged
mechanical ventilation in Guillain-Barré 75 Brown RH, Al-Chalabi A. Amyotrophic lateral
syndrome. Neurocrit Care 2017;26(1):6–13. sclerosis. N Engl J Med 2017;377(2):162–172.
doi:10.1007/s12028-016-0311-5. doi:10.1056/NEJMra1603471.
62 Fulgham JR, Wijdicks EF. Guillain-Barré syndrome. 76 Chiò A, Logroscino G, Traynor BJ, et al. Global
Crit Care Clin 1997;13(1):1–15. epidemiology of amyotrophic lateral sclerosis: a
systematic review of the published literature.
63 Fourrier F, Robriquet L, Hurtevent JF, Spagnolo S. Neuroepidemiology 2013;41(2):118–130. doi:
A simple functional marker to predict the need 10.1159/000351153.
for prolonged mechanical ventilation in patients
with Guillain-Barré syndrome. Crit Care 2011;15(1): 77 Forsgren L, Almay BG, Holmgren G, Wall S.
R65. doi:10.1186/cc10043. Epidemiology of motor neuron disease in
northern Sweden. Acta Neurol Scand 1983;68(1):
64 Lawn ND, Wijdicks EF. Post-intubation pulmonary 20–29. doi:10.1111/j.1600-0404.1983.tb04810.x.
function test in Guillain-Barré syndrome.
Muscle Nerve 2000;23(4):613–616. doi:10.1002/ 78 Kiernan MC, Vucic S, Cheah BC, et al.
(SICI)1097-4598(200004)23:4<613:: Amyotrophic lateral sclerosis. Lancet 2011;
AID-MUS21>3.0.CO;2-4. 377(9769):942–955. doi:10.1016/S0140-
6736(10)61156-7.
65 Winer JB, Hughes RA. Identification of patients at
risk of arrhythmia in the Guillain-Barré syndrome. 79 Haverkamp LJ, Appel V, Appel SH. Natural history
Q J Med 1988;68(257):735–739. doi:10.1093/ of amyotrophic lateral sclerosis in a database
oxfordjournals.qjmed.a068239. population. Validation of a scoring system and a
model for survival prediction. Brain 1995;118(pt 3):
66 Roubenoff RA, Borel CO, Hanley DF. 707–719. doi:10.1093/brain/118.3.707.
Hypermetabolism and hypercatabolism in
Guillain-Barré syndrome. JPEN J Parenter Enteral 80 Chen R, Grand'Maison F, Strong MJ, et al. Motor
Nutr 1992;16(5):464–472. doi:10.1177/ neuron disease presenting as acute respiratory
0148607192016005464. failure: a clinical and pathological study. J Neurol
Neurosurg Psychiatry 1996;60(4):455–458.
67 Raphaël JC, Chevret S, Hughes RA, Annane D.
Plasma exchange for Guillain-Barré syndrome. 81 Miller RG, Jackson CE, Kasarskis EJ, et al. Practice
Cochrane Database Syst Rev 2017;2:CD001798. parameter update: the care of the patient with
doi:10.1002/14651858.CD001798. amyotrophic lateral sclerosis: drug, nutritional,
and respiratory therapies (an evidence-based
68 van der Meché FG, Schmitz PI. A randomized trial
review): report of the Quality Standards
comparing intravenous immune globulin and
Subcommittee of the American Academy of
plasma exchange in Guillain-Baré syndrome.
Neurology. Neurology 2009;73(15):1218–1226.
Dutch Guillain-Barré Study Group. N Engl J
doi:10.1212/WNL.0b013e3181bc0141.
Med 1992;326(17):1123–1129. doi:10.1056/
NEJM199204233261705. 82 US Food and Drug Administration. FDA approves
69 Randomised trial of plasma exchange, drug to treat ALS. fda.gov/newsevents/
intravenous immunoglobulin, and combined newsroom/pressannouncements/ucm557102.
treatments in Guillain-Barré syndrome. Plasma htm. Published May 5, 2017. Accessed
Exchange/Sandoglobulin Guillain-Barré October 18, 2018.
Syndrome Trial Group. Lancet 1997;349(9047): 83 Bedlack RS, Pastula DM, Hawes J, Heydt D.
225–230. doi:10.1016/S0140-6736(96)09095-2. Open-label pilot trial of levetiracetam for cramps
70 van Koningsveld R, Schmitz PI, Meché FG, et al. and spasticity in patients with motor neuron
Effect of methylprednisolone when added disease. Amyotroph Lateral Scler 2009;10(4):
to standard treatment with intravenous 210–215. doi:10.1080/17482960802430773.
immunoglobulin for Guillain-Barré syndrome:
randomised trial. Lancet 2004;363(9404):192–196.
doi:10.1016/S0140-6736(03)15324-X.
1774 DECEMBER 2018

84 EFNS Task Force on Diagnosis and Management 90 Bourke SC, Tomlinson M, Williams TL, et al.
of Amyotrophic Lateral Sclerosis, Andersen PM, Effects of non-invasive ventilation on survival
Abrahams S, et al. EFNS guidelines on the clinical and quality of life in patients with amyotrophic
management of amyotrophic lateral sclerosis lateral sclerosis: a randomised controlled trial.
(MALS)–revised report of an EFNS task force. Lancet Neurol 2006;5(2):140–147. doi:10.1016/
Eur J Neurol 2012;19(3):360–375. doi:10.1111/ S1474-4422(05)70326-4.
j.1468-1331.2011.03501.x.
91 Sancho J, Servera E, Díaz JL, et al. Home
85 Qureshi MM, Cudkowicz ME, Zhang H, Raynor E. tracheotomy mechanical ventilation in patients
Increased incidence of deep venous thrombosis with amyotrophic lateral sclerosis: causes,
in ALS. Neurology 2007;68(1):76–77. doi: complications and 1-year survival. Thorax 2011;
10.1212/01.wnl.0000250444.30622.ee. 66(11):948–952. doi:10.1136/thx.2011.160481.
86 Jackson CE, Rosenfeld J, Moore DH, et al. A 92 Genton L, Viatte V, Janssens JP, et al. Nutritional
preliminary evaluation of a prospective study of state, energy intakes and energy expenditure
pulmonary function studies and symptoms of of amyotrophic lateral sclerosis (ALS) patients.
hypoventilation in ALS/MND patients. J Neurol Clin Nutr 2011;30(5):553–559. doi:10.1016/
Sci 2001;191(1–2):75–78. doi:10.1016/S0022-510X(01) j.clnu.2011.06.004.
00617-7.
93 Wills AM, Hubbard J, Macklin EA, et al.
87 Lechtzin N, Wiener CM, Shade DM, et al. Hypercaloric enteral nutrition in patients with
Spirometry in the supine position improves amyotrophic lateral sclerosis: a randomised,
the detection of diaphragmatic weakness in double-blind, placebo-controlled phase 2 trial.
patients with amyotrophic lateral sclerosis. Lancet 2014;383(9934):2065–2072. doi:10.1016/
Chest 2002;121(2):436–442. doi:10.1378/ S0140-6736(14)60222-1.
chest.121.2.436.
94 Kak M, Issa NP, Roos RP, et al. Gastrostomy
88 Velasco R, Salachas F, Munerati E, et al. tube placement is safe in advanced
Nocturnal oxymetry in patients with amyotrophic amyotrophic lateral sclerosis. Neurol Res
lateral sclerosis: role in predicting survival 2017;39(1):16–22. doi:10.1080/01616412.2016.
[in French]. Rev Neurol (Paris) 2002;158(5 pt 1): 1259028.
575–578.
95 ProGas Study Group. Gastrostomy in patients
89 Radunovic A, Annane D, Rafiq MK, Mustfa N. with amyotrophic lateral sclerosis (ProGas): a
Mechanical ventilation for amyotrophic lateral prospective cohort study. Lancet Neurol 2015;
sclerosis/motor neuron disease. Cochrane 14(7):702–729. doi:10.1016/S1474-4422(15)00104-0.
Database Syst Rev 2013;(3):CD004427. doi:
96 Gutmann L, Besser R. Organophosphate
10.1002/14651858.CD004427.pub3.
intoxication: pharmacologic, neurophysiologic,
clinical, and therapeutic considerations. Semin
Neurol 1990;10(1):46–51. doi:10.1055/s-2008-
1041253.

REVIEW ARTICLE

Multimodality Monitoring
ONLINE
in the Neurocritical
Care Unit
By Lucia Rivera Lara, MD, MPH; Hans Adrian Püttgen, MD
ABSTRACT
PURPOSE OF REVIEW: This article focuses on the multiple neuromonitoring
devices that can be used to collect bedside data in the neurocritical care
unit and the methodology to integrate them into a multimodality monitoring
system. The article describes how to apply the collected data to
appreciate the physiologic changes and develop therapeutic approaches
to prevent secondary injury.
RECENT FINDINGS: Theneurologic examination has served as the primary

monitor for secondary brain injury in patients admitted to the neurocritical
care unit. However, the International Multidisciplinary Consensus
CITE AS:
Conference on Multimodality Monitoring in Neurocritical Care concluded
2018;24(6, NEUROCRITICAL CARE): that frequent bedside examinations are not sufficient to detect and
1776–1788. prevent secondary brain injury and that integration of multimodality
monitoring with advanced informatics tools will most likely enhance our
Dr Lucia Rivera Lara, Johns
assessments compared to the clinical examinations alone. This article
Hopkins University School of reviews the invasive and noninvasive technologies used to monitor focal
Medicine, 4940 Eastern Ave, and global neurophysiologic cerebral alterations.
Baltimore, MD 21224,
lriver14@jhmi.edu.
SUMMARY: Multimodal monitoring is still in the early stages of development.
Research is still needed to establish more advanced monitors with the
Dr Rivera Lara has received
research/grant support from bioinformatics to identify useful trends from data gathered to predict
Covidien Ltd/Medtronic; the clinical outcome or prevent secondary brain injury.
Johns Hopkins Anesthesiology
and Critical Care Medicine
(ACCM) Stimulating and
Advancing ACCM Research
(StAAR) award; and Ornim, Inc.
INTRODUCTION
F
Dr Püttgen reports no disclosure. or more than 2 decades, the neurologic examination has served as the
primary monitor for secondary brain injury in patients admitted to
UNLABELED USE OF
the neurocritical care unit. However, experts in this field concluded
USE DISCLOSURE: at the International Multidisciplinary Consensus Conference on
Drs Rivera Lara and Püttgen Multimodality Monitoring in Neurocritical Care that frequent bedside
discuss the unlabeled/
investigational use of examinations are not enough to prevent secondary injury and that integration
near-infrared spectroscopy of multimodality monitoring with next-generation informatics tools will
as a surrogate of cerebral blood
flow to measure cerebral
substantially augment our assessments compared to the conservative
autoregulation. management of serial clinical examinations alone.1
Multimodality monitoring includes the integration of data from multiple
devices to monitor dynamic physiologic changes that may occur after a primary
of Neurology. neurologic injury. The aim of multimodality monitoring is to understand the
1776 DECEMBER 2018

complexity of the changes that lead to secondary brain injury and develop KEY POINTS
therapeutic strategies before the onset of devastating complications, such as
● The aim of multimodality
delayed cerebral ischemia, cognitive impairment, and motor disability. monitoring is to understand
This manuscript focuses on the multiple neuromonitoring devices that can be the complexity of the
used to collect bedside data and the methodology to integrate them into a changes that lead to
multimodality monitoring system. secondary brain injury.
The article describes how to apply the collected data to appreciate the
● The most severe
physiologic changes and develop therapeutic approaches to prevent secondary limitation to the application
injury. The recommendations of the 2014 International Multidisciplinary of cerebral microdialysis in
Consensus Conference on Multimodality Monitoring in Neurocritical Care have bedside decision making is
been summarized in TABLE 9-1,2 and this article outlines each device’s the unfinished work of
defining markers of health
indications, techniques, recommendation strength, and quality of evidence. and crisis in different
clinical contexts.
NEUROMONITORING OF LOCALIZED PHYSIOLOGY
Many devices monitor specific regions of the brain. Some devices are designed to
be placed in certain areas (eg, near-infrared spectroscopy sensors placed on the
forehead to monitor changes in the frontal lobes 1 cm to 2 cm deeper), and
placement location is key for other probes, such as microdialysis and brain tissue
oxygenation, in deciding to monitor physiologic changes in specific areas at risk
for injury.
Cerebral Microdialysis
Introduced in the 1990s, microdialysis allows for analysis of brain parenchyma
physiology through the measurement of levels of metabolic intermediaries.
Clinical application of the technique involves implantation of a catheter
composed of an inlet and outlet joining at a semipermeable membrane tip. This
catheter is then infused with a perfusate solution approximating CSF in
composition, which allows frequent sampling of dialysate and analysis of
molecules of interest as markers of metabolic crisis, most commonly including
glucose, glutamate, lactate, and pyruvate.
Requiring only a small caliber catheter, microdialysis has the advantage of
relative safety in implementation, but the method has important limitations.
Drawing samples and performing bedside analysis demands a significant
investment of clinician time. In addition, episodic collection and evaluation of
dialysate results in low temporal resolution. The volume of analysis is limited to only
a few cubic millimeters. For this reason, the question of placement is of central
importance when developing a strategy of detecting secondary injury with hopes
of intervention. While a perilesional monitor might have better sensitivity for
evidence of cellular distress, placement more remote from the primary injury might
have more direct applicability toward the goal of preserving unaffected brain.
The most severe limitation to the application of cerebral microdialysis in
bedside decision making is the unfinished work of defining markers of health and
crisis in different clinical contexts. The literature suggests that in the context of
acute brain injury (most often traumatic brain injury [TBI] and subarachnoid
hemorrhage [SAH]), findings such as an elevated lactate to pyruvate ratio,
elevated glutamate, and low glucose signal metabolic dysfunction, which, in
turn, correlates to general changes in clinical outcome.3 This approach does not
take into full account the relationship between the astrocyte and neuron with
regard to metabolic support and the use of lactate as an energy substrate,
especially under stresses other than hypoxia.4 More broadly, lack of trials

MULTIMODALITY MONITORING
designed to prospectively demonstrate the value of cerebral microdialysis in

successfully directing clinical care prevents the development of
recommended practices.
Brain Tissue Oxygenation

As disrupted oxygen delivery exacerbates secondary cerebral injury, it stands
to reason that a method for evaluating moment-to-moment brain tissue
oxygenation would prove beneficial in guiding therapy. One such monitoring
technology is the implanted brain tissue oxygen tension (PbtO2) sensor. Despite
requiring implantation, PbtO2 monitors have shown low rates of complications
TABLE 9-1 Summarized Recommendations of the International Multidisciplinary

Consensus Conference on Multimodality Monitoring in Neurocritical Carea
Quality of Prevalence in
Technology Indication Recommendation Evidence Clinical Practice
Intracranial Patients with acute brain injury who are at risk Strong Moderate High
pressure monitors of elevated intracranial pressure based on
clinical or imaging features
Patients with imminent brain herniation to guide Strong High High

therapy
Cerebral Targeting of cerebral perfusion pressure Weak Moderate Emerging

autoregulation management goals and prognostication in
acute brain injury; pressure reactivity has been
commonly used for this purpose, but many
different approaches may be equally valid
EEG Patients with persistent and unexplained Strong Low High

alteration of mental status; convulsive status
epilepticus that does not return to baseline
within 60 minutes of treatment; refractory
status epilepticus; comatose patients after
cardiac arrest during therapeutic hypothermia
and within 24 hours of rewarming
Patients with aneurysmal subarachnoid Weak Low Low

hemorrhage who have unreliable neurologic
examination, at risk for delayed cerebral
ischemia
Jugular venous Patients with or at risk for cerebral ischemia Strong Low Low
bulb oximetry and/or hypoxia
Brain tissue oxygen Patients with or at risk for cerebral ischemia Strong Low Emerging
monitoring and/or hypoxia
Cerebral Patients with or at risk of cerebral ischemia, Strong Low Low

microdialysis hypoxia, energy failure, and glucose deprivation
Thermal diffusion Patients with risk of focal cerebral ischemia Weak Low Low
flowmeter
EEG = electroencephalography.
a
Data from Le Roux P, et al, Neurocrit Care.2
1778 DECEMBER 2018

(hemorrhage, migration, or infection). Once inserted, verification of KEY POINTS
placement through CT scanning is of significant value, as normal values of
● Brain tissue oxygen
PbtO2 depend on the depth of the probe and its proximity to the area of primary tension monitors have
injury. This variation is a result of PbtO2 not simply reflecting perfusion, but shown low rates of
rather the combination of cerebral blood flow, arteriovenous oxygen tension, complication.
and tissue oxygen extraction. Consequently, many factors affect PbtO2
● Many factors affect brain
including cerebral perfusion pressure (CPP), hemoglobin concentration,
tissue oxygen tension
oxygen saturation, metabolic rate (from fever, agitation, or shivering), and including cerebral perfusion
cerebral vasospasm. pressure, hemoglobin
Measurement of brain tissue oxygenation has been studied, for the most part, concentration, oxygen
in patients with SAH and TBI, and in these populations, many observational saturation, metabolic rate
(from fever, agitation, or
studies have shown a close correlation between reduced PbtO2 (below thresholds shivering), and cerebral
varying from 20 mm Hg to 10 mm Hg) and worse outcome either according to vasospasm.
the Glasgow Outcome Scale or mortality.5 A multicenter phase 2 trial (BOOST-II
[Brain Tissue Oxygen Monitoring in Traumatic Brain Injury])6 has been
completed that compares care based on intracranial pressure (ICP) alone to care
guided by these measurements plus PbtO2 monitoring.7,8 The design of this trial
reflects the current understanding of the optimal use of brain tissue oxygen
monitors. Investigators designated placement of the device in normal-appearing
tissue at shallow depth and contralateral to traumatic intracerebral hemorrhage.
PbtO2 below 20 mm Hg or ICP above 20 mm Hg for more than 5 minutes were set
as triggers for interventions including ventilator management, pharmacologic
management of ICP, transfusion, and repositioning. The trial, which was
stopped early after showing a significant result for the primary outcomes of
decreasing the average duration and depth of brain tissue hypoxia, showed
enough of a positive trend in the secondary outcome measures to prompt
a phase 3 trial.
Regional Cerebral Flowmetry

The ability to directly measure cerebral blood flow would serve to unpack the
combined perfusion and tissue extraction data provided by the PbtO2. Two types
of flowmetry sensors seek to provide this type of analysis.
A thermal diffusion flowmeter consists of a solid-state probe with proximal
and distal thermistors combined with a distal heating element. When the heating
element is activated, the relative difference in temperature readings allows one to
deduce heat lost through dissipation into blood flow.
In contrast, a laser Doppler flowmeter directly measures erythrocyte flux. At
this time, laser Doppler flowmeter remains a tool for research limited by an
incomplete understanding of how to apply reported values as well as a strong
sensitivity of the device to outside factors such as ambient light and temperature.9
Thermal diffusion flowmeter, in contrast, has had some clinical exposure
and some amount of comparison to other methods of cerebral blood flow
measurement, such as xenon-enhanced CT. It is nevertheless limited by several
critical factors. First, the device is highly sensitive to positioning and only offers a
very small volume of measurement. Second, the sensor does not retain accuracy
across a range of patient temperatures and cannot function at all above 39°C
(86°F) because of the risk of injury from the heating element. Finally, not enough
data regarding the predictive value of thermal diffusion flowmeter monitoring
or treatment strategies based on its use have been published to validate regular
clinical use.

Near-infrared Spectroscopy
Because living tissue will absorb light in the 700 nm to 950 nm wavelength
band differently based on the level of oxygen saturation in blood, monitors
based on measuring the attenuation of reflected light have long provided
heartbeat-to-heartbeat measurement of peripheral oxygen saturation (SaO2).
More recently, several device manufacturers have adapted this technology into
noninvasive sensors for brain parenchyma oxygenation.
As noninvasive sensors, near-infrared spectroscopy devices have an excellent
safety profile. Spatial resolution is limited by the depth of light penetration and
further reduced by factors such as hair follicle density, skin tone, and skull
thickness. As with many new monitoring technologies, the most significant
obstacle to clinical implementation is a lack of evidence that decisions based on
the output of the device can improve patient outcomes. However, new evidence
has revealed that near-infrared spectroscopy shows promise as a method of
assessing cerebral autoregulation when used in concert with systemic blood
pressure and ICP monitors.10
NEUROMONITORING OF GLOBAL PHYSIOLOGY

Understanding global physiology is key to detecting significant changes that
affect large areas in the brain after TBI, aneurysmal SAH, ischemic stroke,
intracranial hemorrhage, and status epilepticus, among other conditions. These
changes can be related to increased ICP detected by ICP monitors, metabolic
changes detected by continuous EEG, and perfusion changes detected by
cerebral autoregulation monitoring and brain oxygenation.
Electroencephalography
EEG is frequently used in neurocritical care units, not only to detect epileptiform
activity, but also for prognostication after cardiac arrest and prediction of
cerebral vasospasm in patients with aneurysmal SAH. Clinical symptoms of
seizures in the neurocritical care setting can be negligible; therefore, EEG is
indicated to manage clinical seizures with no return to baseline mental status
within 60 minutes of seizure medication (strong recommendation, low-quality
evidence from the 2014 International Multidisciplinary Consensus Conference
on Multimodality Monitoring guidelines).11
Nonconvulsive seizures have been described in 5% to 15% of patients
with acute intracranial injury (eg, TBI, SAH, intracerebral hemorrhage,
postneurosurgery), and nonconvulsive status epilepticus has been described in
5% to 20% of such patients.11 Nonconvulsive seizures are associated with the
same detrimental effects as convulsive seizures, such as toxic concentrations of
intracellular calcium, oxygen free radicals, increases in intracellular osmolality
that produce neuronal swelling, and, ultimately, failure of ATP production. These
effects translate into increased mortality and worse outcomes with mass effect and
worsening of midline shift. Studies with microdialysis have shown toxic levels of
glutamate spikes and an increased lactate to pyruvate ratio in patients with
nonconvulsive seizures.12 However, no adequately powered randomized
controlled trials have demonstrated that treating nonconvulsive seizures or
nonconvulsive status epilepticus improves outcomes. Therefore, according to the
2014 International Multidisciplinary Consensus Conference on Multimodality
Monitoring guidelines, the concern for potential harm from unrecognized and
therefore untreated seizures and the relatively low risk of the EEG need to be
1780 DECEMBER 2018

considered.11 The cost of the procedure may be considerable and should be KEY POINTS
weighed against the potential benefit.
● Near-infrared
EEG is also recommended for patients during therapeutic hypothermia spectroscopy devices have
after cardiac arrest and within 24 hours of rewarming, not only to diagnose an excellent safety profile.
and treat seizures and myoclonic status epilepticus, but also because of its
prognostic factors (strong recommendation, low-quality evidence from the ● Near-infrared
spectroscopy shows
2014 International Multidisciplinary Consensus Conference on Multimodality
promise as a method of
Monitoring guidelines).11 EEG reactivity has been shown to predict recovery of assessing cerebral
consciousness in patients after cardiac arrest, but the converse is also true. In a autoregulation when used in
study of patients receiving continuous EEG, those with nonreactive and concert with systemic blood
discontinuous EEG had a 0% survival rate (positive predictive value of 100%).13 pressure and intracranial
pressure monitors.
One of the caveats with this type of study is the self-fulfilling prophecy, where
clinical decisions are made to discontinue support potentially based on the ● Understanding global
EEG findings. physiology is key to
Another poor prognostic EEG pattern is burst suppression. Quantitative detecting significant
changes that affect large
EEG for the detection of delayed cerebral ischemia after aneurysmal SAH is areas in the brain after
recommended for patients in whom neurologic examination is unreliable traumatic brain injury,
(weak recommendation, low quality of evidence from the 2014 International aneurysmal subarachnoid
Multidisciplinary Consensus Conference on Multimodality Monitoring hemorrhage, ischemic
stroke, intracranial
guidelines).11 Changes in EEG can be seen when cerebral blood flow declines;
hemorrhage and status
the EEG first loses faster frequencies, then as the cerebral blood flow decreases epilepticus, among other
to approximately 17 mL/100 g/min to 18 mL/100 g/min, slower frequencies conditions.
gradually increase.14 A drop in the alpha/delta ratio or in the percent alpha
variability occurs up to 3 days before any clinical or radiographic evidence of ● Nonconvulsive seizures
have been described in 5%
delayed cerebral ischemia.15 to 15% of patients with acute
Some drawbacks to EEG are the high expense and the need for technicians intracranial injury (eg,
to place the EEG leads and for experts to interpret the recordings. traumatic brain injury,
Furthermore, some variability can be seen between expert readers. Newer subarachnoid hemorrhage,
intracerebral hemorrhage,
technologies offer algorithms for seizure detection and remote access for postneurosurgery), and
interpretation that may help small centers that lack the expertise needed to nonconvulsive status
interpret the data. epilepticus has been
described in 5% to 20% of
such patients.
Jugular Venous Bulb Oximetry
One technology used less frequently in global monitoring of the brain is jugular ● EEG is recommended for
venous bulb oximetry, which has several limitations. It has been studied mostly patients during therapeutic
in patients with TBI and is less accurate than other brain oxygenation markers hypothermia after cardiac
arrest and within 24 hours of
(eg, brain tissue oxygen monitor, PbtO2). Moreover, it is invasive and associated
rewarming, not only to
with several complications, such as catheter misplacement, infection, and jugular diagnose and treat seizures
venous thrombosis. The International Multidisciplinary Consensus Conference and myoclonic status
on Multimodality Monitoring guidelines state that if jugular venous bulb epilepticus, but also
oximetry is used for monitoring, it should be part of a multimodal monitoring because of its prognostic
factors.
approach or at least used in combination with ICP monitoring (low-quality
evidence).11 Therefore, if brain oxygenation monitoring is desired, the preferred ● If brain oxygenation
choice is PbtO2. monitoring is desired,
The best candidates for jugular venous bulb oximetry are patients with TBI the preferred choice is
brain tissue oxygen
and global injury. The recognized threshold for ischemia and trigger for an tension sensors.
intervention is less than 55% oxygen saturation. Ischemia can be treated by
augmenting CPP. Elevations in ICP can also cause ischemia, and in those
cases, management should be focused on treating the intracranial
hypertension.

Intracranial Pressure Monitors

ICP monitoring is recommended as part of protocol-driven care for patients with
acute brain injury who are at risk of elevated ICP based on clinical or imaging
features (strong recommendation, moderate quality of evidence from the 2014
International Multidisciplinary Consensus Conference on Multimodality
Monitoring in Neurocritical Care guidelines).2
Two types of the available ICP monitors provide reliable and accurate data.
The gold standard ICP monitor is the ventriculostomy. A major advantage of this
monitor is that it can be rezeroed after placement; therefore, it is not subject to
drift as other ICP monitors are. It can also be therapeutic in patients who have
hydrocephalus, making it the preferred method of ICP monitoring when it is safe
to use. The second recommended ICP monitor is the intraparenchymal monitor,
which is useful in patients with TBI who need only ICP and CPP monitoring and
are not at risk for developing hydrocephalus.2 Other types of ICP monitors, such
as subdural, subarachnoid, and epidural bolts, are less recommended as they are
subject to daily drift, and their ICP measurement is less accurate. Real-time
measurements of ICP with an epidural probe and an intraparenchymal monitor
showed a difference of 4.3 mm Hg (95% confidence interval, ±17 mm Hg).16 In
comparison, real-time measurements of ICP by an intraparenchymal monitor
and a ventriculostomy presented a bias of –1.2 and a 95% confidence interval
of ±6.8 mm Hg.17
Although ICP monitoring has been available since 1951, only one randomized
controlled trial (BEST TRIP [Benchmark Evidence From South American Trials:
Treatment of Intracranial Pressure]) has compared the brain trauma foundation
guidelines protocol based on ICP monitoring to a protocol based on imaging and
clinical examination without monitoring.15 This trial failed to show significant
differences in composite functional and cognitive outcomes at 6 months or in
14-day mortality (primary and secondary outcomes): P=.49 and P=.18,
respectively.18 This trial raised more questions than answers; the two groups
compared two treatment protocols for mass effect (imaging/clinical
examination–based treatment versus ICP-based treatment), demonstrating that
a difference in a neuromonitoring trial is mostly about the type of protocol or
intervention driven by the monitor. One of the main concerns about this trial was
its generalizability, as it was performed in South America, where prehospital and
posthospital care is different (paramedic training and the availability of
prehospital care is limited, as are rehabilitation centers, physical medicine and
rehabilitation doctors, and physical therapists for posthospital care).19
According to the Guidelines for the Management of Severe Traumatic
Brain Injury, ICP monitors are recommended in patients with a Glasgow Coma
Scale score of 8 or less and who have an abnormal head CT (Level II
recommendation).20 It is recommended that their use be restricted to patients
who are severely ill because these monitors are invasive and carry a risk for
hemorrhage, brain tissue lesions, and infection. However, with the development
of noninvasive ICP monitors, this recommendation may change in the future.
Noninvasive Intracranial Pressure Monitors

Noninvasive ICP monitors are still under development and are not yet
recognized as providing an accurate ICP measurement. Noninvasive monitoring
of ICP using transcranial Doppler (TCD) ultrasonography emerged from
observations that some TCD-derived parameters changed during increased ICP,
1782 DECEMBER 2018

such as the shape of the flow velocity pulse waveform and the pulsatility index.20 KEY POINTS
TCD-based methods in their current state are not able to predict mean values of
● Intracranial pressure
ICP with great confidence; the overall accuracy for TCD-based methods is monitoring is recommended
around ±12 mm Hg. Additional studies are still needed to systematically compare as part of protocol-driven
TCD-based methods and validate them across different populations.20 Tympanic care for patients with acute
membrane displacement is another technique that is being developed to measure brain injury who are at risk
of elevated intracranial
ICP. This method is based on the communication of the perilymph and the CSF
pressure based on clinical
via the perilymphatic duct. CSF pressure is transmitted to the perilymph of the or imaging features.
cochlea, affecting the excursion of tympanic membrane and stapes. Indirect
measurement of perilymphatic pressure may be investigated by observing ● The gold standard
tympanic membrane displacement during stapedial reflex contraction.21 This intracranial pressure
monitor is the
technique needs to be developed further as it has several limitations and poor ventriculostomy.
accuracy compared to that of invasive ICP monitoring, with very wide predictive
values of ±25 mm Hg.22 ● Intracranial pressure
monitors are recommended
in patients with a Glasgow
Cerebral Autoregulation Using Multimodal Monitoring Coma Scale score of 8 or
Over the past 2 decades, dynamic cerebral autoregulation monitoring has less and who have an
evolved to allow bedside calculation of optimal CPP and optimal mean arterial abnormal head CT.
pressure (MAP) with invasive and noninvasive methods, respectively. This
● Noninvasive intracranial
evolving technology has shown that the cerebral autoregulation plateau in
pressure monitors are still
patients with acute brain injury is much narrower than was thought half a under development and are
century ago (80 mm Hg to 120 mm Hg versus 50 mm Hg to 150 mm Hg). not yet recognized as
Furthermore, the MAP at the lower limit of autoregulation varies widely from providing an accurate
43 mm Hg to 90 mm Hg.23 intracranial pressure
measurement.
These measurements of optimal CPP and MAP were derived from the secondary
analysis of continuous MAP and CPP data acquired from the hemodynamic ● Dynamic cerebral
monitor (patients need an arterial line) and continuous data from a cerebral blood autoregulation monitoring
flow surrogate. All data are transferred to software installed in a computer that has evolved to allow
bedside calculation of
calculates real-time prespecified time frequency correlations between the two optimal cerebral perfusion
variables and results in an index of autoregulation. Different brands of commercial pressure and optimal
software are available.24 These software brands are restricted for research use mean arterial pressure
only and are not approved by the US Food and Drug Administration (FDA). with invasive and
noninvasive methods.
The technology used to obtain surrogates of cerebral blood flow can be
noninvasive (continuous TCD, near-infrared spectroscopy, and ● The importance of
ultrasound-tagged near-infrared spectroscopy) or invasive (brain tissue oxygen cerebral autoregulation
monitors and ICP monitors). The most accurate cerebral autoregulatory indices multimodal monitoring
to predict outcome in patients with acute TBI are the pressure reactivity index technology is that
calculation of optimal mean
(derived from a correlation between ICP and MAP using 30 consecutive arterial pressure and
10-second windows), the mean velocity index based on CPP (derived from a cerebral perfusion pressure
correlation between cerebral blood flow velocity from the middle cerebral artery would allow physicians to
and CPP using 30 consecutive 10-second windows), and the autoregulatory individualize cerebral
perfusion pressure or mean
reactivity index (a correlation between cerebral blood flow velocity from the arterial pressure goals and
middle cerebral artery and MAP during thigh cuff release or tilt table potentially improve a
declination). These indices have been less studied in other populations with acute patient’s outcomes.
brain injury.
The importance of this cerebral autoregulation multimodal monitoring
technology is that calculation of optimal MAP and CPP would allow physicians to
individualize CPP or MAP goals and potentially improve a patient’s outcomes.
Over the past 7 years, 12 observational studies have determined the feasibility of
using this technology to delineate optimal MAP or optimal CPP at the bedside in

adults with acute TBI, intracerebral hemorrhage, and aneurysmal SAH, as well as
other populations. Of these studies, only three studied the feasibility of calculating
optimal MAP, and 66% (six of nine studies) showed that patients in whom actual
MAP or CPP was widely different from optimal MAP or CPP were more likely
to have an unfavorable outcome.25 The largest study used the pressure reactivity
index to determine optimal CPP in a retrospective study of 327 patients with acute
TBI.26 The authors of this study found that CPP below the optimal level increased
the incidence of fatal outcome, whereas excessively high CPP was associated with an
CASE 9-1 A 52-year-old woman was

brought to the emergency
department in a comatose
state after experiencing a
thunderclap headache. She
had no prior medical history.
Head CT showed an
aneurysmal subarachnoid
hemorrhage (SAH), modified
Fisher Scale grade 4. Her
Hunt and Hess Scale score
was 5 upon admission. She
underwent coiling of a left
posterior inferior cerebellar FIGURE 9-1
artery aneurysm. Mean arterial pressure findings of the patient in
Her cerebral autoregulation CASE 9-1. Graphs showing the optimal mean arterial
pressure defined by noninvasive neuromonitoring
was monitored with
with the cerebral oximetry index derived from
continuous near-infrared near-infrared spectroscopy. A–D, Near-infrared
spectroscopy and daily spectroscopy of days 1 to 4 post–subarachnoid
continuous transcranial hemorrhage. E–F, Follow-up near-infrared
spectroscopy of days 15 to 16 post–subarachnoid
Doppler (TCD) from day 1 to
hemorrhage. Arrows show the optimal mean
day 4 after SAH onset arterial pressure per day defined by the
(FIGURE 9-1A–D). The optimal second-order polynomial formula26 (U-shaped
mean arterial pressure (MAP) curve). The Y axes depict the cerebral oximetry
was 77 mm Hg on day 1 of index, and the X axes depict the clinically
observed mean arterial pressure.
monitoring and increased to
90 mm Hg on day 4. On days
15 to 16 of hospitalization, she
developed severe clinical and sonographic vasospasm in the posterior
circulation, which was confirmed with CT angiography and magnetic
resonance angiography (MRA). Her TCD velocities in the basilar artery
reached 200 cm/s (FIGURE 9-2). The vasospasm involved the left posterior
inferior cerebellar artery, basilar artery, and left posterior cerebral artery.
Her cerebral autoregulation was monitored again, and her autoregulation
curve had moved to the right (FIGURE 9-1E–F). Her optimal MAP on day 16
of hospitalization was 110 mm Hg, 30 points higher than it was on day 1.
The highest intracranial pressure during the monitoring period
was 15 mm Hg.
1784 DECEMBER 2018

increased proportion of severe disability.26 These results highlight the importance of
preventing not only hypoperfusion but also hyperperfusion, which can lead to
worsening cerebral edema, intracranial hypertension, and disability.
Noninvasive continuous cerebral autoregulatory indices have also been
validated against long-standing invasive and noninvasive technology. The cerebral
oximetry index derived from near-infrared spectroscopy was validated against the
TCD-derived mean velocity index in patients who were acutely comatose with
multiple coma etiologies.10 The study showed a moderate correlation between the
The neurocritical care of COMMENT

patients with acute brain injury
is intended to prevent
secondary injury and promote
healing. This patient had a
severe aneurysmal SAH, for
which the most devastating
complication is delayed
cerebral ischemia. Currently,
the management involves
hypertension and hypervolemia
(double H) to augment cerebral
perfusion pressure (CPP) and
MAP, but the targets for those
increments of CPP and MAP are
not defined. Multimodal
monitoring may provide the
technology needed to define
daily cerebral autoregulatory
curves and calculation of
optimal MAP or CPP in the near
future, allowing prevention of
hypoperfusion and
FIGURE 9-2 hyperperfusion.
Transcranial Doppler velocities of the patient in Additional studies are
CASE 9-1. A, Graph showing the augmentation of the
optimal mean arterial pressure (MAP) target
needed to define the efficacy
defined by the near-infrared spectroscopy–derived of these analysis techniques
cerebral oximetry index, the upper and lower (calculation of optimal MAP
limits of autoregulation showing on the extremes with TCD and near-infrared
of the optimal MAP, and the increase in the basilar
artery velocities obtained for daily vasospasm
spectroscopy) to improve
screening (X axes indicate days of monitoring). B, patient outcome, as most of
Graph showing the augmentation of the optimal the studies have been
MAP target defined by transcranial Doppler–derived retrospective, with small
mean velocity index and the increase in the basilar
sample sizes, and limited to
artery velocities obtained for daily vasospasm
screening. patients with acute traumatic
brain injury.

cerebral oximetry index and the mean velocity index and strong correlation and
good agreement in the optimal MAP calculated from each.10 Similarly, some
studies have validated other near-infrared spectroscopy devices, which are mostly
used in Europe, against the pressure reactivity index and have shown that
combining near-infrared spectroscopy with multimodal monitoring techniques
has the potential to deliver noninvasive surrogate measures of autoregulation to
guide therapy (CASE 9-1).27
CLINICAL INFORMATICS INTEGRATION

Real-time calculation of cerebral autoregulation indices as described above gives
a clear glimpse into the future of neurocritical care monitoring. Individual
sensors give an incomplete assessment of patient physiology, and it stands to
reason that, in addition to the growing research, monitoring systems will find
their greatest utility when appreciated in concert with each other and with other
forms of clinical data, including laboratory values, imaging results, and medical
record documentation. If data integration only occurs in the minds of bedside
clinicians, observations of correlation will be ephemeral and limited in scope. The
neurocritical care unit already generates patient data at a substantial rate, but
most hospitals do not have a corresponding means of data acquisition into a time
series database system, meaning that monitoring systems capable of signals in
the kHz range operate under total data loss save the hourly samples recorded in
the medical record (and perhaps the memories clinicians keep of observed
waveforms). Without informatics systems ready to receive data, new sources of
data cannot meet their potential for improving patient care. Acquisition alone, of
course, is merely an exercise in purchasing electronic storage capacity unless
coupled to a means of organizing and integrating data to make it available for
clinical decision making.28
Lack of common formatting standards stands as a significant barrier to true data
integration, which necessitates a patient data management system that translates
monitor output from proprietary formats and merges them into a common
stream.29 A first patient data management system solution was to create a portable
computer with specialized software that could accept multiple monitor outputs
and integrate them for storage and review. This device has the advantages of
expediency and a relatively low cost, but it has the disadvantages of needing to
establish cable connections to begin data processing and the inability to acquire
data from more than one patient without investing in another machine.
Full investment in multimodal monitoring requires placement of a data
management system within the informatics architecture of the hospital. Such an
arrangement acknowledges multimodal monitoring as integral to patient care by
linking a central data management system to other patient-level data servers
(FIGURE 9-3). As a result, clinicians have access to integrated data displays for all
monitored patients in real time as well as for review within the medical record
system. Although several “home grown” informatics solutions may be developed,
commercial systems have reached the market to provide true device and electronic
medical record integration.
As solutions to the more rudimentary problem of data compatibility become
more readily available and as standardized time series database structures allow
for sharing information and experience between institutions, the more
compelling work of establishing clinical decision support tools and treatment
algorithms can begin in earnest.30
1786 DECEMBER 2018

KEY POINTS
● Cerebral perfusion
pressure below the optimal
level increases the
incidence of fatal outcome,
whereas excessively high
cerebral perfusion pressure
is associated with an
increased proportion of
severe disability.
● Monitoring systems will

find their greatest utility
when appreciated in
concert with each other and
with other forms of clinical
data including laboratory
values, imaging results,
and medical record
FIGURE 9-3
documentation.
Schematic representation of health care informatics architecture with an integrated patient
data management system.
● Lack of common
formatting standards stands
as a significant barrier to
true data integration, which
CONCLUSION necessitates a patient data
Multimodal monitoring is still at the early stages of development. Additional management system that
translates monitor output
research into the monitors and the bioinformatics needed to identify useful from proprietary formats
trends from data gathered is critical to advancing this field. We must emphasize and merges them into a
that the monitors themselves contribute little to patients’ outcomes; the main common stream.
outcome drivers will instead be how the data are combined in secondary analysis
with multimodal monitoring, the protocols in place that dictate the management,
and whether therapy is available to treat the physiologic changes detected by
the monitors.
REFERENCES
1 Vespa P, Menon D, Le Roux P, Participants in the 3 Timofeev I, Carpenter KL, Nortje J, et al. Cerebral
International Multi-disciplinary Consensus extracellular chemistry and outcome following
Conference on Multimodality Monitoring. The traumatic brain injury: a microdialysis study of
International Multi-disciplinary Consensus 223 patients. Brain 2011;134(pt 2):484–494.
Conference on Multimodality Monitoring: future doi:10.1093/brain/awq353.
directions and emerging technologies. Neurocrit
4 Bouzat P, Oddo M. Lactate and the injured brain:
Care 2014;21(suppl 2):S270–S281. doi:10.1007/
friend or foe? Curr Opin Crit Care 2014;20(2):
s12028-014-0049-x.
133–140. doi:10.1097/MCC.0000000000000072.
2 Le Roux P, Menon DK, Citerio G, et al. The
5 Oddo M, Bösel J, Participants in the International
International Multidisciplinary Consensus
Multidisciplinary Consensus Conference on
Conference on Multimodality Monitoring in
Multimodality Monitoring. Monitoring of brain
Neurocritical Care: a list of recommendations
and systemic oxygenation in neurocritical care
and additional conclusions: a statement for
patients. Neurocrit Care 2014;21(suppl 2):
healthcare professionals from the Neurocritical
S103–S120. doi:10.1007/s12028-014-0024-6.
Care Society and the European Society of
Intensive Care Medicine. Neurocrit Care 2014; 6 Clinicaltrials.gov. Brain tissue oxygen monitoring
21(suppl 2):S282–S296. doi:10.1007/s12028- in traumatic brain injury (BOOST 2),
014-0077-6. NCT00974259. clinicaltrials.gov/ct2/show/
NCT00974259. Updated January 1, 2016.

7 Okonkwo DO, Shutter LA, Moore C, et al. Brain 19 Chesnut RM, Bleck TP, Citerio G, et al. A
oxygen optimization in severe traumatic brain consensus-based interpretation of the
injury phase-II: a phase II randomized trial. Crit benchmark evidence from South American trials:
Care Med 2017;45(11):1907–1914. doi:10.1097/ Treatment of Intracranial Pressure Trial.
CCM.0000000000002619. J Neurotrauma 2015;32(22):1722–1724. doi:10.1089/
neu.2015.3976.
8 Stocchetti N, Taccone FS, Citerio G, et al.
Neuroprotection in acute brain injury: an up-to- 20 Brain Trauma Foundation, American Association
date review. Crit Care 2015;19:186. doi:10.1186/ of Neurological Surgeons, Congress of
s13054-015-0887-8. Neurological Surgeons, et al. Guidelines for the
management of severe traumatic brain injury. VI.
9 Miller C, Armonda R, Participants in the Indications for intracranial pressure monitoring.
International Multi-disciplinary Consensus J Neurotrauma 2007;24(suppl 1):S37–S44.
Conference on Multimodality Monitoring. doi:10.1089/neu.2007.9990.
Monitoring of cerebral blood flow and ischemia
in the critically ill. Neurocrit Care 2014;21(suppl 2): 21 Robba C, Bacigaluppi S, Cardim D, et al. Non-
S121–S128. doi:10.1007/s12028-014-0021-9. invasive assessment of intracranial pressure. Acta
Neurol Scand 2016;134(1):4–21. doi:10.1111/ane.12527.
10 Rivera-Lara L, Geocadin R, Zorrilla-Vaca A, et al.
Validation of near-infrared spectroscopy for 22 Shimbles S, Dodd C, Banister K, Mendelow AD,
monitoring cerebral autoregulation in comatose Chambers IR. Clinical comparison of tympanic
patients. Neurocrit Care 2017;27(3):362–369. membrane displacement with invasive intracranial
doi:10.1007/s12028-017-0421-8. pressure measurements. Physiol Meas 2005;26(6):
1085–1092. doi:10.1088/0967-3334/26/6/017.
11 Claassen J, Vespa P, Participants in the
International Multi-disciplinary Consensus 23 Joshi B, Ono M, Brown C, et al. Predicting the
Conference on Multimodality Monitoring. limits of cerebral autoregulation during
Electrophysiologic monitoring in acute brain cardiopulmonary bypass. Anesth Analg 2012;
injury. Neurocrit Care 2014;21(suppl 2):S129–S147. 114(3):503–510. doi:10.1213/ANE.0b013e31823d292a.
doi:10.1007/s12028-014-0022-8. 24 Cambridge Enterprise, University of Cambridge
12 Vespa PM, Miller C, McArthur D, et al. website. enterprise.cam.ac.uk/opportunities/
Nonconvulsive electrographic seizures after icm-software-for-brain-monitoring-in-
traumatic brain injury result in a delayed, neurological-intensive-care-research/. Updated
prolonged increase in intracranial pressure and September 19, 2011. October 9, 2018.
metabolic crisis. Crit Care Med 2007;35: 25 Rivera-Lara L, Zorrilla-Vaca A, Geocadin RG, et al.
2830–2836. Cerebral autoregulation-oriented therapy at
13 Rossetti AO, Urbano LA, Delodder F, et al. the bedside: a comprehensive review.
Prognostic value of continuous EEG monitoring Anesthesiology 2017;126(6):1187–1199. doi:10.1097/
during therapeutic hypothermia after cardiac ALN.0000000000001625.
arrest. Crit Care 2010;14(5):R173. doi:10.1186/ 26 Aries MJ, Czosnyka M, Budohoski KP, et al.
cc9276. Continuous determination of optimal cerebral
14 Foreman B, Claassen J. Quantitative EEG for the perfusion pressure in traumatic brain injury. Crit
detection of brain ischemia. Crit Care 2012;16(2): Care Med 2012;40(8):2456–2463. doi:10.1097/
216. doi:10.1186/cc11230. CCM.0b013e3182514eb6.
15 Rots ML, van Putten MJ, Hoedemaekers CW, 27 Highton D, Ghosh A, Tachtsidis I, Panovska-
Horn J. Continuous EEG monitoring for early Griffiths J, et al. Monitoring cerebral
detection of delayed cerebral ischemia in autoregulation after brain injury: multimodal
subarachnoid hemorrhage: a pilot study. assessment of cerebral slow-wave oscillations
Neurocrit Care 2016;24(2):207–216. doi:10.1007/ using near-infrared spectroscopy. Anesth
s12028-015-0205-y. Analg 2015;121(1):198–205. doi:10.1213/
ANE.0000000000000790.
16 Eide PK, Sorteberg W. Simultaneous
measurements of intracranial pressure 28 Flechet M, Grandas FG, Meyfroidt G. Informatics
parameters in the epidural space and in brain in neurocritical care: new ideas for big data. Curr
parenchyma in patients with hydrocephalus. Opin Crit Care 2016;22(2):87–93. doi:10.1097/
J Neurosurg 2010;113(6):1317–1325. doi:10.3171/ MCC.0000000000000287.
2010.7.JNS10483. 29 Hemphill JC, Andrews P, De Georgia M.
17 Koskinen LO, Olivecrona M. Clinical experience Multimodal monitoring and neurocritical care
with the intraparenchymal intracranial pressure bioinformatics. Nat Rev Neurol 2011;7(8):451–460.
monitoring Codman MicroSensor system. doi:10.1038/nrneurol.2011.101.
Neurosurgery 2005;56(4):693–698; discussion 30 Schmidt JM, De Georgia M, Participants in the
693–698. doi:10.1227/01.NEU.0000156609. International Multidisciplinary Consensus
95596.24. Conference on Multimodality Monitoring.
18 Chesnut RM, Temkin N, Carney N, et al. A trial of Multimodality monitoring: informatics,
intracranial-pressure monitoring in traumatic integration data display and analysis. Neurocrit
brain injury. N Engl J Med 2012;367(26):2471–2481. Care 2014;21(suppl 2):S229–238. doi:10.1007/
doi:10.1056/NEJMoa1207363. s12028-014-0037-1.
1788 DECEMBER 2018

Ethical Considerations in ETHICAL AND
MEDICOLEGAL ISSUES
End-of-life Care in the

Face of Clinical Futility
By Joseph S. Kass, MD, JD, FAAN; Ariane Lewis, MD;
ABSTRACT CITE AS:

Management of patients with terminal brain disorders can be medically, CONTINUUM (MINNEAP MINN)
socially, and ethically complex. Although a growing number of feasible 1789–1793.
treatment options may exist, there are times when further treatment can
no longer meaningfully improve either quality or length of life. Clinicians Address correspondence to
and patients should discuss goals of care while patients are capable of Dr Joseph S. Kass, Baylor
College of Medicine, One Baylor
making their own decisions. However, because such discussions can be Plaza M-210, Houston, TX 77030,
challenging, they are often postponed. These discussions are then kass@bcm.edu.
conducted with patients’ health care proxies after patients lose the
capacity to make their own decisions. Disagreements may arise when a Dr Kass serves as associate
patient’s surrogate desires continued aggressive interventions that are editor of ethical and
medicolegal issues for
either biologically futile (incapable of producing the intended Continuum, as an associate
physiologic result) or potentially inappropriate (potentially capable of editor for Continuum Audio, as a
producing the patient’s intended effect but in conflict with the medical neurology section editor of
Ferri’s Clinical Advisor for
team’s ethical principles). This article explores best practices in Elsevier, and as co-editor of
addressing these types of conflicts in the critical care unit, but these Neurology Secrets, Sixth Edition.
concepts also broadly apply to other sites of care. Dr Kass has received personal
compensation for CME lectures
from Pri-Med Medical Group
and has received personal
compensation as a medicolegal
CASE consultant in legal cases
Note: This is a hypothetical case. involving criminal cases,
malpractice, and personal injury.
A 50-year-old man with a history of a left frontal glioblastoma was Dr Lewis has received personal
brought to the emergency department by his wife after she found him compensation as a workshop
experiencing a seizure in bed. He was admitted to the neurocritical care lecturer for the American
Academy of Neurology and for
unit with status epilepticus. grand rounds lectures from
The patient had been diagnosed with a glioblastoma 7 months earlier Mount Sinai Hospital, Mount
after he had a seizure at work. The tumor was resected, and he was treated Sinai Beth Israel Hospital, and
Newark Beth Israel Medical
with radiation and chemotherapy. For the next 5 months, he remained Center, and for work as a legal
functional and continued to work, but 2 months prior to admission, he had consultant for David A. Axelrod
& Associates, PC. Dr Rubin
to stop working after he began to develop right hemiparesis and aphasia.
reports no disclosure.
At his next follow-up visit, the patient insisted on additional surgery but
was informed that he was no longer a candidate for further tumor UNLABELED USE OF
resection. However, believing firmly that surgery would save his life, the USE DISCLOSURE:
patient consulted several other neurosurgeons in an unsuccessful attempt Drs Kass, Lewis, and Rubin
to find a physician willing to offer him his desired intervention. report no disclosures.
On this admission, the patient’s tumor was found to have grown © 2018 American Academy
significantly. Controlling the status epilepticus proved to be difficult and of Neurology.

END-OF-LIFE CARE AND CLINICAL FUTILITY
required lorazepam, fosphenytoin, propofol, and levetiracetam, and he

remained comatose.
The patient’s surrogate decision maker, his wife, pleaded with the
neurocritical care and neurosurgery teams to perform surgery on the
glioblastoma, despite the physicians’ explanation that the tumor’s
progression was relentless and would very soon result in her husband’s
death. The neurocritical care and neurosurgery teams felt that the wife’s
demands were unreasonable on the basis of “futility,” but they debated
whether they were (1) ethically obligated to yield to the wife’s demands or
(2) ethically justified in not offering an intervention that in reasonable
medical probability would not offer the patient any benefit.
DISCUSSION
M
any medical conditions have more than one reasonable
treatment option, but deciding on the most appropriate
treatment option for an individual patient requires shared
decision making between the clinician and the patient. Shared
decision making allows the patient and the clinician to gauge
treatment benefits and harms in the context of the patient’s unique personal,
family, religious, cultural, psychological, and economic circumstances.1 (For
more information on shared decision making, refer to the Practice Issues article
“End-of-life Considerations and Shared Decision Making in Neurocritical Care”
by Christos Lazaridis, MD, EDIC,2 in this issue of Continuum.)
The “best” treatment strategy is fashioned during a conversation in which the
physician listens to the patient express his or her values and helps the patient
analyze how the various medically reasonable treatment options align with these
values.3 In the context of an incurable disease, shared decision making should
ideally help the patient balance mortality benefit (the potential for a longer life)
with morbidity risk (the risk of serious treatment-related suffering).1,4
In this case, the patient pursued clinically appropriate interventions at the
time of his diagnosis, but he continued to believe that surgery was a viable
option, even when it no longer would benefit him. Thus, the wife’s demands for
additional operative interventions when the patient was in the neurocritical
care unit appear to be consistent with her husband’s previous demands. She is
demonstrating the appropriate decision-making standard of a substituted
judgment expected of a surrogate decision maker. She is acting as her husband,
the patient, would have acted had he been able to express his choice. However,
the results the patient had sought while he still possessed decision-making
capacity were unachievable then and are certainly unachievable now considering
his grave deterioration.
Discussing goals of care with patients with terminal brain disorders has an
added layer of complexity, since some patients may appear to possess full
decision-making capacity but may actually have compromised decision-making
capabilities because of the effect of the neurologic disorder or treatment on
cognition. For example, a patient may be alert, interactive, and capable of
engaging in routine conversation, yet that apparent normalcy may mask an
occult impairment in either executive function or impulse control. Depending on
the disorder, the patient may have difficulty with long-term memory, planning,
and abstract thinking and may not be able to reason through risks and benefits
1790 DECEMBER 2018

or regulate emotional responses in the face of high-stakes choices. A significant
suppression of consciousness announces an obvious lack of decisional capacity;
however, an isolated physical deficit such as hemiparesis or mild aphasia may not
prompt the treatment team to assess the status of a patient’s decision-making
capacity. Furthermore, if a patient’s expressed wishes agree with the
recommendations of the provider, clinicians will likely be biased and not
question the patient’s capacity.
In this case, a critical decision about the patient’s future must be made,
yet the patient is comatose and no longer possesses decision-making capacity.
He meets none of the capacity requirements: he is unable to understand his
situation, appreciate his options, reason through the consequences of his decision,
and express his preferences, the very fundamental cognitive tasks underpinning
decision-making capacity.1 (For more information on a patient’s decision-making
capacity, refer to the article “Assessment of Medical Decision-making Capacity in
Patients With Dementia” by Joshua J. Rodgers, MD, and Joseph S. Kass, MD, JD,
FAAN,5 in the 2018 Behavioral Neurology and Psychiatry issue of Continuum.)
Therefore, the patient’s surrogate decision maker, his wife, must now make
decisions about goals of care.6 As the patient’s surrogate, his wife has a fiduciary
obligation to her husband to make decisions as he would have in these changed
circumstances. Although his wife’s voice is important in representing her husband’s
wishes, the treatment team is not ethically obligated to provide every intervention
she demands. The medical team must offer medically reasonable interventions
for this patient with an incurable terminal illness, but how should the treating
physician respond if asked to initiate treatment that in reasonable professional
judgment is deemed either futile or potentially inappropriate?
In 2015, the American Thoracic Society issued an official policy statement that
provides recommendations to guide medical decision-making conflicts arising
in the intensive care unit. In addition to offering recommendations to improve
communication between the patient’s surrogate decision maker and the medical
team, this policy statement, “Responding to Requests for Potentially Inappropriate
Treatments in Intensive Care Units,”7 offers a useful reconceptualization of
requests for interventions that clinicians may find ethically objectionable. Rather
than conceive of all such objectionable requests as “futile,” the guideline reserves
the term futile for that “rare circumstance that an intervention simply cannot
accomplish the intended physiologic goal.” The guideline is clear that physicians
“should not provide futile intervention and should carefully explain the rationale
for the refusal.” However, the guideline uses the term potentially inappropriate
to describe the more common situation where the requested intervention has
“some chance of accomplishing the effect sought by the patient,” but which the
medical team still believes should not be offered because of a conflict with ethical
principles.7 This differentiation between a futile and potentially inappropriate
intervention separates purely technical medical judgments that privilege a
physician’s expertise over the patient’s demands from value-laden judgments that
should not intrinsically privilege the physician’s values over the patient’s values.
These guidelines create a paradigm by which physicians can say that a collection of
interventions is physiologically futile and therefore will not be offered regardless
of the preference of the patient, as communicated through the surrogate
decision maker.
Physicians are called upon to advocate for the plan they feel is most
appropriate and to follow a fair and principled dispute resolution process,

END-OF-LIFE CARE AND CLINICAL FUTILITY
which the 2015 American Thoracic Society guideline lays out in detail, if an
insurmountable impasse is reached.7 In this case, additional surgical intervention
for the glioblastoma is futile because the physiologic goal of surgical resection
is to prevent tumor expansion/invasion and prolong life in a proportion greater
to the risk it imposes. The widespread extension of the tumor and intractable
status epilepticus renders this physiologic goal unobtainable. Although the
patient’s clinical situation is bound to change as the tumor continues to invade
and herniation progresses, currently, other interventions such as continued
intubation and cardiac resuscitation should be considered potentially
inappropriate treatments, rather than futile, in that a decision about their
appropriateness is value laden.
Rather than allow frustration to erode the lines of communication with the
patient’s wife, the neurocritical care team has a number of communication tools
at its disposal in conducting contentious goals-of-care conversations. In this case,
the patient’s wife should first be invited to express her understanding of her
husband’s condition, including her perception of his prognosis, given the
unfortunate dramatic deterioration in her husband’s clinical condition. The team
should hear the wife out before “correcting” her perception with the medical
facts.8 The team may inquire about her motivation for insisting on aggressive
interventions, gauging whether this wish represents incomplete understanding
of the medical facts, fear, anxiety, faith in God’s will, or any number of other
possible motivations.4 The team should strive to determine if the wife is acting as
her husband’s fiduciary, truly representing his wishes under these changed
circumstances, or conflating her own fears of her husband’s impending death
with a misrepresentation of his wishes.1
The team should ask the wife what her husband would have done had he been
able to choose a course of action under these new circumstances. All probing
should employ empathic statements, exploratory questions, and validating
patient-centered responses, since these communication techniques help bond
family and clinicians.6,8 Phrases such as “there is nothing more we can do” or
“withdrawal of care” must be avoided, as they are untrue and engender a sense of
abandonment rather than signal the intended transition from disease-modifying
to palliative care.9 A phrase such as “transition to comfort care only” provides
a clearer message about the medical team’s continued care and concern for the
patient but refocuses the goal of this continued care on palliation of suffering.9 Of
course, the focus of the goals-of-care meeting should be on the patient, but every
effort should be made to recognize the emotional toll the discussion will have
on the decision maker. Including social work, chaplaincy, palliative care, or other
family members in the discussion can help support the decision maker.1
Nonetheless, the decision maker should not be expected to make decisions
during the initial meeting, and another meeting should be planned for the very
near future.1
CASE CONTINUED
The neurocritical care team held a series of meetings using these effective
communication techniques. The active listening approach of the
neurocritical care team helped repair the relationship between the team
and the patient’s wife. She communicated her frustrations and fears,
and the neurocritical care team validated her feelings and did not force her
1792 DECEMBER 2018

into a decision after the first few meetings. Eventually, the patient’s wife
understood the neurocritical care team’s motivation in recommending
palliative therapy, and she consented to changing the goals of care to
palliative care.
CONCLUSION
The shared medical decision-making model inevitably leads to occasions where
the stakeholders have a disagreement over what the appropriate choice may be.
Each individual brings his or her own value systems, knowledge of the disease,
and personal experiences to the conversation. In order to avoid conflicts,
physicians must first understand that we cannot completely separate our own
values and acknowledge when we are making a scientific medical claim and
when we are advocating for a value-laden decision. A patient’s exploration of
these factors is essential and will often require multiple discussions. When an
impasse occurs, using available resources such as palliative care and an ethics
consultation may help bring a resolution, especially if physicians implement
the existing ethics literature that provides guidance on the establishment of
boundaries in medical decision making.
REFERENCES
1 Cai X, Robinson J, Muehlschlegel S, et al. Patient 6 White DB. Rethinking interventions to improve
preferences and surrogate decision making in surrogate decision making in intensive care units.
neuroscience intensive care units. Neurocrit Care Am J Crit Care 2011;20(3):252–257. doi:10.4037/
2015;23(1):131–141. doi:10.1007/s12028-015-0149-2. ajcc2011106.
2 Lazaridis C. End-of-life considerations and 7 Bosslet GT, Pope TM, Rubenfeld DG, et al. An
shared decision making in neurocritical care. official ATS/AACN/ACCP/ESICM/SCCM policy
Continuum (Minneap Minn) 2018;24(6, statement: responding to requests for potentially
Neurocritical Care):1794–1799. inappropriate treatments in intensive care units.
Am J Respir Crit Care Med 2015;191(11):1318–1330.
3 Rubin MA. The collaborative autonomy model of
doi:10.1164/rccm.201505-0924ST.
medical decision-making. Neurocrit Care 2014;
20(2):311–318. doi:10.1007/s12028-013-9922-2. 8 Baile WF, Buckman R, Lenzi R, et al. SPIKES-A
six-step protocol for delivering bad news:
4 Quill TE, Arnold R, Back AL. Discussing treatment
application to the patient with cancer. Oncologist
preferences with patients who want “everything.”
2000;5(4):302–311. doi:10.1634/theoncologist.
Ann Intern Med 2009;151(5):345–349.
5-4-302.
doi:10.7326/0003-4819-151-5-200909010-00010.
9 Pantilat SZ. Communicating with seriously ill
5 Rodgers JJ, Kass JS. Assessment of medical
patients: better words to say. JAMA 2009;301(12):
decision-making capacity in patients with
1279–1281. doi:10.1001/jama.2009.396.
dementia. Continuum (Minneap Minn) 2018;
24(3, Behavioral Neurology and Psychiatry):
920–925. doi:10.1212/CON.0000000000000600.

PRACTICE ISSUES

End-of-life
ONLINE
Considerations and
Shared Decision Making
in Neurocritical Care
By Christos Lazaridis, MD, EDIC
ABSTRACT
The goal of shared decision making in the neurocritical care setting is to
form plans of care that are consistent with best medical practice and are
respectful of the patient’s values. Close cooperation and meaningful
interaction must be achieved with family members so that the patient’s
“person can emerge” through discussions. This article highlights several
caveats that can subvert this complex process, including the cognitive
biases that affect both clinicians and surrogates. Impact, optimism, and
gain-framing biases may be particularly relevant when considering patients
who are receiving neurocritical care. Practitioners need to be cognizant
of the distorting influences of these biases and make attempts to neutralize
them. Quality of survival and the nature and degree of deficits are often
the dominant concerns after patients experience acute severe brain injuries.
Care should be taken to avoid conflating medical facts and value
judgments when discussing prognoses.
CITE AS:
1794–1799.
CASE
Address correspondence to A 60-year-old woman was admitted to the neurocritical care unit with
Dr Christos Lazaridis, University
of Chicago, Departments of
severe traumatic brain injury after being involved in a car accident. Within
Neurology and Surgery the first week of care following admission, she developed refractory
(Neurosurgery), 5841 S Maryland intracranial hypertension despite conventional treatments.
Ave, Chicago, IL 60637,
lazaridis@uchicago.edu.
An urgent family meeting was held. Her surrogates described the patient
as an independent person (she lived alone) who mostly spent her time
RELATIONSHIP DISCLOSURE: reading at home but who would also seek opportunities to travel. She had
Dr Lazaridis serves on the
editorial board of Neurocritical consistently said that she would prefer to be dead than vegetative, but no
Care and has received advance directive existed.
research/grant support from
As the next step in management, the critical care team offered
Cheetah Medical, Inc.
either a decompressive craniectomy or barbiturate coma for control
UNLABELED USE OF of intracranial pressure (ICP); the family was informed about the results
USE DISCLOSURE:
of a recent clinical trial showing that for every 100 patients treated with
Dr Lazaridis reports no decompressive craniectomy rather than barbiturates, there were 22 more
disclosure. survivors; of these, six were in a vegetative state, and the other 16 were
© 2018 American Academy dependent on daily support for combinations of cognitive and physical
of Neurology. disabilities.1
1794 DECEMBER 2018

The patient’s family members disagreed about the best course of action,
with opinions ranging from support of surgery, the use of barbiturates, and
withdrawal of aggressive treatments. Barbiturates were instituted with
temporary ICP control.
One day later, in the face of new refractory ICP, a second family
meeting was held. This time the critical care team stated the opinion that
surgery would be necessary to save her life. The surrogates remained
divided on whether the patient would have chosen continued aggressive
management. After further deliberation, the surrogates concluded that
the patient’s values and interests would be most appropriately served
by transitioning to palliative care, and a consensus was reached to
withdraw life-sustaining treatments.
DISCUSSION
F
ew events, interventions, or “procedures” in the neurocritical care
setting can affect patient outcome as much as the family conference;
most deaths after acute brain injury result from withdrawal of
life-sustaining treatments.2–4 Little attention has been paid to the
conceptual and practical challenges involved in carrying out such
discussions. This article discusses the caveats involved in shared decision making
as they pertain to patients who are neurocritically ill; shared decision making is
the recommended framework for establishing goals of care in the intensive care
unit (ICU). The definition of shared decision making, as endorsed by the
American College of Critical Care Medicine, is “a collaborative process that
allows patients, or their surrogates, and clinicians to make health care decisions
together, taking into account the best scientific evidence available, as well as the
patient’s values, goals, and preferences.”5 Clinicians should first consider the
involved parties and the interests their particular viewpoints generate.
Parties and Interests

Several desired outcomes to a family meeting are possible where quality of
survival and end of life are among the considerations. These outcomes or goals
can be helpfully explored by considering the involved parties: patients, family/
surrogates, clinicians, and the medical system resources (and society at large).
PATIENTS. Patients are the most vulnerable members in this conversation. For
the majority of patients, their voices are represented and their futures are
decided without immediate participation. A central aim is for the patient’s
“person to emerge” as authentically as possible via different sources of evidence
(eg, written records expressing preference or opinion) and testament (eg,
surrogate narratives). This will often require more than relying on an advance
directive (if it exists), since the circumstances described in such a document
are usually not complex enough to address difficult end-of-life considerations
in the face of uncertainty and value-laden judgments of surviving with
neurologic disability. Ideally, if patients were given an opportunity to evaluate
the decisions made, all parties would want them to endorse these decisions as
their own, as representative and expressive of their autonomy.
FAMILIES/SURROGATES. Inapproaching families, clinicians should focus on the

two-dimensional nature of their roles. Families are the advocates of the patient

NEUROCRITICAL CARE END-OF-LIFE CONSIDERATIONS
and the most direct source of the patient’s voice. Concurrently, they are
undergoing a heavily stressful, potentially life-changing, and often traumatic
experience.6,7 A successful family conference is one where their advocacy role is
not only facilitated and augmented but also supports and serves as a healing
process for them. It would be hard to imagine a family meeting as optimal if it
leaves family members alienated and traumatized, even if the decisions genuinely
respect the interests of the patient.
CLINICIANS. The role of the clinician is complex. As the above two points
suggest, the clinician’s duties extend to both patient and surrogates. The
mixture of uncertainty (clinical, empirical, and moral) with the individual
provider and the societal value systems makes the family meeting a great
balancing act requiring knowledge, skill, empathetic understanding, an ability
to be a leader, and, simultaneously, a great level of humility in the face of the
gravity of decisions made.8 Importantly, clinicians should be aware of the
cognitive biases that often operate at an unconscious level yet may influence
behavior and potentially the care provided. A number of these biases are further
explored below.
Communication skills training has been recommended to enhance the ability
of clinicians to appropriately navigate and conduct complex family meetings.5
Such training needs further evaluation to test its efficacy; however, data
suggest that disease-specific communication training can aid discussion of
prognosis and preferences and may beneficially alter clinicians’ interview
styles.9,10
MEDICAL SYSTEM RESOURCES. Decisions in intensive care have obvious

implications for allocation of scarce resources. One-fourth of US health care
spending goes to the 6% of people who die every year; ICUs account for 20% of
all health care costs.11 This tension has a growing potential in the face of an
increasingly aging population. At the same time, premature or forced (perceived
or real) withdrawal of life-sustaining treatment recommendations threaten the
trust of surrogates; in the grander scheme, they risk propagating notions of
neuronihilism and potential prejudice against a life with disability.12
CAVEATS IN PLANS OF CARE

Merely sharing decision making may not accomplish the above goals or avoid
the multiple nuances involved. Below are caveats that should be acknowledged in
the process of forming plans of care; these relate to several cognitive biases
among clinicians and surrogates and the synergy between perceived windows of
opportunity and value-laden judgments.
Biases and Heuristics

Different types of cognitive biases and heuristics have been identified in decision
making13 that can affect shared decision making by distorting the understanding
of the nature of a certain choice or decision and the foreseeable consequences.14
Impact, optimism, and gain-framing biases may be particularly relevant.
Impact bias refers to “Failure to anticipate our remarkable ability to adapt to
new states. People tend to overestimate the long-term impact of both positive
events . . . and negative events.”15 Fear of disability and limited imagination in
terms of the ability to adapt and potentially have a meaningful life can favor
withdrawal of life-sustaining treatments. This relates to the disability paradox, a
1796 DECEMBER 2018

significant underestimation of actual quality of life associated with a certain
disability.16,17 Examples of our limited ability to make evaluative projections in
regard to quality of life include surveys showing a disparity between what is
considered a favorable outcome among healthy adults and actual patients treated
with surgical decompression for space-occupying hemispheric infarction or
patients in a locked-in state. Most patients treated with surgical decompression in
randomized trials were satisfied with the treatment received at 1 year after the
procedure, although most had remained dependent in their activities of daily
living.18,19 A survey of well-being in a cohort of chronically locked-in patients
showed that, although most self-report severe restrictions in community
reintegration, they profess good subjective well-being.20
Clinicians should be cognizant of the optimism bias that has been shown to be
a common response to threatening health information. This is well documented
in the ICU environment where inaccurate interpretations of physicians’
prognostications by surrogates have been shown to arise partly from optimistic
biases rather than simply from misunderstandings.21
A third example introducing distortions in high-stake decision making can
be seen in the loss/gain-framing bias or loss aversion bias; people react to a
particular choice in different ways depending on how it is presented (eg, as a loss
or gain). People tend to avoid risk when a positive frame is presented but seek
risks when a negative frame is presented. An example is a shift of preferences
when presenting the choice of decompressive craniectomy for refractory
intracranial hypertension. If it is framed as a life-saving measure with a risk of
disability, people may prefer to take a risk to save a life. On the other hand, if it
is framed as a gain in controlling intracranial pressure with a high risk for a
vegetative state or severe disability, people may risk-aversely shift their
preference against decompressive craniectomy.
Windows of Opportunity and Value-laden Judgments

Physicians involved in family conferences of patients in the ICU tend to discuss
more about projected quality of life rather than the chances for survival.22 No explicit
data pertaining to patients with brain injuries exist, but it is plausible to think
that quality of survival is the dominant concern to the family and clinical team.
This is particularly applicable to otherwise healthy younger patients with severe
brain injury; it is in such cases that death may be considered the lesser of two evils
as compared to gravely disabled survival by both the family and clinical team.
A way that this has been conceptualized is the so-called window of
opportunity: the opportunity of limiting or withdrawing life-sustaining
treatments before they become unnecessary (eg, mechanical ventilation) in a
patient with devastating cognitive or functional deficits.23,24 Related concerns
could at least partly explain findings showing that decisions to withdraw
life-sustaining treatments leading to death in traumatic brain injury are often
taken very early (>50% within 3 days).4 The subtler confounder here is that, in
the face of uncertainty, there may be a tendency to conflate predictions about
neurologic deficits with predictions about quality of life or patient’s degrees of
acceptance and adaptation to different levels of disability. In a sense, clinicians
are trying to predict the combined outcome, ie, degree of a certain deficit or
disability, and how this is experienced and valued by the individual patient, a
task that may be incrementally difficult as compared to the uncertainties
pervading the two components of prediction (the nature/degree of deficit or

NEUROCRITICAL CARE END-OF-LIFE CONSIDERATIONS
disability and how that is experienced by the individual patient).25 As a result,

clinicians may inadvertently step beyond assessing medical facts to making
questionable value judgments.26 Proper shared decision making ought to
highlight and inform such judgments and allow these decisions to be made in
accordance with the values of the patient.
CONCLUSION
This case highlights some of the uncertainties involved in shared decision making
for patients with severe brain injury. Apart from the epistemic uncertainty
having to do with projected prognosis in terms of functional outcome, there is
also experiential uncertainty related to the quality of survival as well as the
practical uncertainty of required resources. In this case, an open and thorough
dialogue with surrogates led to the limitation of further aggressive measures that
could preserve life yet confer a serious likelihood of severe disability (deemed
most likely to be unacceptable to this patient).
Shared decision making regarding goals of care and the limits of treatment
for patients who are neurocritically ill is a highly nuanced process. The
combination of a nihilistic approach toward brain injury with judgments that
certain disabilities make lives not worth living is conducive to self-fulfilling (and
self-reinforcing) prophecies. At the same time, there are outcomes that many
patients (and their families) would have reasons to deem unacceptable, ie,
outcomes that they would consider a disservice. The inherent, multidimensional
nature of uncertainty in conjunction with pervasive cognitive biases and
differing evaluative standpoints can make the family meeting a daunting
endeavor. Communication training, acknowledging and handling uncertainty,
alertness to numerous cognitive biases, and avoiding a conflation of medical facts
and values when prognosticating are strategies and principles that can enhance
our ability to conduct shared decision making in a way that is respectful to all
parties involved.
ACKNOWLEDGMENT
The author thanks Jenny Blumenthal-Barby, PhD, for her careful reading of the
manuscript and for providing feedback.
REFERENCES
1 Hutchinson PJ, Kolias AG, Timofeev IS, et al. Trial 4 Turgeon AF, Lauzier F, Simard JF, et al. Mortality
of decompressive craniectomy for traumatic associated with withdrawal of life-sustaining
intracranial hypertension. N Engl J Med 2016; therapy for patients with severe traumatic
375(12):1119–1130. doi:10.1056/NEJMoa1605215. brain injury: a Canadian multicentre cohort study.
CMAJ 2011;183(14):1581–1588. doi:10.1503/
2 Mayer SA, Kossoff SB. Withdrawal of life
cmaj.101786.
support in the neurological intensive care unit.
Neurology 1999;52(8):1602–1609. doi:10.1212/ 5 Kon AA, Davidson JE, Morrison W, et al. Shared
WNL.52.8.1602. decision making in ICUs: an American College of
Critical Care Medicine and American Thoracic
3 Becker KJ, Baxter AB, Cohen WA, et al.
Society policy statement. Crit Care Med 2016;44(1):
Withdrawal of support in intracerebral
188–201. doi:10.1097/CCM.0000000000001396.
hemorrhage may lead to self-fulfilling
prophecies. Neurology 2001;56(6):766–772.
doi:10.1212/WNL.56.6.766.
1798 DECEMBER 2018

6 Azoulay E, Pochard F, Kentish-Barnes N, et al. Risk 16 Ubel PA, Loewenstein G, Schwarz N, Smith D.
of post-traumatic stress symptoms in family Misimagining the unimaginable: the disability
members of intensive care unit patients. Am J paradox and health care decision making. Health
Respir Crit Care Med 2005;171(9):987–994. Psychol 2005;24(4S):S57–S62. doi:10.1037/
doi:10.1164/rccm.200409-1295OC. 0278-6133.24.4.S57.
7 Adelman RD, Tmanova LL, Delgado D, et al. 17 Albrecht GL, Devlieger PJ. The disability paradox:
Caregiver burden: a clinical review. JAMA 2014; high quality of life against all odds. Soc Sci
311(10):1052–1060. doi:10.1001/jama.2014.304. Med 1999;48(8):977–988. doi:10.1016/S0277-
9536(98)00411-0.
8 Blumenthal-Barby JS, Lazaridis C. A woman in her
30s whose family becomes distrustful after an 18 Klein A, Kuehner C, Schwarz S. Attitudes in the
initial prognosis proves inaccurate. Chest 2016; general population towards hemi-craniectomy
149(4):e115–e117. doi:10.1016/j.chest.2015.09.021. for middle cerebral artery (MCA) infarction. A
population-based survey. Neurocrit Care 2012;
9 Au DH, Udris EM, Engelberg RA, et al. A
16(3):456–461. doi:10.1007/s12028-012-9677-1.
randomized trial to improve communication
about end-of-life care among patients with 19 Geurts M, van der Worp HB, Kappelle LJ, et al.
COPD. Chest 2012;141(3):726–735. doi:10.1378/ Surgical decompression for space-occupying
chest.11-0362. cerebral infarction: outcomes at 3 years in the
randomized HAMLET trial. Stroke 2013;44(9):
10 Jenkins V, Fallowfield L. Can communication skills
2506–2508. doi:10.1161/STROKEAHA.113.002014.
training alter physicians’ beliefs and behavior
in clinics? J Clin Oncol 2002;20(3):765–769. 20 Bruno MA, Bernheim JL, Ledoux D, et al. A survey
doi:10.1200/JCO.2002.20.3.765. on self-assessed well-being in a cohort of
chronic locked-in syndrome patients: happy
11 Curtis JR, Engelberg RA, Bensink ME, Ramsey
majority, miserable minority. BMJ Open 2011;1(1):
SD. End-of-life care in the intensive care unit:
e000039. doi:10.1136/bmjopen-2010-000039.
can we simultaneously increase quality and
reduce costs? Am J Respir Crit Care Med 2012; 21 Zier LS, Sottile PD, Hong SY, et al. Surrogate
186(7):587–592. doi:10.1164/rccm. decision makers' interpretation of prognostic
201206-1020CP. information: a mixed-methods study. Ann
Intern Med 2012;156(5):360–366. doi:10.7326/
12 Hemphill JC 3rd, White DB. Clinical nihilism in
0003-4819-156-5-201203060-00008.
neuroemergencies. Emerg Med Clin North
Am 2009;27(1):27–37, vii–viii. doi:10.1016/ 22 White DB, Engelberg RA, Wenrich MD, et al.
j.emc.2008.08.009. Prognostication during physician-family
discussions about limiting life support in intensive
13 Blumenthal-Barby JS, Krieger H. Cognitive biases
care units. Crit Care Med 2007;35(2):442–448.
and heuristics in medical decision making: a
doi:10.1097/01.CCM.0000254723.28270.14.
critical review using a systematic search
strategy. Med Decis Making 2015;35(4):539–557. 23 Wilkinson D. The window of opportunity for
doi:10.1177/0272989X14547740. treatment withdrawal. Arch Pediatr Adolesc Med
2011;165(3):211–215. doi:10.1001/archpediatrics.
14 Blumenthal-Barby JS. Biases and heuristics in
2011.4.
decision making and their impact on autonomy.
Am J Bioeth 2016;16(5):5–15. doi:10.1080/15265161. 24 Johnson LSM. Inference and inductive risk in
2016.1159750. disorders of consciousness. AJOB Neurosci 2016;
7(1):35–43. doi:10.1080/21507740.2016.1150908.
15 Tversky A, Griffin D. Endowments and contrast in
judgments of well-being. In: Kahneman D, 25 Lazaridis C. What does it mean to
Tversky A, editors. Choices, values, and frames. neuro-prognosticate? AJOB Neurosci 2016;7(1):
New York, NY: Russell Sage Foundation, 48–50. doi:10.1080/21507740.2016.1146369.
Cambridge University Press, 2000:709–725.
26 Ubel PA. Medical facts versus value
judgments—toward preference-sensitive
guidelines. N Engl J Med 2015;372(26):2475–2477.
doi:10.1056/NEJMp1504245.

PRACTICE ISSUES
Neurocritical Care
Coding for Neurologists
By Marc R. Nuwer, MD, PhD, FAAN; Paul M. Vespa, MD, FAAN
ABSTRACT
CITE AS:
Coding specifies the work performed when providing patient care. Critical
2018;24(6, NEUROCRITICAL CARE): care services mostly use code 99291, and other codes specify additional
1800–1809. time and procedures. Current Procedural Terminology defines critically ill
as “a high probability of imminent or life-threatening deterioration in the
Dr Marc R. Nuwer, University of patient’s condition,” a condition necessary for use of the critical care
California Los Angeles, code. A patient may be critically ill for neurologic reasons even when
Department of Neurology,
Room 1190, Reed Neurological
stable from a cardiorespiratory status. Rules govern who can use these
Research Center, 710 Westwood codes, whether they can be used by more than one physician, the locations
Plaza, Los Angeles, CA 90095, where the code may be used, and what services are included and
MRN@UCLA.edu.
excluded. Physicians need to document the medical necessity of visits and
RELATIONSHIP DISCLOSURE: nature of critical illness or high-risk medical decision making because
Dr Nuwer has received personal auditors may not understand the nature of serious neurologic illness.
compensation for serving as an
honorary editor for Clinical
Neurophysiology; for serving on
the board of directors of
INTRODUCTION
CortiCare, Inc; and for serving on
P
the editorial boards of the hysicians, carriers, hospitals, and regulators use the Current Procedural
Journal of Clinical Terminology (CPT) codes to accurately identify and report their
Neurophysiology and Neurology
Clinical Practice. Dr Nuwer
services. The American Medical Association owns, maintains, and
receives royalties from copyrights the CPT and updates it annually.1 Use of the correct
Cambridge University Press and standardized code set allows physicians, carriers, hospitals, coders,
has received research/grant
funding from the National patients, and their representatives to know what services were provided. Codes
Institutes of Health, Second are available for daily visits as well as for procedures provided in the intensive
Sight, and the United States care unit (ICU). The same codes are used either in a neurocritical care unit or in
Army. Dr Nuwer has given expert
medicolegal testimony in a court any other ICU.
deposition and a court hearing.
Dr Vespa has received personal
compensation as an editor for EVALUATION AND MANAGEMENT CODING
Critical Care Medicine and Use of the critical care daily visit Evaluation and Management code 99291
Neurocritical Care. Dr Vespa has
received personal compensation depends on the severity of illness. Coding policies specify what services should be
as a consultant for Sage coded separately, whose services may be counted as critical care, and how to
Therapeutics, has received
count the time spent providing critical care.
research/grant support from the
National Institutes of Health, and
holds stock options in InTouch
Determining If Patients Are Critically Ill
Health.
To use the critical care visit code 99291 (first 30 to 74 minutes), the patient must
UNLABELED USE OF be critically ill, which must be documented and explained. The phrase critically ill
USE DISCLOSURE:
indicates “a high probability of imminent or life-threatening deterioration in the
Drs Nuwer and Vespa report patient’s condition.”2 The decision making and treatment must include an
no disclosures. indication that the patient’s condition met that definition as well as
© 2018 American Academy considerations and plans to prevent life-threatening deterioration or organ
of Neurology. system failure.3,4
1800 DECEMBER 2018

The threat of central nervous system failure alone can require critical care
services even without risk of cardiorespiratory collapse. Brain instability or
risk of “brain collapse” is as grave a risk to death and quality of life as
cardiorespiratory collapse. The following is a list of the commonly encountered
neurologic disorders that often justify use of critical care codes:
u Acute spinal cord compression or injury

u Acute stroke
u Coma after cardiac arrest
u Coma of unknown etiology
u Guillain-Barré syndrome
u Intracerebral hemorrhage
u Malignant intracranial pressure
u Meningoencephalitis
u Myasthenic crisis
u Neuroleptic malignant syndrome
u Paraneoplastic encephalitis
u Status epilepticus
u Subarachnoid hemorrhage
u Traumatic brain injury
Some patients in the ICU are not critical, such as a patient who is simply on
a respirator. For those cases, use other hospital visit codes. Consulting on a
patient who is critically ill is not necessarily a critical care service (eg, consulting
for altered mental state on a patient in the medical ICU).
The patient’s location is not key. A patient may be in the emergency
department or still on a floor unit and yet be critically ill by the definition given
above. The neurologist may code for critical care if the patient is critically ill
and the neurologist is treating the critical illness. The use of the critical illness
code is dependent on the patient’s critical illness and the physician’s actions to
address the illness. If the neurologist consults for a minor problem on a patient
critically ill for medical reasons in a medical ICU, then that consultation is
routine (ie, not critical care).
Who Can Code for the Service

Different specialties may use critical care codes on the same day as long as they
are not used at the same time of day. The neurologist should use the diagnosis
code for the diagnosis he or she addresses, which may be different from other
reasons for being in the ICU. By using different diagnoses than other providers
seeing the patient, the physician may avoid denials by avoiding the perception
that multiple concurrent physicians are serving duplicate roles.
Only one neurologist may submit critical care code 99291 on a particular
calendar date. That neurologist must have documented at least 30 minutes of
critical care time. Additional time beyond that may be aggregated together with a
second neurologist from the same practice group (ie, one physician may code for
the sum of time of both physicians). Or, the additional time may be separately
submitted by the second neurologist with code 99292 (each additional 30 minutes

NEUROCRITICAL CARE CODING FOR NEUROLOGISTS
of critical care). A physician may not aggregate time with nonphysician

practitioners or residents.
A nurse practitioner, as a nonphysician practitioner, may code for critical care
if that is within his or her scope of practice. A nonphysician practitioner’s time
is counted separately from that of any physician. The nonphysician practitioner’s
time cannot be added into the physician’s time when counting time for patients
with Medicare (ie, the split-shared coding method cannot be used).
For example, if a neurointensivist performs 40 minutes of critical care services
on a patient, and then a nurse practitioner in the same group performs 35 minutes
of additional critical care services on the same patient that day, coding for this
same patient on the same day is as follows:
u 99291 Critical care, evaluation and management of the critically ill or critically injured
patient; first 30 to 74 minutes [under the physician’s name]
u 99292 each additional 30 minutes (list separately in addition to code for primary service)
[under the nurse practitioner’s name]
CPT © 2018 American Medical Association. All rights reserved. CPT is a registered trademark of
the American Medical Association.
Rules for physician assistants vary. Some hospitals, carriers, and states
allow physician assistants to code similar to nurse practitioners for critical
care, and others are more restrictive. Some states, hospitals, and carriers
combine nurse practitioners and physician assistants under the term
nonphysician practitioner.
Documenting Time
Code for the total time the patient received critical care. Include the time spent
on the patient’s unit reviewing test results or imaging studies, time spent
discussing the patient’s case with other medical staff, and time spent
documenting critical care services. Also include the time spent caring for the
patient in the emergency department or in the radiology department while the
patient is in radiology. Activities off the unit away from the patient may not be
reported as critical care time because the physician is not immediately available
to the patient.
Use CPT codes 99291 and 99292 together to report total time. Code 99291
covers the initial 30 to 74 minutes on that day. Code 99292 covers the additional
30-minute time increments. Use 99292 when time totals 75 minutes or more.
When using 99292, documentation should show why time was needed beyond
the first hour. A resident’s time and teaching sessions with residents do not count
toward billing or coding. Coded time is considered the attending physician’s time
spent providing patient care. The following is a summary of neurocritical care
coding guidelines:
u No split-shared services allowed; cannot add together physician plus nonphysician

practitioner time
u Only one provider can code at any specific time of day
u Document the time spent on the unit dealing with patient’s care
u Time can be continuous or intermittent and aggregated for that day
u Must meet minimum time requirements
1802 DECEMBER 2018

u Billing provider’s time counts, not resident’s
u Critical care is based on the patient’s condition
u Location is where the patient is, which could be in the ICU or anywhere else
Family Discussions
Providing routine updates to the family also does not count toward billable time
spent caring for a patient. Time with a surrogate decision maker can count as
critical care time if the patient is unable to give a history or make decisions.
Document the discussion as necessary to determine treatment decisions,
preferably summarizing the conclusions or options discussed.
PROCEDURES AND OTHER SPECIAL CODES

Procedures bundled into CPT codes 99291 and 99292 that do not require separate
coding include the following:
u Blood draw for specimens including for blood gases

u Gastric intubation
u Information data stored in computers (eg, ECGs)
u Interpretation of cardiac output measurements
u Interpretation of chest x-rays
u Pulse oximetry
u Temporary transcutaneous pacing
u Ventilator management
u Vascular access procedures
Include time spent performing these services in the total critical care time.
Other procedure codes may be coded separately (eg, lumbar puncture,
endotracheal intubation, placement of a flow-directed catheter, cardiopulmonary
resuscitation, placement of a ventricular catheter, interpretation of an EEG,
or performance of nerve conduction studies). Exclude the time spent
providing these procedures from the total critical care time. When performing
these other procedures, use modifier 25 with the critical care codes to indicate
that procedures and evaluation and management were performed on the
same day.
Lumbar puncture has three different CPT codes.1 One code is for simple
CSF drainage, such as may be performed for idiopathic intracranial
hypertension. Another code is for blood patch for a CSF leak. The three
CPT codes are1:
u 62270 Spinal puncture, lumbar, diagnostic

u 62272 Spinal puncture, therapeutic, for drainage of CSF (by needle or catheter)
u 62273 Injection, epidural, of blood or clot patch
The physician may code separately for neurodiagnostic and monitoring

procedures for EMG, nerve conduction studies, and EEG. Specific coding advice

for these procedures is beyond the scope of this article. The following monitoring
and emergency procedures also are among those commonly coded separately in
the critical care unit:
u 31500 Intubation, endotracheal, emergency procedure

u 93503 Insertion and placement of flow-directed catheter (eg, Swan-Ganz) for monitoring
purposes
u 92950 Cardiopulmonary resuscitation (eg, in cardiac arrest)
End-of-life care planning is important for patients, which is apparent for

many patients who are admitted to the ICU. To encourage discussion and
preparation of advance care plans, two CPT codes (listed below) allow for coding
of the time spent in these discussions. However, the same physician cannot code
these on the same day as he or she uses CPT code 99291. These codes may be used
on a separate day, such as by the nurse practitioner prior to surgery. While the
codes are often used in office practice, they can be used when providing initial or
subsequent day evaluation and management (eg, code 99233 for subsequent
hospital care per day).
Examples of written advance directives include such forms as the health care
proxy, the durable power of attorney for health care, the living will, and medical
orders for life-sustaining treatment. The two CPT codes for time spent in
discussion and preparation of these forms are:
u 99497 Advance care planning including the explanation and discussion of advance directives
such as standard forms (with completion of such forms, when performed), by the
physician or other qualified health care professional; first 30 minutes, face-to-face with
the patient, family member(s), and/or surrogate
u 99498 each additional 30 minutes (list separately in addition to code for primary procedure)
Prolonged services are used occasionally in the ICU, but they are not used with
the primary codes 99291 and 99292. When a patient is no longer critically ill, the
daily visit should be coded as subsequent day hospital management (eg, 99233).
The base time for this code is 35 minutes. When the unit time spent for that
patient exceeds the base time by more than 30 minutes, then the physician may
add a prolonged service code to identify the additional time spent. For example,
if the patient is no longer critically ill but requires 80 minutes of documented
attending physician unit time, then code 99233 plus 99356. The inpatient
prolonged service CPT codes are:
u 99356 Prolonged service in the inpatient or observation setting, requiring unit/floor time
beyond the usual service; first hour (list separately in addition to code for Evaluation
and Management service)
u 99357 each additional 30 minutes (list separately in addition to code for prolonged service)
1804 DECEMBER 2018

Telemedicine
Consultations over the Internet are now more common than they were a few
years ago. Many commercial payers are required by states to reimburse for
telemedicine. Payers often cover general telemedicine regardless of mandates
because of cost savings by preventing emergency department visits and better
management of chronic conditions.
For most carriers, the majority of telemedicine is coded using the standard
CPT codes plus a modifier. In January 2017, the modifier of choice changed from
GT to 95, although some carriers still prefer to use the old GT modifier. Modifier
95 identifies a “synchronous telemedicine service rendered via a real-time
interactive audio and video telecommunications system.”1 For routine inpatient
and outpatient teleneurology visits, append the modifier to the same CPT code
that otherwise would have been used for that visit. The same documentation
standards apply in those cases, which does limit or preclude some codes that
require extensive physical examinations.
The CPT Appendix P lists codes for which the telemedicine modifier may be
used. The list includes the usual inpatient and outpatient Evaluation and
Management codes, but not the critical care codes 99291 and 99292. Instead, CPT
specifies two codes for telehealth critical care:
u 0188T Remote real-time interactive video-conferenced critical care, evaluation,

and management of the critically ill or critically injured patient; first 30 to
74 minutes
u 0189T each additional 30 minutes (list separately in addition to code for primary service)
These codes recognize the difference between the critical care services that
can be delivered at the bedside with 99291 and the limitation imposed by the
distance. Creation of the two separate codes flags telehealth as a distinct service
rather than as delivering critical care services at a distance. Use modifier 95 with
codes 0188T and 0189T.
When providing telehealth, use Place of Service 02, which is a new place
of service code as of January 2017. Some carriers may not yet recognize
this new place of service code. These codes cover interactive audio-video
telehealth sessions. They do not apply to telephone consultation or
management sessions.
Avoid using simple audio-visual communication that is typical for personal
communication because of the lack of full Health Insurance Portability and
Accountability Act (HIPAA) compliance. The system must allow private,
interactive, two-way audio and visual communication. The website telehealth.
org/video claims to list platforms that are HIPAA compliant and encrypted.
Remember that the physician may need to be licensed in the state where the
patient is hospitalized as well as be privileged in that facility.
Medicare is different than most carriers in that Medicare is much more
restrictive in which patients are eligible for telehealth. Most patients covered
by Medicare Part B are eligible for remote critical care only if they are
hospitalized in a rural area, defined by the United States Census Bureau
as a county outside of a metropolitan statistical area or in a rural health
professional shortage area.5 Medicare uses different codes for remote critical

care. Medicare also makes an exception by covering certain teleconsultations

for acute stroke within 4.5 hours of symptom onset. The critical care telehealth
codes for patients with Medicare are included in the Healthcare Common
Procedure Coding System6:
u G0508 Telehealth consultation, critical care, initial, physicians typically spend 60 minutes
communicating with the patient and providers via telehealth
u G0509 Telehealth consultation, critical care, subsequent, physicians typically spend
50 minutes communicating with the patient and providers via telehealth
The physician might provide a telephone consultation directly with a physician

who is caring for a patient at a remote hospital. Some carriers and contracts
allow payment for this model. These interprofessional telephone or Internet
consultations involve the treating attending or primary physician requesting the
advice of a physician with specific specialty expertise without the need for
face-to-face contact or telehealth directly with the patient. These circumstances
include urgent situations where a timely face-to-face service with the consultant
may not be feasible (eg, because of distance). The codes may not be used if the
consultant has or will see the patient within 14 days. For example, this code is not
to arrange for transfer of care. The time for the service may include review of
records and images if the time consulting with the primary physician is more
than half of the documented time. The code may be used only once per week for
the same patient by the same physician. The CPT codes for interprofessional
telephone/Internet consultations are as follows:
u 99446 Interprofessional telephone/Internet assessment and management service provided

by a consultative physician including a verbal and written report to the patient’s
treating/requesting physician or other qualified health care professional; 5 to 10 minutes
of medical consultative discussion and review
u 99447 11 to 20 minutes of medical consultative discussion and review
u 99448 21 to 30 minutes of medical consultative discussion and review
u 99449 31 minutes or more of medical consultative discussion and review
CPT © 2018 American Medical Association. All rights reserved. CPT is a registered trademark of the
American Medical Association.
The policies, reimbursement, and licensing for telemedicine are actively

evolving. Over the next few years, these codes and rules may change. There is
much interest in expanding telemedicine, so the future of telemedicine seems
bright, especially for population-based care plans that emphasize access to
specialty care.
CASE
A 28-year-old man was brought to the emergency department for urgent
evaluation and management of status epilepticus. Three neurologists who
shared a neurology practice worked to provide critical care for this patient.
The first neurologist spent 40 minutes in the emergency department
assessing and managing this critical patient and initiated therapy to break
the status epilepticus. After his seizures stopped for 20 minutes, the
patient was moved to the neurocritical care unit.
1806 DECEMBER 2018

Subsequently in the neurocritical care unit, the second neurologist
continued to assess for imminent risk of continuing status epilepticus and
searched for any provoking causes. The second neurologist spent a total of
35 minutes assuring that the patient’s seizures did not recur and that his
clinical state did not deteriorate and searched for the cause that provoked
the status epilepticus.
The first neurologist documented 40 minutes of critical care time and
reported code 99291. The second neurologist documented 35 minutes of
critical care time and reported code 99292 for the additional time, noting
that their combined time caring for the patient was 75 minutes. The
emergency department physician coded critical care time earlier that
morning but not simultaneously with the first neurologist.
On the following morning, the second neurologist performed a lumbar
puncture and provided critical care to the patient, who had a fever and two
seizures overnight. Later that day, a third neurologist took over for the
second neurologist and went on to provide additional critical care services.
Together they documented 80 minutes of time for their critical care
services. The second neurologist coded 62270 separately for the lumbar
puncture and excluded the time she took for the lumbar puncture from her
critical care time. The second neurologist coded 99291 for the morning
services. The third neurologist coded 99292 for his afternoon services.
Both neurologists used modifier 25 with the critical care codes to indicate
that the lumbar puncture was a separate procedure.
DISCUSSION
This case presents two examples of the use of codes 99291 and 99292 and how
practice partners can aggregate their time to meet the 75-minute minimum for
use of the two codes. This case shows the codes being used in different locations,
shows coding for a procedure separate from the critical care codes, and
exemplifies the use of the modifier in that circumstance.
AUDITOR ISSUES
Physicians can avoid typical problems that cause downcoding when audited.
Downcoding is when a CPT level of service is changed to a lower level of service.
An auditor may change a 99291 code to a 99233 code. The latter code indicates a
hospital follow-up visit with a high level of care.
That change might happen when an attending physician’s note uses accurate
language in referring to the patient as critically ill with a high probability of
imminent or life-threatening deterioration, but a resident’s note on the same
day makes such a statement as, “stable and may be transferred to the floor
tomorrow.” An auditor will consider that patient stable and therefore no longer
critically ill. The team should be consistent in documentation.
Critical care may be provided on multiple days, even if no changes are made in
the treatment rendered to the patient, provided that the patient’s condition
continues to require the level of attention necessary in critical care. However,
notes simply cut and pasted day after day are discouraged. Auditors frown upon
critical care notes that fail to reflect the work done and thoughts considered on
that day. Notes that fail to change over days give the auditor the impression
that the patient’s condition is stable, even if that is not true. It is best to tailor

notes to the patient’s condition in a given day. Details in the note should show the
work accomplished and planned that day.
Consider that auditors may be poorly educated in neurocritical care. The
neurocritical care patient is at risk of imminent death due to further brain injury,
which can occur unpredictably and rapidly cause multiorgan dysfunction and
death. Auditors are trained in cardiorespiratory critical care, and the clinical
neurosciences are foreign to many of them.
Physicians may need to walk auditors through the high-risk rationale,
although it should be their job to have this knowledge. Emphasis on specific
reasons for neuroprotective strategies in critical care such as mechanical
ventilation, osmolar therapy, temperature management, and similar
interventions may be useful in educating the coders.
Documentation of existing protocols and specific interventions in the
physician’s progress notes may be useful. For example, document a 50% risk of
imminent stroke or death if that is the case for that patient with new-onset
vasospasm. While a physician may prevail when presenting his or her appeal of
an auditor’s downcoding, it is best to act proactively to educate the auditors and
establish documentation of imminent danger. Avoiding the downcoding in the
first place is better than appealing it later.
Auditors may seem ignorant about how high-risk, unstable, and critical
such neurocritical care situations are, such as in cases of vasospasm after
a subarachnoid hemorrhage or when tapering sedation from a patient who is
pharmacologically suppressed to an isoelectric or burst-suppression state to
stop status epilepticus. Document the high risk of failure to carefully treat
the vasospasm or to promptly identify and treat recurrent or refractory
status epilepticus.
This raises the issue of treating the chart rather than treating the patient.
The chart becomes a place for educating auditors. The physician’s time
documenting critical care services in the medical record may be reported as
critical care unit time (eg, time spent documenting the rationale for treatment
and management).
CONCLUSION
Coding specifies what work was performed when providing patient care.
Code 99291 describes the evaluation and management of patients who are
critically ill, which is defined as “a high probability of imminent or
life-threatening deterioration in the patient’s condition.” Documentation
must clarify that the patient is in a critically ill state so that an auditor can
understand the neurologic critical illness. A patient may be critically ill for
neurologic reasons even when stable from a cardiorespiratory status. Other
codes specify additional time and procedures used in the ICU. Physicians can
combine time with other physician practice partners, but not with
nonphysician providers or residents. Only one provider can code at any
specific moment of time. Time spent on the unit dealing with patient’s care
counts toward the critical care codes, and that time can be continuous or
intermittent and added together across that day. Location for critical care
services is wherever the patient is, which could be in the ICU or elsewhere
such as the emergency department.
1808 DECEMBER 2018

REFERENCES
1 American Medical Association. Current 4 Centers for Medicare & Medicaid Services.
procedural terminology (CPT) 2019. Chicago, IL: Medicare claims processing manual. Chapter 12:
American Medical Association Press, 2018. physician/nonphysician practitioners. cms.gov/
Regulations-and-Guidance/Guidance/Manuals/
2 Department of Health & Human Services.
Downloads/clm104c12.pdf. Accessed October 2,
Centers for Medicare & Medicaid Services
2018.
manual system: publication 100-04 Medicare
claims processing transmittal 1548. cms.gov/ 5 United States Census Bureau. Metropolitan
Regulations-and-Guidance/Guidance/ and micropolitan statistical areas. census.gov/
Transmittals/downloads/R1548CP.pdf. Updated programs-surveys/metro-micro.html. Accessed
July 8, 2008. Accessed October 2, 2018. October 2, 2018.
3 Centers for Medicare & Medicaid Services. MLN 6 Centers for Medicare & Medicaid Services.
Matters, Medical Learning Network. Critical care HCPCS coding questions: G-codes. cms.gov/
visits and neonatal care (codes 99291–99292). Medicare/Coding/MedHCPCSGenInfo/HCPCS_
cms.gov/Outreach-and-Education/Medicare- Coding_Questions.html. Updated July 22, 2013.
Learning-Network-MLN/MLNMattersArticles/ Accessed October 2, 2018.
Downloads/MM5993.pdf. Updated July 9, 2008.

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

Postreading Self-Assessment and CME Test and CME Test Responses
Continuum is now using the AAN’s online learning After completing this tally sheet,
center and all Continuum CME activities have moved to please enter your answers online at
continpub.com/CME.
continpub.com/CME. For information on Continuum
Audio CME, go to continuum.audio-digest.org.
You must be a Continuum subscriber or have purchased a NEUROCRITICAL CARE
back issue to earn self-assessment and CME (SA-CME) Vol. 24 No. 6 December 2018
credit for this issue. Institutional access does not include
individual CME. To earn credits, complete the Postreading
Self-Assessment and CME Test in one of three ways: 1 A B C D E 21 A B C D E
1 Go to continpub.com/CME and complete the test online
(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 3 A B C D E 23 A B C D E

answers on the adjacent tally sheet before entering them
online at continpub.com/CME, or 4 A B C D E 24 A B C D E
3 Email ContinuumCME@aan.com for online access if you
5 A B C D E 25 A B C D E
are a nonsubscriber who has a single issue.
Participants have up to 3 years from the date of 6 A B C D E 26 A B C D E

publication to earn AMA PRA Category 1 Credit™ from this
issue. No credit will be awarded for this issue after 7 A B C D E 27 A B C D E
December 31, 2021.
US Participants
The ABPN has reviewed Continuum: Lifelong Learning in
Neurology and has approved this product as part of a 9 A B C D E 29 A B C D E
comprehensive lifelong learning and self-assessment
program, which is mandated by the ABMS as a necessary 10 A B C D E 30 A B C D E
component of maintenance of certification.
US participants may earn up to 20 AMA PRA Category 1 11 A B C D E 31 A B C D E
Credits in SA-CME by completing the Postreading Self-
Assessment and CME Test. AAN members can view and 12 A B C D E 32 A B C D E
download a report of SA-CME credits earned on the
AAN’s NeuroTracker™ on aan.com within 2 business days. 13 A B C D E 33 A B C D E
AAN nonmember subscribers may request a transcript
of credits earned by contacting AAN Member Services at
memberservices@aan.com or (800) 879-1960. 14 A B C D E 34 A B C D E
Canadian Participants 15 A B C D E 35 A B C D E
This program is an Accredited Self-Assessment Program
(Section 3) as defined by the Maintenance of Certification 16 A B C D E 36 A B C D E
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. 18 A B C D E 38 A B C D E
Canadian participants should visit MAINPORT
(www.mainport.org) to record learning and outcomes.
Canadian participants can claim a maximum of 20 hours
(credits are automatically calculated). 20 A B C D E 40 A B C D E
For all questions regarding Continuum CME, email

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

Postreading
SELF-ASSESSMENT
AND CME
Self-Assessment
and CME Test
By Adam G. Kelly, MD; James W. M. Owens Jr, MD, PhD
NEUROCRITICAL CARE
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
December 31, 2021.
CANADIAN PARTICIPANTS: This program is an Accredited Self-Assessment

Program (Section 1) 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: CEREBRAL EDEMA AND ELEVATED INTRACRANIAL PRESSURE
1 Which of the following pathophysiologic processes is most likely

associated with a combination of both vasogenic and cytotoxic
cerebral edema?
A cerebral metastasis from breast cancer

B ischemic stroke
C normal pressure hydrocephalus
D obstructive hydrocephalus
E traumatic brain injury
2 A 43-year-old man is brought to the emergency department after

sustaining injuries from a motorcycle accident. His Glasgow Coma Scale
score is 7 (eyes 2, verbal 2, motor 3). No cervical spine injury is noted on
imaging, although head CT shows some cerebral contusions and
associated edema in the bilateral frontal lobes. Which of the following
best describes the role of corticosteroid treatment in this patient?
A corticosteroids could be given if his Glasgow Coma Scale score

improves to 8 or higher
B corticosteroids improve mortality but do not change the
probability of disability
C corticosteroids should be avoided because of the presence of
hemorrhage
D corticosteroids should be given to all patients with moderate and
severe traumatic brain injury
E no evidence supports corticosteroid treatment in this scenario
3 Routine monitoring of intracranial pressure is typically recommended in

which of the following clinical scenarios?
A bacterial meningitis
B malignant middle cerebral artery infarction
C newly diagnosed glioblastoma
D severe traumatic brain injury
E spontaneous lobar intracerebral hemorrhage
1814 DECEMBER 2018

4 A 56-year-old woman is admitted to the neurocritical care unit with
thrombosis of the superior sagittal sinus. Two days into her hospital
course, she becomes obtunded, and a head CT shows increased edema in
both frontal and parietal lobes. Osmotic therapy is considered. Which of
the following factors would support the use of mannitol as opposed to
hypertonic saline in this patient?
A acute kidney dysfunction

B age older than 50 years
C cerebral sinus thrombosis as a cause of edema
D hypotension
E intravascular volume overload
5 A 23-year-old man is seen in consultation in the intensive care unit for

prognosis following a traumatic brain injury. He had been treated with
high-dose pentobarbital for refractory elevated intracranial pressure,
which had been stopped 18 hours prior to the consult. Intracranial
pressure is now well controlled, but the patient remains unresponsive.
On examination, he does not respond to noxious stimulation, and the
pupillary, corneal, vestibuloocular, and gag reflexes are all absent.
Which of the following is the best next step in management?
A begin hyperventilation to prevent recurrent intracranial pressure

elevation
B continue supportive interventions and serial neurologic
examinations
C perform apnea test
D perform nuclear medicine brain perfusion scan
E withdraw supportive interventions as patient meets criteria for
brain death
ARTICLE 2: INTRACEREBRAL HEMORRHAGE
6 A 54-year-old man presents to the emergency department with acute-

onset aphasia. He has hypertension that has been well controlled for
many years on hydrochlorothiazide. His examination is notable for a
blood pressure of 136/85 mm Hg and expressive aphasia. Brain imaging
shows an acute hemorrhage in the left temporal lobe without associated
ventricular bleeding. Which of the features present in this case increase
the likelihood of an underlying structural abnormality as the cause of
this patient’s hemorrhage?
A blood pressure at time of presentation

B history of hypertension
C lack of intraventricular hemorrhage
D male sex
E temporal lobe location

POSTREADING TEST
7 Which of the following is most accurate regarding the use of the

Intracerebral Hemorrhage Score and other prognostic scores in patients
presenting with acute intracerebral hemorrhage?
A initial examination should always be used, not an examination

following stabilization
B scores can only predict outcomes at time of hospital discharge
C scores may be associated with a self-fulfilling prophecy and
premature limitations in care
D scores only predict mortality, not functional outcomes
E scores rely on variables that are not readily available clinically
8 A 79-year-old woman presents to the emergency department with

sudden-onset headache and visual dysfunction. Examination shows a
left homonymous hemianopsia, and CT shows an acute hemorrhage in
the right temporal-parietal region. She takes warfarin for atrial
fibrillation, and her international normalized ratio (INR) is 2.4. In
addition to prothrombin complex concentrate, which of the following
should also be administered to this patient?
A andexanet alfa
B emergent hemodialysis
C fresh frozen plasma
D idarucizumab
E vitamin K
9 A 67-year-old man presents to the emergency department for acute-

onset right-sided weakness, with brain imaging showing an acute
hemorrhage in the left basal ganglia. He has a history of coronary artery
disease for which he takes aspirin. He is admitted to the intensive care
unit for blood pressure control; no acute neurosurgical interventions are
planned in his case. Which of the following best describes the role of
platelet transfusion in this case?
A platelets should be transfused

B platelets should be transfused only if his blood pressure cannot
be controlled below 140/90 mm Hg
C platelets should be transfused only if his hemorrhage volume is
more than 50 cc
D platelets should be transfused only if his platelet count is less than
150,000 per mL
E platelets should not be transfused
1816 DECEMBER 2018

ARTICLE 3: SUBARACHNOID HEMORRHAGE
10 Which of the following imaging modalities is most sensitive for

subacute or chronic subarachnoid hemorrhage?
A head CT with contrast

B head CT without contrast
C susceptibility-weighted imaging (SWI) of the brain
D T1-weighted MRI of the brain
E time-of-flight magnetic resonance angiography
11 Which of the following factors favors surgical clipping over

endovascular coiling in the treatment of an unsecured aneurysm after
subarachnoid hemorrhage?
A middle cerebral artery location of aneurysm

B multiple medical comorbidities
C older age of the patient
D poor clinical grade
E top of the basilar artery location of aneurysm
12 A 51-year-old man in the step-down unit is asymptomatic 10 days after

presentation with a nontraumatic subarachnoid hemorrhage isolated to
the perimesencephalic cisterns. No source of the subarachnoid
hemorrhage was found on two six-vessel digital subtraction
angiograms performed 1 week apart, and the patient has had an
uncomplicated hospital course. What is the most appropriate next step
in the management of this patient?
A CT angiography of head and neck

B discharge home
C repeat cerebral digital subtraction angiography
D repeat MRI of brain and cervical spine
E transfer to medical floor
13 In a patient with an aneurysmal subarachnoid hemorrhage who is not

comatose and does not have clinically evident seizures, when may
prophylactic anticonvulsant treatment be discontinued?
A 2 weeks
B 6 months
C 1 year
D when an EEG reveals no epileptiform activity
E when the aneurysm is secured

POSTREADING TEST
14 A 45-year-old man in the neurocritical care unit 5 days after a

subarachnoid hemorrhage develops an expressive aphasia. He is
afebrile, and an emergent EEG reveals no epileptiform activity. His
glucose, sodium, and PaO2 are unchanged as compared to earlier in the
day, when he was asymptomatic. Which of the following prophylactic
treatments, when initiated at initial diagnosis of aneurysmal
subarachnoid hemorrhage, is most useful in decreasing the risk of
poor functional outcome from this complication?
A apixaban
B aspirin
C hemodilution
D intravascular hypervolemia
E nimodipine
15 A 28-year-old man who is clinically stable in the neurocritical care unit

2 days after aneurysmal subarachnoid hemorrhage develops confusion
followed by a decreasing level of consciousness, impaired upgaze, and
increased mean arterial pressure. Which of the following tests is most
useful in identifying the likely cause for these clinical changes?
A digital subtraction angiography

B echocardiography
C EEG
D head CT without contrast
E measurement of serum sodium
ARTICLE 4: MANAGEMENT OF STROKE IN THE NEUROCRITICAL CARE UNIT
16 A 62-year-old man presents with a left middle cerebral artery stroke

and a National Institutes of Health Stroke Scale (NIHSS) score of 15.
Having no contraindications, he is treated with IV recombinant tissue
plasminogen activator with improvement in his NIHSS score to 5.
However, he neurologically worsens 6 hours later, and a head CT
reveals an intracranial hemorrhage. His fibrinogen level is found to be
low. Administration of which of the following medications is most
likely to be beneficial?
A activated factor VII

B aminocaproic acid
C cryoprecipitate
D idarucizumab
E vitamin K
1818 DECEMBER 2018

17 A 70-year-old man with a history of atrial fibrillation, for which he
takes dabigatran, presents with a right middle cerebral artery stroke
and undergoes mechanical thrombectomy with a significant
improvement in his National Institutes of Health Stroke Scale (NIHSS)
score. He experiences a recrudescence of his symptoms and worsening
obtundation. Head CT reveals hemorrhage in the right middle cerebral
artery territory and edema. Which of the following agents is most likely
to decrease the risk of hemorrhage expansion?
A activated factor VII

B aminocaproic acid
C idarucizumab
D tranexamic acid
E vitamin K
18 Which of the following clinical factors is most predictive of the need for
critical care intervention in a patient with an acute ischemic stroke?
A age
B female sex
C history of diabetes mellitus
D National Institutes of Health Stroke Scale (NIHSS) score
E systolic blood pressure
19 A 52-year-old man with a left middle cerebral artery ischemic stroke

occupying 75% of the arterial territory experiences significantly
worsening clinical status 36 hours after symptom onset. Head CT
reveals worsening cerebral edema with mass effect. Which of the
following interventions is most likely to reduce the risk of death in
this patient?
A hemicraniectomy
B IV mannitol
C IV recombinant tissue plasminogen activator
D mechanical embolectomy
E suboccipital decompressive craniectomy
20 Which of the following medications is useful in the treatment of

post–recombinant tissue plasminogen activator angioedema?
A fresh frozen plasma

B IV furosemide
C IV methylprednisolone
D oral lisinopril
E oral spironolactone

POSTREADING TEST
ARTICLE 5: STATUS EPILEPTICUS, REFRACTORY STATUS EPILEPTICUS, AND

SUPER-REFRACTORY STATUS EPILEPTICUS
21 A 25-year-old man develops right-sided hemiclonic jerking that quickly

progresses to a generalized convulsion. He receives one dose of IV
lorazepam en route to the emergency department and then another
dose in the emergency department 10 minutes later. After another
10 minutes, clear convulsive activity continues unabated. Which of
the following medications should be administered next?
A IV valproic acid
B ketamine infusion
C midazolam infusion
D pentobarbital infusion
E propofol infusion
22 A 37-year-old woman presented to the emergency department with a

flurry of brief convulsive seizures, which then became continuous and
persisted despite treatment with IV lorazepam, IV valproic acid, and
continuous IV infusion of midazolam and pentobarbital titrated to a
burst-suppression pattern. Now, 24 hours after she was admitted to the
intensive care unit, which of the following therapies is most likely to
be beneficial?
A allopregnanolone
B increased midazolam infusion dose
C IV phenobarbital
D ketogenic diet
E therapeutic hypothermia
23 A 12-year-old girl is admitted to the intensive care unit with altered

mental status of unclear etiology. She has a history of diabetes mellitus,
controlled with insulin, but has no history of epilepsy. No convulsive
seizure activity was noted prior to admission. On examination, she is
now unresponsive. Which of the following factors in this case is
significantly associated with an increased likelihood of finding seizures
on EEG?
A absence of a history of epilepsy

B age younger than 18 years
C female sex
D history of diabetes mellitus
E lack of convulsive seizures before onset of coma
1820 DECEMBER 2018

24 Which of the following factors is associated with worse functional
outcomes for patients in status epilepticus?
A age older than 60 years

B convulsive status epilepticus
C history of epilepsy
D lack of need for intubation
E male sex
ARTICLE 6: COMA AND BRAIN DEATH
25 Which of the following medications has been shown in a randomized

controlled trial to accelerate recovery of consciousness after traumatic
brain injury?
A amantadine
B fluoxetine
C methylphenidate
D pramipexole
E zolpidem
26 Which of the following causes of coma is generally associated with the

poorest prognosis regardless of treatment?
A anoxic brain injury

B autoimmune encephalitis
C drug intoxication
D refractory status epilepticus
E traumatic brain injury
27 A 36-year-old previously healthy man sustained a severe traumatic

brain injury in a motor vehicle accident 4 months ago. Initially entirely
unresponsive with his eyes persistently closed, he now has periods of
eye opening during which he will, at times, although not consistently,
track objects with his eyes. He does not follow commands or express
himself in any way detectable by bedside testing. Which of the
following classifications best fits the patient’s current condition?
A brain dead
B coma
C minimally conscious state minus
D minimally conscious state plus
E vegetative state/unresponsive wakefulness syndrome

POSTREADING TEST
28 A patient suspected to be dead by neurologic criteria undergoes an apnea

test. Before the ventilator is disconnected, the patient is normothermic,
with a PaCO2 of 40 mm Hg, a PaO2 of 210 mm Hg, and a systolic blood
pressure of 120 mm Hg. In which of the following scenarios would the
test be appropriately interpreted as confirming apnea?
A no respiratory movements except agonal gasps after 2 minutes

B no respiratory movements except agonal gasps after 8 minutes
C no respiratory movements of any kind with a PaCO2 of 65 mm Hg
D no respiratory movements of any kind with a PaO2 of 55 mm Hg
E no respiratory movements of any kind with a systolic blood
pressure of 70 mm Hg and falling
ARTICLE 7: MANAGEMENT OF COMATOSE SURVIVORS OF CARDIAC ARREST
29 A 69-year-old woman with a known history of cardiomyopathy

collapses and is found by paramedics to be in ventricular fibrillation.
She is successfully resuscitated after 20 minutes. She is unresponsive
on hospital arrival and is placed on a temperature management
protocol targeting a core temperature of 35°C (95°F). Which of the
following factors best describes important aspects of her temperature
management treatment?
A electrolyte disturbances and immunosuppression are known

complications
B goal temperature should be maintained for 1 week postarrest
C rewarming should be done quickly to minimize complications
D shivering does not need to be controlled; it is an important
component of the therapeutic benefit of hypothermia
E temperature management should start 48 to 72 hours postarrest
30 A 46-year-old man experiences a cardiac arrest and is successfully

resuscitated after 25 minutes without a pulse. He is treated with
targeted temperature management and awakens after 72 hours with
grossly normal motor function. His wife and children ask about the
possibility that he will experience difficulty with memory and about
his ability to return to work as a schoolteacher. Which of the following
is the most appropriate counseling of this patient’s family?
A cognitive dysfunction is almost certain and likely irreversible

B cognitive dysfunction is common in survivors of cardiac arrest and
may be amenable to rehabilitation
C he is unlikely to have cognitive dysfunction given his age
D he is unlikely to have cognitive dysfunction given his brief duration
of arrest
E he is unlikely to have cognitive dysfunction given his reassuring
motor examination
1822 DECEMBER 2018

31 Which of the following best describes the impact of sedation and
anesthesia on the prognostic capabilities of absent N20 responses on
somatosensory evoked potentials following resuscitation from cardiac
arrest?
A N20 responses are affected by even light sedation

B N20 responses are not affected by high-dose benzodiazepine
infusion
C N20 responses are not affected by sedation or anesthesia
D N20 responses can be predictive in patients receiving sedation
but are discouraged with suppressive levels of
sedation/anesthesia
E N20 responses should be checked prior to beginning targeted
temperature management
32 A 36-year-old woman is resuscitated following a cardiac arrest with an

estimated pulseless time of 30 minutes. After undergoing targeted
temperature management for 48 hours, she begins experiencing
rhythmic episodes of left-sided twitching suggestive of seizures; in
addition, her EEG shows other electrical events consistent with
nonconvulsive seizures. Which of the following best describes the
prognostic and therapeutic implications of this patient’s seizures and
EEG findings?
A her convulsive seizures should be treated, but nonconvulsive

events do not require treatment
B the presence of nonconvulsive events is highly predictive of a
poor neurologic outcome in this case
C seizures should be considered a treatable form of secondary brain
injury postarrest and should be treated
D seizures show preservation of brain electrical function and should
be considered a positive prognostic sign
E treatment of either seizure type (convulsive, nonconvulsive) will
not alter her prognosis
ARTICLE 8: CRITICAL CARE OF NEUROMUSCULAR DISORDERS
33 For a patient with amyotrophic lateral sclerosis, which of the following

measures is the best predictor of respiratory failure?
A erect forced vital capacity

B peak cough expiratory flow
C seated expiratory tidal volume
D sniff nasal pressure
E supine forced vital capacity

POSTREADING TEST
34 A 10-year-old girl is brought to the emergency department with

weakness and worsening respiratory effort. She had been playing in an
old shed on her grandfather’s farm about an hour before initial
symptoms of sweating and coughing were noted. She then developed
“twitching” of her eyelids followed by twitching in the muscles of
her face. In the emergency department she is normothermic and
normotensive with a heart rate of 48 beats/min and shallow
respirations. She is diaphoretic with abundant oral secretions. Which
of the following interventions is most likely to be helpful for this
patient?
A corticosteroids
B IV immunoglobulin (IVIg)
C plasma exchange
D pralidoxime
E pyridostigmine
35 Which of the following best describes the pattern and distribution of

symptoms seen in patients with intensive care unit–acquired weakness?
A facial muscles are often preferentially involved

B ocular muscles are most commonly spared
C respiratory muscles are rarely involved
D weakness is most commonly focal
E weakness is usually accompanied by hyperreflexia
36 A 34-year-old woman with a diagnosis of acetylcholine receptor–

positive myasthenia gravis is hospitalized after developing ptosis,
double vision, severe swallowing dysfunction, and 4/5 weakness in
her proximal arms and legs. These symptoms began 1 week after she
stopped all medications because of cost issues. Her respiratory function
is stable based on subjective report and bedside pulmonary function
testing. Which of the following best describes treatment options
available for this patient?
A acute thymectomy should be performed with or without presence

of thymoma
B corticosteroids are contraindicated at this time
C IV immunoglobulin (IVIg) and plasma exchange are equally
effective in this scenario
D IVIg and plasma exchange used together would provide additional
benefit
E pyridostigmine monotherapy should be used first in this patient
1824 DECEMBER 2018

ARTICLE 9: MULTIMODALITY MONITORING IN THE NEUROCRITICAL CARE
UNIT
37 A 48-year-old woman is admitted to the neurocritical care unit

following a subarachnoid hemorrhage from a ruptured middle cerebral
artery aneurysm. A cerebral microdialysis catheter is placed. Which of
the following findings on dialysate analysis is most suggestive of
ongoing metabolic dysfunction?
A elevated lactate to pyruvate ratio

B high glucose
C high pyruvate
D low glutamate
E low lactate
38 Changes in quantitative EEG findings, such as a drop in the alpha/delta

ratio or drop in the percent of alpha variability, may be predictive of
which of the following events following aneurysmal subarachnoid
hemorrhage?
A brain death
B delayed cerebral ischemia
C nonconvulsive status epilepticus
D recurrent aneurysmal bleeding
E uncal herniation
39 A 51-year-old man is admitted to the neurocritical care unit with an

aneurysmal subarachnoid hemorrhage. Head CT shows evidence of
ventricular enlargement, and elevated intracranial pressure is
suspected. Which of the following types of intracranial pressure
monitors is most indicated in this patient?
A epidural bolt monitor

B intraparenchymal monitor
C noninvasive transcranial Doppler monitor
D subdural bolt monitor
E ventriculostomy

POSTREADING TEST
40 A 67-year-old man develops a cardiac arrest and is resuscitated after

15 minutes without a pulse. He is treated with therapeutic hypothermia
to 34°C (93°F) for 24 hours and is then rewarmed slowly per
institutional protocol. Which of the following forms of neurologic
monitoring is recommended during this patient’s rewarming period?
A cerebral microdialysis
B EEG
C implantable brain tissue oxygenation monitoring
D jugular venous bulb oximetry
E near-infrared spectroscopy
1826 DECEMBER 2018

Postreading
SELF-ASSESSMENT
AND CME
Self-Assessment
and CME Test—Preferred
Responses
By Adam G. Kelly, MD; James W. M. Owens Jr, MD, PhD
NEUROCRITICAL CARE
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 December 31, 2021.
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: CEREBRAL EDEMA AND ELEVATED INTRACRANIAL PRESSURE
1 The preferred response is E (traumatic brain injury). Cerebral edema is usually

divided into three main categories: vasogenic, cytotoxic, and hydrostatic edema.
Vasogenic edema arises from increased permeability of the blood-brain barrier
and is most commonly seen in brain tumors (intraaxial or extraaxial, primary, or
metastatic) or cerebral abscesses. Cytotoxic edema develops because of fluid
shifts from the extracellular to intracellular compartment and is typical of
cerebral ischemia. Hydrostatic edema results from the altered flow of CSF, most

POSTREADING TEST—PREFERRED RESPONSES
commonly in obstructive hydrocephalus, and causes transependymal

displacement of CSF. Traumatic brain injury and intracerebral hemorrhage are
most commonly associated with a combination of cytotoxic and vasogenic
edema. Normal pressure hydrocephalus is not typically associated with cerebral
edema. For more information, refer to page 1589 of the Continuum article
“Cerebral Edema and Elevated Intracranial Pressure.”
2 The preferred response is E (no evidence supports corticosteroid treatment in

this scenario). In patients with severe traumatic brain injury (defined as a
Glasgow Coma Scale score of 8 or less), no evidence currently supports the
routine use of corticosteroids. The CRASH (Corticosteroid Randomisation After
Significant Head Injury) trial showed increased mortality in patients with severe
brain injury treated with methylprednisolone compared to those treated with
placebo. This, along with other data, has led the Brain Trauma Foundation and
other guideline-writing organizations to recommend against the use of
corticosteroids in patients with severe traumatic brain injury. For more
information, refer to pages 1589–1590 of the Continuum article “Cerebral Edema
and Elevated Intracranial Pressure.”
3 The preferred response is D (severe traumatic brain injury). Measurement of

intracranial pressure on a continuous basis is indicated in certain clinical
scenarios, and even when measured, treatment of elevated intracranial pressure
should be individualized to the specific patient situation. Of the choices listed
here, intracranial pressure monitoring is indicated most often in a patient with
severe traumatic brain injury (defined as a Glasgow Coma Scale score of 8 or
less). Monitoring could be considered in select patients with some of the other
listed disorders, but routine monitoring is not recommended in these patients.
For more information, refer to pages 1593–1594 of the Continuum article
“Cerebral Edema and Elevated Intracranial Pressure.”
4 The preferred response is E (intravascular volume overload). Osmotic therapy,

in the form of either mannitol or hypertonic saline, is commonly employed in the
treatment of cerebral edema. Either can be used as a treatment alone or as a
bridge to definitive treatment such as decompressive surgery for focal lesions.
Little evidence supports the widespread use of one form of osmotic treatment
over the other, and, as such, decisions on whether to use mannitol or hypertonic
saline should be individualized to the specific patient. As a diuretic, mannitol is
relatively contraindicated in cases of acute kidney injury and hypotension;
however, it is the preferred option in patients with concomitant fluid overload
who may benefit from diuresis. No clear evidence supports the preference of one
strategy over another based on patient age or underlying pathophysiology. For
more information, refer to pages 1590–1591 of the Continuum article “Cerebral
Edema and Elevated Intracranial Pressure.”
1828 DECEMBER 2018

5 The preferred response is B (continue supportive interventions and serial
neurologic examinations). In patients with refractory intracranial pressure
elevation, pentobarbital (as well as other anesthetics such as propofol) can be
used in select patients. The half-life of pentobarbital is very long (up to 50 hours
in some patients); therefore, any prognostic decision making should be
postponed until providers can be more confident that the effects of
pentobarbital are not causing concerning examination findings. Because the
intracranial pressure is now controlled, the apparent loss of neurologic function
is more likely to be attributable to residual pentobarbital effects rather than
progression to brain death. For this reason, clinical brain death testing with an
apnea test or perfusion scan is not indicated. Hyperventilation may have a very
transient intracranial pressure lowering effect, but prophylactic hyperventilation
should be avoided. For more information, refer to page 1594 of the Continuum
article “Cerebral Edema and Elevated Intracranial Pressure.”
ARTICLE 2: INTRACEREBRAL HEMORRHAGE
6 The preferred response is E (temporal lobe location). One of the important

aspects of intracerebral hemorrhage management is determining whether a
hemorrhage is a primary event (eg, related to hypertension, amyloid angiopathy)
or due to an underlying structural lesion such as a vascular malformation.
Although the presence or absence of any single factor is not definitive in making
this distinction, some characteristics are associated with a higher risk of an
underlying vascular anomaly, which include female sex, presence of primary
intraventricular bleeding, absent history of hypertension or coagulopathy, and
lobar hemorrhage location in younger patients. In this patient’s case, his
temporal lobe hemorrhage at age 54 is concerning for an underlying vascular
lesion, and vascular imaging is warranted. For more information, refer to
page 1604 of the Continuum article “Intracerebral Hemorrhage.”
7 The preferred response is C (scores may be associated with a self-fulfilling

prophecy and premature limitations in care). The Intracerebral Hemorrhage
Score, the Functional Outcome in Patients With Primary Intracerebral Hemorrhage
(FUNC) score, and other similar tools can be used to estimate prognosis following
intracerebral hemorrhage and can be helpful in assessing severity, guiding
treatment decisions, and counseling patients and families. Positive aspects
include their ability to predict both mortality and functional outcomes at a variety
of meaningful time points (30 days, 90 days) and their use of relevant clinical
variables that are easily obtained (including in the use of the Intracerebral
Hemorrhage Score and the Glasgow Coma Scale following medical stabilization).
Important limitations of these tools include their generation of an average, or
expected, outcome from a population of patients as opposed to a definitive
prognosis for an individual patient and concerns that predictions of poor
outcomes result in early limitations of high intensity treatments with subsequent
increases in mortality (the so-called self-fulfilling prophecy). For more information,
refer to pages 1604–1605 of the Continuum article “Intracerebral Hemorrhage.”

8 The preferred response is E (vitamin K). This patient presents with an acute brain
hemorrhage in the setting of oral anticoagulation with warfarin. She is at risk
for hematoma expansion (and associated worsening of clinical outcomes) if
her anticoagulation is not reversed. Factor replacement with prothrombin
complex concentrate or fresh frozen plasma should be given (with preference
to prothrombin complex concentrate, given more rapid correction of the INR)
along with IV vitamin K. Idarucizumab is a reversal agent for dabigatran but has
no role in warfarin-associated hemorrhage. Andexanet alfa has been recently
cleared by the US Food and Drug Administration (FDA) for reversal of novel oral
anticoagulant–induced severe hemorrhages. Broader availability is expected
in 2019. Neither andexanet alfa nor hemodialysis would be indicated in this
case. For more information, refer to pages 1608–1609 of the Continuum article
“Intracerebral Hemorrhage.”
9 The preferred response is E (platelets should not be transfused). This patient

presents with a probable hypertensive hemorrhage in the setting of antiplatelet
medication use. The PATCH (Platelet Transfusion Versus Standard Care After
Acute Stroke Due to Spontaneous Cerebral Haemorrhage Associated With
Antiplatelet Therapy) trial did not show improved outcomes or positive effects
on the risk of hematoma expansion with the routine use of platelet transfusions
in patients with brain hemorrhage who were taking antiplatelet agents, and
severe adverse effects were more common in patients treated with transfusion.
Except in cases where neurosurgical intervention is likely to be needed, platelet
transfusion should not be used in patients with intracerebral hemorrhage who
are otherwise being treated with conservative supportive management. For
more information, refer to page 1610 of the Continuum article “Intracerebral
Hemorrhage.”
ARTICLE 3: SUBARACHNOID HEMORRHAGE
10 The preferred response is C (susceptibility-weighted imaging [SWI] of the

brain). Subacute or chronic subarachnoid hemorrhage is best detected by a
hemosiderin-sensitive MRI sequence such as SWI. CT without contrast is the
most appropriate test to perform for an acute subarachnoid hemorrhage. Once
a subarachnoid hemorrhage has been diagnosed, blood vessel imaging is useful
in identifying the source, but this is better accomplished with digital subtraction
angiography or CT angiography. For more information, refer to page 1627 of the
Continuum article “Subarachnoid Hemorrhage.”
11 The preferred response is A (middle cerebral artery location of aneurysm).

A middle cerebral artery location of the aneurysm favors surgical clipping using
a temporal craniotomy approach, especially when it is located more distally.
Factors associated with poor surgical risk, such as older age and multiple
medical comorbidities, favor endovascular treatment as would a top of the
1830 DECEMBER 2018

basilar artery location for the aneurysm. For more information, refer to Table 3-5
of the Continuum article “Subarachnoid Hemorrhage.”
12 The preferred response is B (discharge home). A patient with a perimesencephalic

subarachnoid hemorrhage and no source found with repeat digital subtraction
angiography may be discharged after approximately 8 to 10 days. Further
observation on the medical floor is not necessary in this circumstance as the
patient’s course had remained uncomplicated. For more information, refer to
pages 1630–1631 of the Continuum article “Subarachnoid Hemorrhage.”
13 The preferred response is E (when the aneurysm is secured). Prophylactic

anticonvulsant treatment should be continued in a patient with aneurysmal
subarachnoid hemorrhage who does not have seizures and with an examination
that can be followed until the aneurysm is secured and for no more than 3 to
7 days. For more information, refer to page 1641 of the Continuum article
“Subarachnoid Hemorrhage.”
14 The preferred response is E (nimodipine). Prophylactic administration of

nimodipine, a calcium channel blocker, improves functional outcome, although
it does not prevent angiographic vasospasm. Intravascular euvolemia (not
hypervolemia) is also important in preventing delayed cerebral ischemia.
Hemodilution is no longer recommended because of associated adverse
outcomes. Antiplatelet agents and anticoagulation are not indicated for delayed
cerebral ischemia prevention following subarachnoid hemorrhage. For more
information, refer to page 1645 of the Continuum article “Subarachnoid
Hemorrhage.”
15 The preferred response is D (head CT without contrast). This patient is most

likely developing acute symptomatic hydrocephalus, a condition most rapidly
diagnosed by head CT in the context of suggestive clinical symptoms. EEG
could reveal subclinical seizures that could be associated with altered
mentation but would not be expected to produce a sustained increase in blood
pressure or impaired upgaze. Poor cardiac output and hyponatremia could also
produce decreasing level of consciousness but would not persistently impair
upgaze or increase arterial blood pressure. For more information, refer to
page 1640 of the Continuum article “Subarachnoid Hemorrhage.”
ARTICLE 4: MANAGEMENT OF STROKE IN THE NEUROCRITICAL CARE UNIT
16 The preferred response is C (cryoprecipitate). In the context of hemorrhage

following recombinant tissue plasminogen activator administration,
cryoprecipitate can be quickly administered to normalize plasma fibrinogen

levels (goal of >150 mg/dL). Aminocaproic acid is an antifibrinolytic agent that

can be used to prevent hematoma expansion. Activated factor VII and vitamin K
are used to reverse warfarin. Idarucizumab is a specific reversal agent for
dabigatran. For more information, refer to page 1671 and Case 4-2 of the
Continuum article “Management of Stroke in the Neurocritical Care Unit.”
17 The preferred response is C (idarucizumab). Idarucizumab reverses the effect

of dabigatran. Aminocaproic acid and tranexamic acid are antifibrinolytic
agents that may help prevent hematoma expansion. Activated factor VII and
vitamin K can be used to reverse warfarin. For more information, refer to
page 1674 and Table 4-4 of the Continuum article “Management of Stroke in the
Neurocritical Care Unit.”
18 The preferred response is E (systolic blood pressure). Of the factors listed,

systolic blood pressure above 140 mm Hg is a predictor of the need for
critical care intervention. Other such factors include male sex, a higher NIHSS
score, older age, and hyperlipidemia. For more information, refer to
page 1659 of the Continuum article “Management of Stroke in the
19 The preferred response is A (hemicraniectomy). A hemicraniectomy is most

likely to reduce the risk of death in this patient. He is outside the window for
thrombolysis or embolus retrieval. A suboccipital decompressive craniectomy
is useful for cerebellar strokes involving 25% to 33% of the cerebellar
hemisphere. Osmotic therapy may be temporarily helpful in this setting but is
much less likely to make a significant difference in terms of outcome as
compared to hemicraniectomy. For more information, refer to pages 1674–1675
of the Continuum article “Management of Stroke in the Neurocritical Care Unit.”
20 The preferred response is C (IV methylprednisolone). Post–recombinant tissue

plasminogen activator angioedema is treated with IV methylprednisolone
and antihistamines. Concurrent use of lisinopril has been noted in most
patients. Swelling of the tongue can lead to airway compromise, and
aggressive airway management may be required. For more information, refer
to page 1665 and Table 4-4 of the Continuum article “Management of Stroke
in the Neurocritical Care Unit.”
ARTICLE 5: STATUS EPILEPTICUS, REFRACTORY STATUS EPILEPTICUS, AND

SUPER-REFRACTORY STATUS EPILEPTICUS
21 The preferred response is A (IV valproic acid). Of the agents listed, IV

valproic acid would be the best medication to administer next. This patient
has not entered stage II (established status epilepticus) and has proven to be
1832 DECEMBER 2018

benzodiazepine resistant. The other agents would be appropriate to consider if
a second-line medication is ineffective. Other second-line agents to consider
would include IV phenobarbital, fosphenytoin, phenytoin, and levetiracetam.
For more information, refer to pages 1695–1696 of the Continuum article “Status
Epilepticus, Refractory Status Epilepticus, and Super-refractory Status
Epilepticus.”
22 The preferred response is D (ketogenic diet). The ketogenic diet has shown
evidence of efficacy in adults with super-refractory status epilepticus in small
trials and would be a reasonable next step in this patient. Allopregnanolone has
not been shown to be efficacious in super-refractory status epilepticus, and
the evidence for therapeutic hypothermia is currently mixed. Given the
tachyphylaxis associated with midazolam as well as the lack of clinical
response thus far, increasing the midazolam drip rate is unlikely to be effective.
Similarly, IV phenobarbital is unlikely to help in a patient refractory to
pentobarbital infusion. For more information, refer to page 1700 of the
Continuum article “Status Epilepticus, Refractory Status Epilepticus, and
Super-refractory Status Epilepticus.”
23 The preferred response is B (age younger than 18 years). This patient’s age,
younger than 18 years, is significantly associated with the risk of finding seizures
on EEG. Her comatose state is another predisposing factor in this case. A
history of epilepsy and convulsive seizures would be predisposing factors,
while female sex and a history of diabetes mellitus have not been associated
with the likelihood of seizures. For more information, refer to page 1694 of
the Continuum article “Status Epilepticus, Refractory Status Epilepticus, and
Super-refractory Status Epilepticus.”
24 The preferred response is A (age older than 60 years). Age older than 60 years is
associated with a worse outcome following status epilepticus. Other factors
include female sex, admission to a smaller hospital, presence of comorbidities,
presence of complications of status epilepticus, nonconvulsive status
epilepticus, and need for intubation. For more information, refer to page 1701
of the Continuum article “Status Epilepticus, Refractory Status Epilepticus,
and Super-refractory Status Epilepticus.”
ARTICLE 6: COMA AND BRAIN DEATH
25 The preferred response is A (amantadine). In a randomized controlled trial,

amantadine was shown to accelerate recovery of consciousness after
traumatic brain injury, although overall outcomes were not affected. The other
medications are supported by anecdotal reports but have not been shown
to be effective in a randomized controlled trial. For more information, refer to
page 1721 of the Continuum article “Coma and Brain Death.”

26 The preferred response is A (anoxic brain injury). While a great deal of

variability in outcome exists for patients in coma, anoxic brain injury is generally
associated with a worse prognosis than other causes of coma. Autoimmune
encephalitis can be associated with very good functional recovery following
appropriate treatment. Delayed recovery with an overall good prognosis may
be seen in drug intoxication and traumatic brain injury, particularly in younger
patients. Good functional recovery after refractory status epilepticus is also
possible. For more information, refer to pages 1721–1723 of the Continuum
article “Coma and Brain Death.”
27 The preferred response is C (minimally conscious state minus). This patient’s

level of consciousness would be best categorized as minimally conscious state
minus. He has had impaired consciousness for more than 28 days and now has
periods of eye opening with purposive but non–language-based function (visual
tracking). If his purposive behavior included language function, his condition would
be considered minimally conscious state plus. If he had periods of eye opening
but no purposive behaviors, his condition would be considered vegetative
state/unresponsive wakefulness syndrome. Since he has periods of eye opening
and some degree of interactivity with the environment, he would no longer be
considered comatose, although he was comatose immediately following his injury.
The patient does not meet neurologic criteria for brain death. For more
information, refer to page 1724 of the Continuum article “Coma and Brain Death.”
28 The preferred response is C (no respiratory movements of any kind with a

PaCO2 of 65 mm Hg). An apnea test would confirm apnea if the patient had
no respiratory movements of any kind with a PaCO2 of greater than or equal to
60 mm Hg (or 20 mm Hg greater than at baseline). Any respiratory movement,
including agonal gasps, would indicate some level of brainstem function, at which
point the test should be aborted. The primary drive to breathe is associated with
hypercarbia rather than hypoxia, and the criteria are therefore based on PaCO2
rather than PaO2. The apnea test has no strict time criteria. While a systolic blood
pressure of 70 mm Hg and falling would be a reason to abort the test, it would not be
consistent with apnea but, rather, would mean that the test was inconclusive. For
more information, refer to Table 6-8 of the Continuum article “Coma and Brain Death.”
ARTICLE 7: MANAGEMENT OF COMATOSE SURVIVORS OF CARDIAC ARREST
29 The preferred response is A (electrolyte disturbances and immunosuppression

are known complications). Targeted temperature management is indicated in
this patient, who is status post–cardiac arrest with evidence of neurologic
dysfunction (unresponsiveness). A goal temperature of 32°C (89.6°F) to 36°C
(96.9°F) for 24 to 72 hours is most reasonable; more specific targets regarding
temperature and duration are less clear at this time. Temperature management,
particularly with lower temperature goals, is associated with a number of side
effects, including immunosuppression and electrolyte derangements; some of
1834 DECEMBER 2018

these can be exacerbated by rewarming too quickly. Shivering should be
controlled while patients are undergoing temperature management, and
temperature management should begin as soon as safely possible following
resuscitation. For more information, refer to page 1740 of the Continuum article
“Management of Comatose Survivors of Cardiac Arrest.”
30 The preferred response is B (cognitive dysfunction is common in survivors of

cardiac arrest and may be amenable to rehabilitation). Even in patients who
make an otherwise good recovery in terms of motor function, gait, and basic
cognition, a large proportion (20% to 50%) will have persistent, measurable
cognitive dysfunction at 3 months postarrest. This can have a large impact on
individuals’ ability to return to work and other prior activities. Patients and
families should be counseled about the possibility of cognitive issues as well as
their potential reversibility and improvement with cognitive rehabilitation. For
more information, refer to page 1748 and Case 7-4 and of the Continuum article
“Management of Comatose Survivors of Cardiac Arrest.”
31 The preferred response is D (N20 responses can be predictive in patients

receiving sedation but are discouraged with suppressive levels of
sedation/anesthesia). Absence of bilateral N20 responses on somatosensory
evoked potential testing is a very strong predictor of a poor neurologic outcome
following resuscitation from cardiac arrest, but the test’s use has limitations.
Specifically, since somatosensory evoked potential testing is EEG-based,
responses can be impacted by high-dose sedation or anesthesia that reaches
suppressive levels, as might be needed to treat postanoxic seizures or status
epilepticus. As such, while N20 response status can be used in a predictive
fashion in patients receiving lower doses of sedation, its use is discouraged in
patients receiving higher-dose sedation as the effects of sedation can
confound results. For more information, refer to page 1743 of the Continuum
article “Management of Comatose Survivors of Cardiac Arrest.”
32 The preferred response is C (seizures should be considered a treatable form

of secondary brain injury postarrest and should be treated). At the present
time, seizures should be considered both a marker of an underlying brain injury
as well as a potentially treatable cause of ongoing brain injury post–cardiac
arrest. As such, this patient’s events, both convulsive and nonconvulsive,
should be treated fairly aggressively as long as other signs of poor neurologic
outcome are not present. For more information, refer to page 1741 of the
Continuum article “Management of Comatose Survivors of Cardiac Arrest.”
ARTICLE 8: CRITICAL CARE OF NEUROMUSCULAR DISORDERS
33 The preferred response is D (sniff nasal pressure). Of the respiratory function

tests listed, sniff nasal pressure has the best predictive value for respiratory

failure. Supine forced vital capacity may have more sensitivity than erect
forced vital capacity, but sniff nasal pressure is more predictive than both.
Peak cough expiratory flow is a better measure of cough effectiveness than
impending respiratory failure. For more information, refer to page 1768 of the
Continuum article “Critical Care of Neuromuscular Disorders.”
34 The preferred response is D (pralidoxime). This patient presents with signs and
symptoms consistent with organophosphate intoxication, and pralidoxime
administration is the best treatment option. Pralidoxime should be used within
a few hours of organophosphate intoxication before phosphorylated
acetylcholinesterase becomes resistant to reactivation. Corticosteroids, IV
immunoglobulin (IVIg), and plasma exchange would not be helpful in this
scenario. Pyridostigmine, an acetylcholinesterase inhibitor, would likely make
symptoms worse by increasing concentrations of acetylcholine. In case of
excessive secretions, atropine should also be considered. For more
information, refer to page 1771 of the Continuum article “Critical Care of
Neuromuscular Disorders.”
35 The preferred response is B (ocular muscles are most commonly spared).

Intensive care unit–acquired weakness is a common finding in patients with
prolonged stays in the intensive care unit and may result from a combination
of myopathic and neuropathic changes. The pattern of weakness in this
disorder is typically generalized (as opposed to focal), with more notable
involvement of proximal muscles. Facial, ocular, and bulbar muscles are most
commonly spared, so the presence of weakness in these muscles should
suggest another diagnosis. Respiratory involvement is common, as intensive
care unit–acquired weakness is often seen as a cause of failure to wean from
mechanical ventilation. Reflexes are usually decreased or absent. For more
information, refer to pages 1757–1758 and Case 8-1 of the Continuum article
“Critical Care of Neuromuscular Disorders.”
36 The preferred response is C (IV immunoglobulin [IVIg] and plasma exchange

are equally effective in this scenario). This patient presents with an acute
exacerbation of her known myasthenia gravis, likely related to abrupt
withdrawal of her medications. Although her respiratory function is stable at
this time, her degree of weakness and bulbar dysfunction warrants acute
treatment. Either IVIg or plasma exchange could be used in this patient; they
are considered to be equally effective in this setting, and using the two
modalities together does not have an additive effect. Corticosteroids are not
contraindicated in this setting but should be given in tandem with another
acute therapy such as IVIg or plasma exchange. Thymectomy could be
considered as a long-term treatment but should not be carried out during an
acute exacerbation. Pyridostigmine or other cholinesterase inhibitors can used
for symptomatic management but would be inappropriate as monotherapy
given her degree of symptoms. For more information, refer to page 1762 of the
Continuum article “Critical Care of Neuromuscular Disorders.”
1836 DECEMBER 2018

ARTICLE 9: MULTIMODALITY MONITORING IN THE NEUROCRITICAL CARE
UNIT
37 The preferred response is A (elevated lactate to pyruvate ratio). Cerebral

microdialysis is a multimodal monitoring device for patients requiring
neurologic intensive care. Although a definitive relationship between dialysate
findings and clinical outcomes is still being established, certain results are
thought to be consistent with ongoing metabolic dysfunction, which include
low glucose levels, elevated glutamate levels, and an elevated lactate to
pyruvate ratio. For more information, refer to pages 1777–1778 of the
Continuum article “Multimodality Monitoring in the Neurocritical Care Unit.”
38 The preferred response is B (delayed cerebral ischemia). In patients with

aneurysmal subarachnoid hemorrhage, several options are available to monitor
for and detect delayed cerebral ischemia, particularly in patients in whom the
clinical examination alone is insufficient or unreliable. One option includes
quantitative EEG, for which several findings, such as drop in the alpha/delta
ratio and drop in the alpha variability, can be seen several days prior to
radiographic changes of delayed cerebral ischemia. For more information, refer
to page 1781 of the Continuum article “Multimodality Monitoring in the
39 The preferred response is E (ventriculostomy). This patient likely has elevated

intracranial pressure as a result of his subarachnoid hemorrhage and associated
hydrocephalus and would benefit from continuous intracranial pressure
monitoring. In this case, a ventriculostomy is the preferred option, since in
addition to providing a measure of global intracranial pressure from his diffuse
brain injury, it will also allow for CSF diversion and thus offer a therapeutic
benefit. For more information, refer to page 1782 of the Continuum article
“Multimodality Monitoring in the Neurocritical Care Unit.”
40 The preferred response is B (EEG). Patients experiencing a cardiac arrest who

then undergo therapeutic hypothermia or targeted temperature management
are at risk for developing seizures or nonconvulsive status epilepticus. As a
result, EEG monitoring is recommended during hypothermia and rewarming
both for detection of these events and for prognostic factors. While the other
options listed can provide some information on brain status, their role in the
care of the patient post–cardiac arrest is less clear. For more information, refer
to page 1781 of the Continuum article “Multimodality Monitoring in the

APPENDIX A
1838 DECEMBER 2018

APPENDIX A

APPENDIX A
1840 DECEMBER 2018

APPENDIX A

APPENDIX A
1842 DECEMBER 2018

APPENDIX A

APPENDIX A
1844 DECEMBER 2018

APPENDIX A

APPENDIX A
1846 DECEMBER 2018

APPENDIX B

APPENDIX B
1848 DECEMBER 2018

ERRATA
In the February 2015 and April 2018 issues of

Continuum (Spinal Cord Disorders, Vol. 21, No. 1 and
Spinal Cord Disorders, Vol. 24, No. 2), the following
errors occurred:
In FIGURE 1-3 of “Spinal Cord Functional Anatomy”

by Tracey A. Cho, MD, FAAN (Continuum: Lifelong
Learning in Neurology 2015;21:17), which was
reprinted as FIGURE 1-1 of “Approach to Myelopathy”
by Tracey A. Cho, MD, FAAN, and Shamik
Bhattacharyya, MD, MS (Continuum: Lifelong
Learning in Neurology 2018;24:387), the
intermediolateral column is incorrectly labeled as
“(C8 to T3),” and should be labeled “(T1 to L3).”
The corrected figure is below.
Cho TA. Spinal cord functional anatomy. Continuum (Minneap Minn)

2015;21(1, Spinal Cord Disorders):13–35. doi:10.1212/01.
CON.0000461082.25876.4a.
Cho TA, Bhattacharyya S. Approach to myelopathy. Continuum

(Minneap Minn) 2018;24(2, Spinal Cord Disorders):386–406. doi:10.1212/
CON.0000000000000583.
The editors regret these errors.

LEARNING OBJECTIVES AND CORE COMPETENCIES
Learning Objectives Core Competencies

Upon completion of this Continuum: Lifelong This Continuum: Lifelong Learning in Neurology
Learning in Neurology Neurocritical Care issue, Neurocritical Care issue covers the following
participants will be able to: core competencies:
◆ Implement the distinct treatment strategies and ◆ Patient Care

protocols for cerebral edema, elevated intracranial
pressure, and cerebral herniation syndromes ◆ Medical Knowledge
◆ Discuss the current best practices for the ◆ Practice-Based Learning and Improvement
management of intracerebral hemorrhage and review
the results of recent major intracerebral hemorrhage ◆ Interpersonal and Communication Skills
clinical trials
◆ Professionalism
◆ Discuss the epidemiology, diagnosis, management,
and medical and neurologic complications of ◆ Systems-Based Practice
aneurysmal subarachnoid hemorrhage
◆ Discuss the management of acute ischemic stroke

including its management in the neurocritical care unit
◆ Define status epilepticus and discuss the etiologies,

diagnostic tools, the urgency for aborting it,
and implement the currently available treatment
options, including escalation algorithms
◆ Describe the diagnostic and therapeutic approach

to acute coma and the principles of brain
death determination
◆ Discuss the assessment and management of patients

who are comatose following cardiac arrest
◆ Review the management options and intensive care

needs and evaluate respiratory failure in patients with
neuromuscular disorders
◆ Discuss the approaches to multimodality

neuromonitoring, the neuromonitoring devices
currently available, and their strengths and limitations
◆ Navigate discussions about advanced care planning

and end-of-life care
◆ Recognize caveats that may undermine the aims

of shared decision making in the neurocritical
care setting
1576 D EC E M B ER 2 0 1 8

LIST OF ABBREVIATIONS
Neurocritical Care GBS Guillain-Barré syndrome

GCS Glasgow Coma Scale
GRE Gradient recalled echo
99M
Tc Technetium 99m HHT Hypertensive and mild hypervolemic therapy
AAN American Academy of Neurology HIPAA Health Insurance Portability and Accountability Act
ABC Airway, breathing, and circulation HMPAO Hexamethylpropyleneamine oxime
AChR Acetylcholine receptor ICAT Intensive Care After Thrombolysis
ACTH Adrenocorticotropic hormone ICH Intracerebral hemorrhage
AED Antiepileptic drug ICP Intracranial pressure
AHA American Heart Association ICU Intensive care unit
AIDP Acute inflammatory demyelinating IM Intramuscular
polyradiculoneuropathy
INR International normalized ratio
ALS Amyotrophic lateral sclerosis
IV Intravenous
AMAN Acute motor axonal neuropathy
IVH Intraventricular hemorrhage
AMPA α-Amino-3-hydroxy-5-methylisoxazole-4
propionic acid IVIg Intravenous immunoglobulin
AMSAN Acute motor-sensory axonal neuropathy LEMS Lambert-Eaton myasthenic syndrome
ASPECTS Alberta Stroke Program Early Computed LRP4 Lipoprotein receptor–related protein 4
Tomography Score MAP Mean arterial pressure
ATP Adenosine triphosphate MicroRNA Microribonucleic acid
BiPAP Bilevel positive airway pressure
MRA Magnetic resonance angiography
BMI Body mass index
MRC Medical Research Council
CK Creatine kinase
MRI Magnetic resonance imaging
CMAP Compound muscle action potential
mRS Modified Rankin Scale
CMS Centers for Medicare & Medicaid Services
MRV Magnetic resonance venography
CNS Central nervous system
MUAP Motor unit action potential
CPP Cerebral perfusion pressure
CPT Current Procedural Terminology MuSK Muscle-specific tyrosine kinase
CSF Cerebrospinal fluid NIH National Institutes of Health
CT Computed tomography NIHSS National Institutes of Health Stroke Scale
CTA Computed tomography angiography NMDA N-methyl-D-aspartate
CTP Computed tomography perfusion NORSE New-onset refractory status epilepticus
DNA Deoxyribonucleic acid PbtO2 Brain tissue oxygen tension
DSA Digital subtraction angiography PET Positron emission tomography
DWI Diffusion-weighted imaging PRES Posterior reversible encephalopathy syndrome
ECD Ethyl cysteinate dimer PTT Partial thromboplastin time
ECG Electrocardiogram RCVS Reversible cerebral vasoconstriction syndrome
ECMO Extracorporeal membrane oxygenation rTMS Repetitive transcranial magnetic stimulation
EEG Electroencephalogram; electroencephalography rtPA Recombinant tissue plasminogen activator
EMG Electromyography SAH Subarachnoid hemorrhage
EMSE Epidemiology-based Mortality Score in Status SAPS II Simplified Acute Physiology Score II
Epilepticus
SIADH Syndrome of inappropriate secretion
ENLS Emergency neurologic life support of antidiuretic hormone
EVD External ventricular drain
SIRS Systemic inflammatory response syndrome
FDA US Food and Drug Administration
SNAP Sensory serve action potential
FDG Fludeoxyglucose
SOFA Sequential Organ Failure Assessment [score]
FDG-PET Fludeoxyglucose positron emission tomography
SPECT Single-photon emission computed tomography
FFP Fresh frozen plasma
STESS Status Epilepticus Severity Score
FLAIR Fluid-attenuated inversion recovery
SWI Susceptibility-weighted imaging
fMRI Functional magnetic resonance imaging
TBI Traumatic brain injury
FOUR Full Outline of UnResponsiveness [score]
TCD Transcranial Doppler
FRESH Functional Recovery Expected After
Subarachnoid Hemorrhage TICI Thrombolysis in Cerebral Infarction
FUNC Functional Outcome in Patients With Primary TSH Thyroid-stimulating hormone

Intracerebral Hemorrhage VGCC Voltage-gated calcium channel
GABA γ-Aminobutyric acid WFNSS World Federation of Neurological Surgeons Scale

Neurocritical Care
Article 1: Cerebral Edema and Elevated
Intracranial Pressure
Matthew A. Koenig, MD, FNCS. Continuum (Minneap Minn). December 2018; 24 (6
Neurocritical Care):1588–1602.
ABSTRACT
PURPOSE OF REVIEW:
This article reviews the management of cerebral edema, elevated intracranial pressure (ICP),
and cerebral herniation syndromes in neurocritical care.
RECENT FINDINGS:
While corticosteroids may be effective in reducing vasogenic edema around brain tumors,
they are contraindicated in traumatic cerebral edema. Mannitol and hypertonic saline use
should be tailored to patient characteristics including intravascular volume status. In patients
with traumatic brain injury who are comatose, elevated ICP should be managed with an
algorithmic, multitiered treatment protocol to maintain an ICP of 22 mm Hg or less. Third-line
ICP treatments include anesthetic agents, induced hypothermia, and decompressive
craniectomy. Recent clinical trials have demonstrated that induced hypothermia and
decompressive craniectomy are ineffective as early neuroprotective strategies and should
be reserved for third-line management of refractory ICP elevation in severe traumatic brain
injury. Monitoring for cerebral herniation should include bedside pupillometry in supratentorial
space-occupying lesions and recognition of upward herniation in patients with posterior
fossa lesions.
SUMMARY:
Although elevated ICP, cerebral edema, and cerebral herniation are interrelated, treatments
should be based on the distinct pathophysiologic process. Focal lesions resulting in brain
compression are primarily managed with surgical decompression, whereas global or multifocal
brain injury requires a treatment protocol that includes medical and surgical interventions.
KEY POINTS
• Corticosteroids are ineffective for the treatment of cytotoxic edema and are contraindicated in the
treatment of patients with severe traumatic brain injury.
• Frequent administration of hypertonic saline may cause hyperchloremic metabolic acidosis, which is
associated with higher mortality in neurocritical care. Buffering hypertonic solutions with acetate may lower
the chance of developing hyperchloremic metabolic acidosis.

• Serum osmolality should be monitored in patients treated with frequent doses of mannitol. Mannitol should
be held when serum osmolality exceeds 320 mOsm/kg to 340 mOsm/kg or when the osmolar gap exceeds
20 mOsm/kg.
• Osmotic agents can be used as a temporizing measure to treat mass effect from cytotoxic edema related to
stroke and intracerebral hemorrhage, but evidence for efficacy is weak. These lesions may require surgical
decompression.
• In states of poor intracranial compliance due to global cerebral edema, small changes in intracranial volume
related to flat head position, hyperemia, hypercarbia, fever, or pain may result in exaggerated increases in
intracranial pressure.
• Spontaneous oscillations in intracranial pressure called Lundberg A and B waves may cause self-limited
increases in intracranial pressure that last several minutes. These oscillations are an indicator of poor
intracranial compliance, but they typically resolve spontaneously without treatment.
• Augmentation of cerebral perfusion pressure with systemic vasopressors may lower intracranial pressure by
causing reflex cerebral vasoconstriction, thereby lowering the intracranial volume of blood. Excessive
cerebral perfusion, however, may contribute to vasogenic edema in regions with a disrupted blood-brain
barrier.
• Current Brain Trauma Foundation guidelines recommend maintaining an intracranial pressure of 22 mm Hg or
less and a cerebral perfusion pressure of at least 50 mm Hg to 60 mm Hg in patients with severe traumatic
brain injury.
• A recent clinical trial of an intracranial pressure–based treatment protocol for severe traumatic brain injury
compared to a treatment protocol based on clinical examination and imaging without intracranial pressure
monitoring showed no difference in outcomes between the two groups.
• Pentobarbital infusion for intracranial pressure reduction can result in severe medication side effects such as
propylene glycol toxicity, immunosuppression, impaired gastrointestinal motility, and distributive shock.
• A recent clinical trial of early induced hypothermia as a neuroprotective strategy in severe traumatic brain
injury showed reduction of intracranial pressure–directed interventions, but neurologic outcomes were
worse in patients treated with hypothermia.
• Induced hypothermia continues to be a useful third-line intervention for refractory intracranial pressure
elevation, but an effective hypothermia protocol that includes multimodality treatment of shivering is
required.
• A recent clinical trial demonstrated that decompressive craniectomy for refractory intracranial pressure
elevation in severe traumatic brain injury improves survival and reduces the chance of severe disability, but
more patients survived in a vegetative state compared to medical management alone.
• With transtentorial herniation, the pupil dilation is ipsilateral to the mass lesion, but hemiplegia may be
contralateral because of direct corticospinal tract involvement or ipsilateral because of compression of the
contralateral cerebral peduncle against the tentorial edge (Kernohan notch phenomenon).
• Serial bedside quantitative pupillometry may detect reduction in the pupillary constriction velocity hours
prior to frank clinical signs of transtentorial herniation, which may afford time for preemptive treatment.
• Although patients can survive and recover from transtentorial herniation in some cases, the sequelae of
herniation can include Duret brainstem hemorrhage, ipsilateral anterior cerebral artery stroke, and
contralateral posterior cerebral artery stroke.
• Posterior fossa mass lesions can cause cerebellar tonsillar herniation or upward herniation of the cerebellum
through the tentorial incisura, which may not be accompanied by intracranial pressure elevation or
pupillary changes.

Article 2: Intracerebral Hemorrhage
Wendy C. Ziai, MD, MPH, FAHA, FNCS, FESO; J. Ricardo Carhuapoma, MD, FAHA.
Continuum (Minneap Minn). December 2018; 24 (6 Neurocritical Care):1603–1622.
ABSTRACT
PURPOSE OF REVIEW:
This article describes the advances in the management of spontaneous intracerebral
hemorrhage in adults.
RECENT FINDINGS:
Therapeutic intervention in intracerebral hemorrhage has continued to focus on arresting
hemorrhage expansion, with large randomized controlled trials addressing the effectiveness of
rapidly lowering blood pressure, hemostatic therapy with platelet transfusion, and other clotting
complexes and clot volume reduction both of intraventricular and parenchymal hematomas
using minimally invasive techniques. Smaller studies targeting perihematomal edema and
inflammation may also show promise.
SUMMARY:
The management of spontaneous intracerebral hemorrhage, long relegated to the management
and prevention of complications, is undergoing a recent evolution in large part owing to
stereotactically guided clot evacuation techniques that have been shown to be safe and that
may potentially improve outcomes.
KEY POINTS
• Treatment for intracerebral hemorrhage revolves around acute management of blood pressure, intracranial
pressure, and cerebral edema.
• Although the majority of intracerebral hemorrhage is attributed to hypertensive small penetrating vessel
arteriopathy, vascular imaging (CT angiography, magnetic resonance angiography, or catheter angiography)
and MRI are essential to rule out other etiologies of intracerebral hemorrhage.
• The Intracerebral Hemorrhage Score and the Functional Outcome in Patients With Primary Intracerebral
Hemorrhage score allow rapid outcome stratification of patients with intracerebral hemorrhage.
Nevertheless, they should not be used independently to guide goals of care discussions of patients with
intracerebral hemorrhage.
• A spot sign, defined by the presence of contrast within the hematoma visualized on CT angiography or
contrast-enhanced CT, is associated with a high risk of early intracerebral hemorrhage expansion.
• Although hematoma expansion is a complication that negatively modifies mortality and functional outcome
after intracerebral hemorrhage, it lacks specific interventions aimed at ameliorating its impact.
• Currently, no proven benefit has been shown from recombinant activated factor VII in spontaneous or
anticoagulation-associated intracerebral hemorrhage.
• In patients with intracerebral hemorrhage with disorders of primary or secondary hemostasis, rapid reversal
of coagulopathy is indicated in an attempt to improve neurologic outcomes and survival.
• In patients with intracerebral hemorrhage with a history of recent use of antiplatelet agents, routine use of
platelet transfusions is not indicated. The only exception for this recommendation is if surgical interventions
are anticipated in the emergency care of these patients.
• Perihematomal edema, which contributes to early neurologic deterioration and poor outcome, develops
rapidly following intracerebral hemorrhage, reaching maximal volume by 2 weeks. There is currently no
proven clinical therapy that both reduces perihematomal edema and improves outcomes.

• The routine use of antiepileptic drugs following intracerebral hemorrhage is currently not recommended.
There should be, however, a high degree of suspicion of electrographic seizures or status epilepticus in
patients with intracerebral hemorrhage with decreased level of consciousness.
• Use of an external ventricular drain is recommended in cases of intracerebral hemorrhage with
intraventricular extension and obstructive hydrocephalus. External ventricular drains can facilitate relief of
obstructive hydrocephalus as well as reduce the neurotoxic effects of intraventricular blood.
• Minimally invasive endoscopic surgery for intraventricular hemorrhage volume reduction is considered
experimental at this time and awaits large-scale studies to evaluate safety and efficacy.
• After several randomized clinical trials and contemporary meta-analyses, no clear recommendation for open
craniotomy and surgical evacuation of supratentorial hypertensive hematomas can be made at this point.
• Several new devices placed via stereotactic imaging enable enhanced drainage of parenchymal clots and
await demonstration of whether successful volumetric clot reduction is matched with improvement in
functional outcomes.
Article 3: Subarachnoid Hemorrhage

Susanne Muehlschlegel, MD, MPH, FNCS, FCCM. Continuum (Minneap Minn).
December 2018; 24 (6 Neurocritical Care):1623–1657.
ABSTRACT
PURPOSE OF REVIEW:
This article reviews the epidemiology, clinical presentation, diagnosis, and management of
patients with aneurysmal subarachnoid hemorrhage (SAH). SAH is a type of hemorrhagic stroke
and is a neurologic emergency with substantial morbidity and mortality. This article reviews
the most common and potentially life-threatening neurologic and medical complications to
promote their early recognition and prevent secondary brain injury.
RECENT FINDINGS:
Over the past 30 years, the incidence of SAH has remained stable; yet, likely because of
improved care in specialized neurocritical care units, discharge mortality has considerably
decreased. Two consensus guidelines by the American Heart Association/American Stroke
Association and the Neurocritical Care Society have outlined best practices for the management
of patients with SAH. The most important recommendations include admission of patients to
high-volume centers (defined as more than 35 SAH admissions per year) under the management
of a multidisciplinary, specialized team; expeditious identification and treatment of the
bleeding source with evaluation by a multidisciplinary team consisting of cerebrovascular
neurosurgeons, neuroendovascular specialists, and neurointensivists; management of patients
in a neurocritical care unit with enteral nimodipine, blood pressure control, euvolemia, and
close monitoring for neurologic and medical complications; and treatment of symptomatic
cerebral vasospasm/delayed cerebral ischemia with induced hypertension and endovascular
therapies. This article also highlights new insights of SAH pathophysiology and provides
updates in the management approach.
SUMMARY:
SAH remains a neurologic emergency. Management of patients with SAH includes adherence to
published guidelines, but some areas of SAH management remain understudied. Clinical trials
are required to elucidate the role of these controversial management approaches in improving
patient outcomes.

KEY POINTS
• Despite a decline in the mortality, subarachnoid hemorrhage remains a highly morbid disease.
• In patients with subarachnoid hemorrhage, aneurysm rupture occurs at an average age of 53 years. This young
age at onset results in a high societal cost and number of years of productivity lost.
• The most common cause for subarachnoid hemorrhage is a ruptured cerebral aneurysm (85%); however, 10%
of subarachnoid hemorrhages may not reveal a bleeding source, while the minority of cases (5%) may be
due to other vascular causes.
• Subarachnoid hemorrhage is predominant in women, African Americans, and Hispanics. Hypertension,
smoking, and excess alcohol intake are modifiable risk factors that individually double the risk of
subarachnoid hemorrhage.
• Current guidelines recommend screening for aneurysms if the patient has two or more first-degree relatives
with aneurysms or subarachnoid hemorrhage.
• Subarachnoid hemorrhage typically presents with a sudden and severe headache (“worst headache of life”),
which is distinctly different from usual headaches.
• Physical examination of a patient with subarachnoid hemorrhage should include determination of the level of
consciousness and the patient’s score on the Glasgow Coma Scale, evaluation for meningeal signs, and
the presence of focal neurologic deficits.
• Transient elevation in the intracranial pressure is the cause of nausea, vomiting, and syncope and may be
associated with additional cardiac and pulmonary complications after subarachnoid hemorrhage.
• The most rapidly available and appropriate initial diagnostic test for patients with suspected subarachnoid
hemorrhage is a noncontrast head CT.
• In cases of negative or equivocal head CT findings in which a high suspicion still exists for subarachnoid
hemorrhage, a lumbar puncture is the immediate next recommended step.
• CSF should be spun down and evaluated for xanthochromia by visual inspection and, if available,
spectrophotometry. Xanthochromia takes approximately 12 hours to develop and may not be present if a
lumbar puncture is performed earlier after headache onset.
• Head CT and MRI are considered equally sensitive in detecting subarachnoid hemorrhage in the first 2 days,
except in the hyperacute first 6 hours after subarachnoid hemorrhage, during which head CT may miss a
small proportion of subarachnoid hemorrhages and MRI may be slightly superior.
• Hemosiderin-sensitive MRI sequences (gradient recalled echo and susceptibility-weighted imaging) or
fluid-attenuated inversion recovery sequences have superior sensitivity to detect subacute or chronic
subarachnoid hemorrhage compared to head CT.
• The “gold standard” vessel imaging remains cerebral digital subtraction angiography.
• Approximately 15% of patients with subarachnoid hemorrhage will have negative imaging studies for a source
of bleeding, of which approximately 38% have nonaneurysmal perimesencephalic subarachnoid hemorrhage.
• The focus in the first few minutes to hours after subarachnoid hemorrhage, until the patient can undergo
treatment of the ruptured aneurysm, should be directed toward the prevention of rebleeding.
• Risk factors for rebleeding include poor-grade subarachnoid hemorrhage, hypertension, a large aneurysm,
and, potentially, the use of antiplatelet drugs.
• The short-term use (up to a maximum of 72 hours until aneurysm securement) of antifibrinolytics (tranexamic acid
or ε-aminocaproic acid) is recommended by guidelines, although there is institutional variation in their use.
• The two most commonly used clinical scales, the World Federation of Neurological Surgeons Scale and the
Hunt and Hess Scale, are strong predictors of outcome in patients with subarachnoid hemorrhage. The
most reliable and validated radiologic scale is the modified Fisher Scale.
• For the treatment of aneurysm, endovascular coiling is preferred over surgical clipping whenever possible,
but the choice of treatment depends on the patient’s age as well as the aneurysm’s location, morphology,
and relationship to adjacent vessels.
• Patients with subarachnoid hemorrhage are at risk for several additional neurologic complications, including
hydrocephalus, brain edema, delayed cerebral ischemia, rebleeding, seizures, and neuroendocrine
disorders, the latter of which can lead to impaired regulation of sodium, volume, and glucose.

• Acute symptomatic hydrocephalus occurs in 20% of patients with subarachnoid hemorrhage, usually within
minutes to days after subarachnoid hemorrhage onset. In cases of hydrocephalus, insertion of an external
ventricular drain can be lifesaving.
• If seizures occur prior to aneurysm securement, they are usually a sign of early rebleeding.
• The occurrence of nonconvulsive seizures (7% to 18%) and nonconvulsive status epilepticus (3% to 13%) is more
common in patients with subarachnoid hemorrhage who are comatose and has been associated with
delayed cerebral ischemia and worse outcomes.
• Continuous EEG monitoring should be considered in patients with high-grade subarachnoid hemorrhage.
• In the absence of randomized controlled trials of antiepileptic drug treatment in subarachnoid hemorrhage,
but with known negative effects of anticonvulsants, particularly phenytoin, on neurocognitive recovery
after subarachnoid hemorrhage, treatment with antiepileptic drugs should be limited to the preaneurysm
treatment time frame only.
• Delayed cerebral ischemia after subarachnoid hemorrhage is defined as any neurologic deterioration that
persists for more than 1 hour and cannot be explained by any other neurologic or systemic condition.
• Delayed cerebral ischemia occurs on average 3 to 14 days after subarachnoid hemorrhage. The risk for
delayed cerebral ischemia increases with subarachnoid hemorrhage thickness and intraventricular
hemorrhage, as demonstrated by the modified Fisher Scale.
• Delayed cerebral ischemia should be suspected if patients with subarachnoid hemorrhage develop a focal or
global neurologic deficit or have a decrease of 2 or more points on the Glasgow Coma Scale that lasts for
at least 1 hour and cannot be explained by any other cause.
• Patients with subarachnoid hemorrhage should undergo physiologic or imaging monitoring routinely during
the risk period for delayed cerebral ischemia.
• Transcranial Doppler has adequate sensitivity and specificity to detect increased cerebral blood flow
velocities secondary to cerebral vasospasm in the middle cerebral and basilar arteries but is highly
dependent on the operator and cranial bone window.
• Digital subtraction angiography remains the gold standard for detection of large- and middle-sized artery
vasospasm.
• Only symptomatic vasospasm, occurring in 30% of patients with subarachnoid hemorrhage, has been
associated with delayed cerebral ischemia and poor outcome after subarachnoid hemorrhage.
• Hypervolemic, hypertensive, and hemodilutional (Triple H) therapy is no longer supported by guidelines
because of the existing evidence of adverse associations with outcomes after the use of hemodilution.
Standard treatment is now hypertensive and mild hypervolemic therapy (HHT).
• In the management of patients with symptomatic vasospasm, hypertension is preferably induced using a1
receptor agonists by a continuous infusion (norepinephrine or phenylephrine).
• Cardiopulmonary dysfunction is a well-known complication of subarachnoid hemorrhage and can range from
minor ECG changes to severe stress cardiomyopathy and neurogenic pulmonary edema.
• The severity of subarachnoid hemorrhage is an independent predictor of cardiopulmonary injury, suggesting
that the cardiopulmonary injury is neurally mediated.
• Takotsubo cardiomyopathy in subarachnoid hemorrhage is associated with higher mortality and worse
long-term outcomes.
• Fever is the most common medical complication after subarachnoid hemorrhage, occurring in up to 70% of
patients.
• Fever has been associated with delayed cerebral ischemia and worse clinical outcomes and is likely related
to systemic inflammatory response syndrome and chemical meningitis, rather than an infectious process.
• Deep vein thrombosis after subarachnoid hemorrhage is common, with rates between 2% and 20%.
• Mechanical venous thromboembolism prophylaxis should be initiated immediately on admission with the use
of pneumatic compression devices. At the author’s institution, chemoprophylaxis with subcutaneous
fractionated or unfractionated heparin is usually initiated immediately after endovascular aneurysm repair
and within 24 hours after craniotomy for clipping.

• In the absence of clinical trials of glucose control in patients with subarachnoid hemorrhage, current
recommendations are to maintain a blood glucose level between 80 mg/dL and 200 mg/dL.
• Hyponatremia is the most common electrolyte disorder in patients with subarachnoid hemorrhage and can
occur in up to 30% of patients.
• The laboratory findings are similar in both cerebral salt wasting and syndrome of inappropriate secretion of
antidiuretic hormone. The only differentiating finding is the patient’s intravascular volume status; cerebral
salt wasting is a hypovolemic state, while patients with syndrome of inappropriate secretion of antidiuretic
hormone are euvolemic or even hypervolemic. It is of utmost importance to correctly differentiate these
two syndromes because treatment is opposite.
• Cerebral salt wasting is treated with fluid administration and sometimes a continuous infusion of hypertonic
saline and fludrocortisone if diuresis and natriuresis impede maintenance of adequate volume status.
Patients with syndrome of inappropriate secretion of antidiuretic hormone are treated with fluid restriction
and sometimes diuresis with loop diuretics.
• Anemia and hemoglobin concentrations of less than 9 g/dL have been associated with delayed cerebral
ischemia and poor clinical outcomes; however, optimal hemoglobin goal levels and transfusion thresholds
are not known.
Article 4: Management of Stroke in the

Neurocritical Care Unit
Chethan P. Venkatasubba Rao, MD, FNCS; Jose I. Suarez, MD, FNCS, FANA. Continuum
(Minneap Minn). December 2018; 24 (6 Neurocritical Care):1658–1682.
ABSTRACT
PURPOSE OF REVIEW:
This article provides updated information regarding the diagnosis and treatment (specifically
critical care management) of acute ischemic stroke. This article also discusses the increased use
of thrombolysis and thrombectomy in clinical practice.
RECENT FINDINGS:
Stroke is the leading cause of disability in the United States. A significant proportion of patients
with acute ischemic stroke require critical care management. Much has changed in the early
evaluation and treatment of patients presenting with acute ischemic stroke. The introduction of
embolectomy in large vessel occlusions for up to 24 hours post–symptom onset has resulted
in one in every three eligible patients with acute ischemic stroke with the potential to lead an
independent lifestyle. These patients increasingly require recognition of complications and
initiation of appropriate interventions as well as earlier admission to dedicated neurocritical care
units to ensure better outcomes.
SUMMARY:
This article emphasizes issues related to the management of patients with acute ischemic stroke
undergoing mechanical thrombectomy and thrombolysis and addresses the complex
physiologic changes affecting neurologic and other organ systems.
KEY POINTS
• One in every four patients with acute ischemic stroke will need critical care intervention.
• Factors predicting critical care admission in patients with acute ischemic stroke include severity of stroke,
age, elevated systolic blood pressure, and hyperglycemia.

• Circulation, airway, and breathing stabilization followed by rapid neurologic assessment, neuroimaging, and
point of care testing should be the initial response in the evaluation of patients presenting with acute
ischemic stroke.
• The main neurologic reasons for intensive care unit admission in patients with acute ischemic stroke include
blood pressure management post-thrombolysis or post-thrombectomy, cerebral edema, symptomatic
hemorrhagic transformation, and seizures.
• The main cardiac and respiratory indications for admission of patients with acute ischemic stroke to the
neurocritical care unit include myocardial infarction, cardiac arrhythmias, heart failure, inability to maintain
the airway, and the need for mechanical ventilation.
• Management of hemodynamics in patients with fluctuating neurologic symptoms should be targeted
individually and adjusted for optimum symptom control.
• Interpretation of the neurologic evaluation of patients with acute ischemic stroke with larger deficits can be
obscured when only using the National Institutes of Health Stroke Scale and no other clinical parameters.
• In patients with acute ischemic stroke, treatment of hemorrhagic transformation should be directed toward
hemodynamic control and coagulopathy reversal.
• Recombinant tissue plasminogen activator–related symptomatic intracranial hemorrhage should be treated
initially with cryoprecipitate.
• While interpreting the levels of fibrinogen, it is important to realize that it is one of the acute phase reactants
and, hence, a lower level is more reliable, and a normal level should be considered with healthy skepticism.
• Patients with acute ischemic stroke with a National Institutes of Health Stroke Scale score of greater than 15,
altered sensorium, and infarction of more than 50% of the middle cerebral artery territory should be
considered for prophylactic hemicraniectomy within 48 hours.
• Patients with acute ischemic stroke who present with cerebellar strokes involving more than 25% to 33% of a
hemisphere should be considered for suboccipital decompressive craniectomy.
• Short-acting IV agents should be used to provide optimal hemodynamic management in patients with acute
ischemic stroke.
• Early mobilization in the intensive care unit should be encouraged in patients with acute ischemic stroke.
• When possible, endotracheal intubation and general anesthesia administration should be avoided for
patients undergoing mechanical thrombectomy.
Article 5: Status Epilepticus, Refractory

Status Epilepticus, and Super-refractory
Status Epilepticus
Sarah E. Nelson, MD; Panayiotis N. Varelas, MD, PhD, FNCS, FAAN. Continuum
(Minneap Minn). December 2018; 24 (6 Neurocritical Care):1683–1707.
ABSTRACT
PURPOSE OF REVIEW:
Status epilepticus, refractory status epilepticus, and super-refractory status epilepticus can be
life-threatening conditions. This article presents an overview of the three conditions and
discusses their management and outcomes.
RECENT FINDINGS:
Status epilepticus was previously defined as lasting for 30 minutes or longer but now is more
often defined as lasting 5 minutes or longer. A variety of potential causes exist for status

epilepticus, refractory status epilepticus, and super-refractory status epilepticus, but all three
ultimately involve changes at the cellular and molecular level. Management of patients with
status epilepticus generally requires several studies, with EEG of utmost importance given the
pathophysiologic changes that can occur during the course of status epilepticus. Status
epilepticus is treated with benzodiazepines as first-line antiepileptic drugs, followed by
phenytoin, valproic acid, or levetiracetam. If status epilepticus does not resolve, these are
followed by an IV anesthetic and then alternative therapies based on limited data/evidence,
such as repetitive transcranial magnetic stimulation, therapeutic hypothermia, immunomodulatory
agents, and the ketogenic diet. Scores have been developed to help predict the outcome of
status epilepticus. Neurologic injury and outcome seem to worsen as the duration of status
epilepticus increases, with outcomes generally worse in super-refractory status epilepticus
compared to status epilepticus and sometimes also to refractory status epilepticus.
SUMMARY:
Status epilepticus can be a life-threatening condition associated with multiple complications,
including death, and can progress to refractory status epilepticus and super-refractory status
epilepticus. More studies are needed to delineate the best management of these three entities.
KEY POINTS
• While in the past, tonic-clonic status epilepticus was defined as continuous seizure activity or two or more
seizures without recovery of consciousness lasting longer than 30 minutes, tonic-clonic status epilepticus
is now defined as seizure activity lasting 5 minutes or longer, with 30 minutes being the cutoff for development
of long-term consequences.
• No standard definition for nonconvulsive status epilepticus currently exists. A working definition
differentiates diagnostic criteria based on whether epileptic encephalopathy is present.
• Refractory status epilepticus is continuous seizure activity not controlled by first-line and second-line
antiepileptic drugs; it occurs in 9% to 43% of all cases of status epilepticus.
• Super-refractory status epilepticus is defined either as status epilepticus not controlled by third-line
anesthetic agents or as status epilepticus continuing for 24 hours or longer after anesthesia is administered.
The exact incidence and associated mortality of super-refractory status epilepticus are unknown.
• The annual incidence of status epilepticus is approximately 12.6 per 100,000 person-years and is increasing over
time. Seizures or status epilepticus may occur in up to 19% of patients hospitalized in the intensive care unit.
• The etiology of status epilepticus may be divided into known or symptomatic causes and unknown or
cryptogenic causes. In general, acute causes appear to be more common than chronic causes.
• Encephalitis appears to be a common cause of both refractory status epilepticus and super-refractory status
epilepticus.
• New-onset refractory status epilepticus (NORSE) has recently emerged as a challenging disorder in which the
etiology of refractory status epilepticus is not immediately apparent.
• Multiple changes at the cellular and molecular level occur in status epilepticus as time passes.
• With increasing duration of seizure activity, the frequency of systemic complications, neurologic injury, and
mortality increase.
• Continuous video-EEG is a valuable tool given its ability to detect nonconvulsive status epilepticus.
• Imaging modalities, including CT, MRI, single-photon emission CT, and positron emission tomography, may
help in the diagnosis and management of status epilepticus.
• Status epilepticus requires urgent treatment because the response to treatments and the prognosis diminish
with elapsed time. A staged approach to treatment has thus been advocated, with different medications
used in early (stage I), established (stage II), refractory (stage III), and super-refractory status epilepticus
(stage IV).
• Treating status epilepticus generally begins with benzodiazepines and is followed by IV administration of
fosphenytoin, valproic acid, or levetiracetam.

• The extent of suppression needed to treat refractory status epilepticus (burst suppression versus merely
suppressing seizures) is not known.
• Continuous anesthetic agents are used to treat refractory status epilepticus and are typically maintained for
24 to 48 hours before being weaned.
• Several alternative therapies can be used in patients with refractory status epilepticus and, especially,
super-refractory status epilepticus, including surgical resection, repetitive transcranial magnetic
stimulation, immunosuppression or immunomodulation, and the ketogenic diet.
• In some cases, seizures and status epilepticus may be epiphenomena for severe brain injury rather than the
primary offender. It is unknown whether treating these conditions will improve outcomes.
• Up to 50% of patients with super-refractory status epilepticus die, and outcomes are worse in
super-refractory status epilepticus compared to nonrefractory status epilepticus.
Article 6: Coma and Brain Death

Alejandro A. Rabinstein, MD, FAAN. Continuum (Minneap Minn). December 2018; 24 (6
Neurocritical Care):1708–1731.
ABSTRACT
PURPOSE OF REVIEW:
This article discusses the diagnostic and therapeutic approach to patients who are comatose and
reviews the current knowledge on prognosis from various causes of coma. This article also
provides an overview of the principles for determination of brain death as well as advice on how
to avoid common pitfalls.
RECENT FINDINGS:
Technologic advances have refined our understanding of the physiology of consciousness and
the spectrum of disorders of consciousness; they also promise to improve our prognostic
accuracy. Yet the clinical principles for the evaluation and treatment of coma remain unaltered.
The clinical standards for determination of death by neurologic criteria (ie, brain death) are also
well established, although variabilities in local protocols and legal requirements remain a
problem to be resolved.
SUMMARY:
Effective evaluation of coma demands a systematic approach relying on clinical information to
ensure rational use of laboratory and imaging tests. When the cause of coma is deemed
irreversible in the setting of a catastrophic brain injury and no clinical evidence exists for brain
and brainstem function, patients should be evaluated for the possibility of brain death by
following the clinical criteria specified in the American Academy of Neurology guidelines.
KEY POINTS
• Coma is a state of unresponsiveness in which the patient is not awake and cannot interact with the
environment, even after vigorous stimulation.
• Bilateral diffuse alterations in cortical function or severe diencephalic or brainstem dysfunction can produce
coma.
• Focused history and physical examination very often provide clues to the etiology of coma.
• Coma scales are useful but should not replace a more complete neurologic examination.
• The differential diagnosis and diagnostic testing should be guided by the history, physical examination, and
location of the patient (out of hospital, in the emergency department, on a hospital floor, or in the intensive
care unit).

• CT is a reasonable and practical first imaging study when structural damage is suspected, but MRI may be
more useful for various causes of coma.
• The value of EEG for the evaluation of patients who are comatose without clinical seizures remains to be
established.
• When evaluating coma of unclear etiology, always think of treatable causes first.
• Do not assume coma is irreversible when the cause remains indeterminate.
• Treatable causes of coma most often represent a medical emergency.
• The value of pharmacologic or nonpharmacologic neurostimulation for coma recovery is unproven.
• The prognosis of coma depends mostly on its cause and duration.
• Prolonged disorders of consciousness (longer than 28 days) are typically the result of widespread brain
damage.
• Prolonged disorders of consciousness constitute a spectrum that includes vegetative state (also referred to
as unresponsive wakefulness syndrome) and minimally conscious state.
• Although the problem is improving, practice variations in the determination of brain death continue to occur
across different states and countries.
• The diagnosis of brain death demands an established cause of irreversible coma, the absence of confounding
factors, the complete absence of motor response and brainstem reflexes, and no breathing efforts on a
formal apnea test.
• Angiography, nuclear scans, EEG, and transcranial Doppler are ancillary tests that can be used to confirm
brain death, but they all have limitations.
• The most appropriate use of ancillary tests to confirm brain death is to complement the neurologic
examination when certain aspects of the examination (such as the apnea test) cannot be performed.
• Not recognizing confounding factors (especially the residual effects of central nervous system depressants)
is a common pitfall in brain death evaluations.
• The concept of brain death is accepted by all major religions, but small cultural and religious groups continue
to challenge the validity of the concept.
Article 7: Management of Comatose

Survivors of Cardiac Arrest
David B. Seder, MD, FCCP, FCCM, FNCS. Continuum (Minneap Minn). December 2018;
24 (6 Neurocritical Care):1732–1752.
ABSTRACT
PURPOSE OF REVIEW:
Because the whole-body ischemia-reperfusion insult associated with cardiac arrest often
results in brain injury, neurologists perform an important role in postresuscitation cardiac arrest
care. This article provides guidance for the assessment and management of brain injury following
cardiac arrest.
RECENT FINDINGS:
Neurologists have many roles in postresuscitation cardiac arrest care: (1) early assessment of
brain injury severity to help inform triage for invasive circulatory support or revascularization;
(2) advocacy for the maintenance of a neuroprotective thermal, hemodynamic, biochemical, and
metabolic milieu; (3) detection and management of seizures; (4) development of an accurate,
multimodal, and conservative approach to prognostication; (5) application of shared
decision-making paradigms around the likely outcomes of therapy and the goals of care; and

(6) facilitation of the neurocognitive assessment of survivors. Therefore, optimal management
requires early neurologist involvement in patient care, a detailed knowledge of postresuscitation
syndrome and its complex interactions with prognosis, expertise in bringing difficult cases to
their optimal conclusions, and a support system for survivors with cognitive deficits.
SUMMARY:
Neurologists have a critical role in postresuscitation cardiac arrest care and are key participants
in the treatment team from the time of first restoration of a perfusing heart rhythm through
the establishment of rehabilitation services for survivors.
KEY POINTS
• Top treatment priorities in the early hours after resuscitation include determining and reversing the cause of
cardiac arrest to prevent circulatory collapse and rearrest as well as maintaining a neuroprotective
hemodynamic and biochemical milieu.
• Restoration and maintenance of a favorable thermal, hemodynamic, biochemical, and metabolic milieu after
cardiac arrest is of great concern and should be a key element of urgent recommendations.
• Because many patients who undergo targeted temperature management after cardiac arrest receive
neuromuscular blockade to control shivering (and therefore never have convulsions) and because others
experience primarily nonconvulsive seizures, it is ideal to monitor for seizures with continuous EEG.
• Prognostication after cardiac arrest is problematic. Many of the frequently cited studies describing
prognostic factors after cardiac arrest are flawed, and imprecise early prognostication can lead to
unnecessary deaths.
• Cognitive dysfunction is a significant source of morbidity among survivors of cardiac arrest.
• Unlike traditional prognostication after a cardiac arrest, the goal of early neurologic evaluation is to classify
patients into categories of high, medium, and low risk of a poor neurologic outcome in a time frame that
allows for triage to individualized treatment pathways.
• Rapid diagnostic testing modalities that allow for accurate early neurologic triage after cardiac arrest include
assessment of the total ischemic time, the no-flow and low-flow intervals, neurologic examination, CT, EEG
background, continuity and reactivity at different time points after resuscitation, and the bispectral index
and suppression ratio.
• Patients whose risk assessment suggests mild or moderate brain injury should be considered for initiation of
aggressive strategies for circulatory support, as they are likely to benefit, up to and including extracorporeal
membrane oxygenation or surgical revascularization, if necessary, and should undergo an accelerated
diagnostic workup to identify the cause of the arrest with the intention of maintaining adequate cerebral
blood flow and preventing rearrest.
• All patients who are comatose after cardiac arrest should be treated in a maximally neuroprotective milieu.
• Because the cerebral microcirculation depends on maintenance of an adequate perfusion pressure in the
hours after resuscitation to prevent progression of vulnerable cortical and subcortical regions from ischemia
to necrosis, the maintenance of adequate blood pressure and cardiac output is crucial.
• Hypotension after resuscitation is associated with worse outcomes and must be avoided.
• Body temperature should be controlled as soon as possible following resuscitation to a target temperature
between 32°C (89.6°F) and 36°C (96.8°F) for 24 to 72 hours after resuscitation, with careful management
of the complications of therapy.
• Data suggest the superiority of a 35°C (95°F) to 36°C (96.8°F) target for patients with severe shock, and the
higher target should also be considered in patients with active bleeding or life-threatening infections.
• Rewarming after targeted temperature management should always be performed slowly to minimize adverse
consequences of the inflammatory and vasodilatory effects of warming.
• Because of the strong independent association of hyperglycemia and poor outcome after cardiac arrest,
plasma glucose levels should be normalized by insulin administration.

• Hypoxic-ischemic encephalopathy–related seizures are both a marker of brain injury severity and a cause of
ongoing secondary injury.
• Patients who are comatose after cardiac arrest should be monitored with continuous EEG, at least through
the process of rewarming to normal body temperature and the resolution or waning of epileptiform activity.
• Patients with hypoxic-ischemic encephalopathy–related seizures who do not have very high neuron-specific
enolase levels and extensive brain injury on MRI may make a good recovery, and ongoing supportive therapy
is appropriate unless circulatory collapse or multiple organ failure preclude ongoing aggressive care.
• When performed too soon or without adequate diagnostic testing, prognostication after resuscitation from
cardiac arrest can cost some patients their lives.
• Multimodal prognostication is standard after cardiac arrest.
• When the prognosis after cardiac arrest is uncertain, the principles of shared decision making are employed
to help the family and treatment team determine a patient- and family-centered pathway of care that
offers comfort, support, hope, and direction.
• Many experts believe formal assessment of prognosis after cardiac arrest should be delayed for at least
72 hours after rewarming from targeted temperature management and potentially longer.
• Neurologists should be familiar with factors associated with delayed awakening after cardiac arrest,
including renal insufficiency, older age, and postresuscitation shock.
• Shared decision making is considered the standard approach to conversations about end-of-life care. When
prognostication has been performed, a multidisciplinary meeting that includes the patient’s surrogates
and health care stakeholders should be held to exchange information, deliberate, and make a decision about
the goals of care.
• More than half of survivors of cardiac arrest have measurable neurocognitive dysfunction, and 20% to 50%
have persistent deficits of memory, learning, or executive function even after 3 months.
Article 8: Critical Care of Neuromuscular

Disorders
Diana Greene-Chandos, MD, FNCS; Michel Torbey, MD, MPH, FCCM, FAHA, FNCS,
FAAN. Continuum (Minneap Minn). December 2018; 24 (6 Neurocritical Care):1753–1775.
ABSTRACT
PURPOSE OF REVIEW:
Weakness is a common reason patients are seen in neurologic consultation. This article reviews
the differential diagnosis of neuromuscular disorders in the intensive care unit (ICU), discusses
the intensive care needs and evaluation of respiratory failure in patients with neuromuscular
disorders, and provides a practical guide for management.
RECENT FINDINGS:
Although primary neuromuscular disorders used to be the most common cause for weakness
from peripheral nervous system disease in the ICU, a shift toward ICU-acquired weakness is
observed in today’s clinical practice. Therefore, determining the cause of weakness is important
and may have significant prognostic implications. Guillain-Barré syndrome and myasthenia
gravis remain the most common primary neuromuscular disorders in the ICU. In patients with
myasthenia gravis, it is important to be vigilant with the airway and institute noninvasive
ventilation early in the course of the disease to attempt to avoid the need for intubation. On the
other hand, patients with Guillain-Barré syndrome should be intubated without delay if the
airway is at risk to avoid further complications. In patients with ICU-acquired weakness, failure

to wean from the ventilator is usually the challenge. Early mobility, glucose control, minimizing
sedation, and avoiding neuromuscular blocking agents remain the only therapeutic regimen
available for ICU-acquired weakness.
SUMMARY:
Critical care management of neuromuscular disorders requires a multidisciplinary approach
engaging members of the ICU and consultative teams. Developing an airway management
protocol could have implications on outcome and length of stay for patients with neuromuscular
disorders in the ICU. Tending to the appropriate nuances of each patient who is critically ill with
a neuromuscular disorder through evidence-based medicine can also have implications on length
of stay and outcome.
KEY POINTS
• Intensive care unit–acquired weakness is a clinical diagnosis.
• Early clinical signs of respiratory failure include tachycardia, tachypnea, increased sweating, and use of
accessory muscles.
• Hypercapnia, as indicated on arterial blood gases, is often a late sign of intensive care unit–acquired
weakness; therefore, arterial blood gases are not very useful in acute triage decisions.
• Early mobilization and glucose control are key in the prevention of intensive care unit–acquired weakness.
• Rhabdomyolysis occurs with a wide spectrum of signs and symptoms, ranging from severe muscle pain and
varying degrees of generalized weakness. Markedly elevated plasma creatine kinase may lead to
myoglobinuria and acute kidney injury.
• Noninvasive ventilation should be considered early in a patient with myasthenia gravis.
• A hallmark of Lambert-Eaton myasthenic syndrome is postexercise facilitation. Tendon reflexes and muscle
strength improve immediately after a brief maximal contraction.
• Lambert-Eaton myasthenic syndrome should be considered in patients with any unexplained weakness after
pharmacologic neuromuscular blockade.
• The potassium channel blocker 3,4-diaminopyridine is an effective symptomatic treatment for
Lambert-Eaton myasthenic syndrome.
• Plasma exchange and IV immunoglobulin are equally effective for the treatment of Guillain-Barré syndrome.
The decision to use either should depend on the availability of these treatments at the specific treatment
center.
• No evidence exists to recommend any specific ventilatory mode in patients with Guillain-Barré syndrome.
• Noninvasive ventilation has a limited role in patients with severe Guillain-Barré syndrome.
• Appropriate psychological support is recommended for patients with Guillain-Barré syndrome in the
intensive care unit.
• The types of amyotrophic lateral sclerosis include limb-onset amyotrophic lateral sclerosis, which has both
upper motor neuron and lower motor neuron signs; bulbar-onset amyotrophic lateral sclerosis, which
features dysphagia and dysarthria and upper motor neuron and lower motor neuron signs that occur after the
disease progresses; primary lateral sclerosis, which has upper motor neuron involvement only; and
progressive muscular atrophy, which has lower motor neuron involvement only.
• Respiratory complications are the main cause of death in patients with amyotrophic lateral scerlosis due to
diaphragmatic weakness, aspiration, or pneumonia.
• Patients with amyotrophic lateral sclerosis without severe bulbar dysfunction treated with noninvasive
positive pressure ventilation had an improved quality of life and survived longer compared with those who
underwent standard care.
• Ventilation via tracheostomy should be offered as an alternative to patients who do not tolerate noninvasive
positive pressure ventilation.

• Organophosphates irreversibly inhibit acetylcholinesterases, resulting in excessive amounts of acetylcholine
at the neuromuscular junction and prolonged endplate potential and desensitization of the postsynaptic
membrane.
Article 9: Multimodality Monitoring in the

Neurocritical Care Unit
Lucia Rivera Lara, MD, MPH; Hans Adrian Püttgen, MD. Continuum (Minneap Minn).
December 2018; 24 (6 Neurocritical Care):1776–1788.
ABSTRACT
PURPOSE OF REVIEW:
This article focuses on the multiple neuromonitoring devices that can be used to collect bedside
data in the neurocritical care unit and the methodology to integrate them into a multimodality
monitoring system. The article describes how to apply the collected data to appreciate the
physiologic changes and develop therapeutic approaches to prevent secondary injury.
RECENT FINDINGS:
The neurologic examination has served as the primary monitor for secondary brain injury in
patients admitted to the neurocritical care unit. However, the International Multidisciplinary
Consensus Conference on Multimodality Monitoring in Neurocritical Care concluded that
frequent bedside examinations are not sufficient to detect and prevent secondary brain injury
and that integration of multimodality monitoring with advanced informatics tools will most likely
enhance our assessments compared to the clinical examinations alone. This article reviews the
invasive and noninvasive technologies used to monitor focal and global neurophysiologic
cerebral alterations.
SUMMARY:
Multimodal monitoring is still in the early stages of development. Research is still needed to
establish more advanced monitors with the bioinformatics to identify useful trends from data
gathered to predict clinical outcome or prevent secondary brain injury.
KEY POINTS
• The aim of multimodality monitoring is to understand the complexity of the changes that lead to secondary
brain injury.
• The most severe limitation to the application of cerebral microdialysis in bedside decision making is the
unfinished work of defining markers of health and crisis in different clinical contexts.
• Brain tissue oxygen tension monitors have shown low rates of complication.
• Many factors affect brain tissue oxygen tension including cerebral perfusion pressure, hemoglobin
concentration, oxygen saturation, metabolic rate (from fever, agitation, or shivering), and cerebral
vasospasm.
• Near-infrared spectroscopy devices have an excellent safety profile.
• Near-infrared spectroscopy shows promise as a method of assessing cerebral autoregulation when used in
concert with systemic blood pressure and intracranial pressure monitors.
• Understanding global physiology is key to detecting significant changes that affect large areas in the brain
after traumatic brain injury, aneurysmal subarachnoid hemorrhage, ischemic stroke, intracranial hemorrhage
and status epilepticus, among other conditions.

• Nonconvulsive seizures have been described in 5% to 15% of patients with acute intracranial injury (eg,
traumatic brain injury, subarachnoid hemorrhage, intracerebral hemorrhage, postneurosurgery), and
nonconvulsive status epilepticus has been described in 5% to 20% of such patients.
• EEG is recommended for patients during therapeutic hypothermia after cardiac arrest and within 24 hours of
rewarming, not only to diagnose and treat seizures and myoclonic status epilepticus, but also because of
its prognostic factors.
• If brain oxygenation monitoring is desired, the preferred choice is brain tissue oxygen tension sensors.
• Intracranial pressure monitoring is recommended as part of protocol-driven care for patients with acute
brain injury who are at risk of elevated intracranial pressure based on clinical or imaging features.
• The gold standard intracranial pressure monitor is the ventriculostomy.
• Intracranial pressure monitors are recommended in patients with a Glasgow Coma Scale score of 8 or less
and who have an abnormal head CT.
• Noninvasive intracranial pressure monitors are still under development and are not yet recognized as
providing an accurate intracranial pressure measurement.
• Dynamic cerebral autoregulation monitoring has evolved to allow bedside calculation of optimal cerebral
perfusion pressure and optimal mean arterial pressure with invasive and noninvasive methods.
• The importance of cerebral autoregulation multimodal monitoring technology is that calculation of optimal
mean arterial pressure and cerebral perfusion pressure would allow physicians to individualize cerebral
perfusion pressure or mean arterial pressure goals and potentially improve a patient’s outcomes.
• Cerebral perfusion pressure below the optimal level increases the incidence of fatal outcome, whereas
excessively high cerebral perfusion pressure is associated with an increased proportion of severe disability.
• Monitoring systems will find their greatest utility when appreciated in concert with each other and with other
forms of clinical data including laboratory values, imaging results, and medical record documentation.
• Lack of common formatting standards stands as a significant barrier to true data integration, which
necessitates a patient data management system that translates monitor output from proprietary formats and
merges them into a common stream.

Issue Overview
Neurocritical Care, Volume 24, Issue 6, December 2018
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 Neurocritical Care issue,
participants will be able to:
 Implement the distinct treatment strategies and protocols for cerebral edema, elevated
intracranial pressure, and cerebral herniation syndromes
 Discuss the current best practices for the management of intracerebral hemorrhage and
review the results of recent major intracerebral hemorrhage clinical trials
 Discuss the epidemiology, diagnosis, management, and medical and neurologic
complications of aneurysmal subarachnoid hemorrhage
 Discuss the management of acute ischemic stroke including its management in the
neurocritical care unit
 Define status epilepticus and discuss the etiologies, diagnostic tools, the urgency for
aborting it, and implement the currently available treatment options, including escalation
algorithms
 Describe the diagnostic and therapeutic approach to acute coma and the principles of
brain death determination
 Discuss the assessment and management of patients who are comatose following cardiac
arrest
 Review the management options and intensive care needs and evaluate respiratory failure
in patients with neuromuscular disorders

 Discuss the approaches to multimodality neuromonitoring, the neuromonitoring devices
currently available, and their strengths and limitations
 Navigate discussions about advanced care planning and end-of-life care
 Recognize caveats that may undermine the aims of shared decision making in the
neurocritical care setting
Core Competencies
This Continuum: Lifelong Learning in Neurology Neurocritical Care issue covers the following
core competencies:
 Patient Care
 Medical Knowledge
 Practice-Based Learning and Improvement
 Interpersonal and Communication Skills
 Professionalism
 Systems-Based Practice
Contributors
Jose I. Suarez, MD, FNCS, FANA, Guest Editor

Professor, Director of Neurosciences Critical Care, Departments of Anesthesiology and Critical
Care Medicine, Neurology, and Neurosurgery, Johns Hopkins University School of Medicine,
Baltimore, Maryland
Relationship Disclosure: Dr Suarez has received research/grant support from the National Institute of Neurological
Disorders and Stroke and as co-investigator in the SETPOINT2 (Stroke-related Early Tracheostomy Versus
Prolonged Orotracheal Intubation in Neurocritical Care Trial) study from the Patient-Centered Outcomes Research
Institute. Dr Suarez is the current president and a member of the board of directors of the Neurocritical Care Society.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Suarez reports no disclosure.

Associate Professor, Departments of Neurology, Neurosurgery, and Anesthesiology and Critical
Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland
Relationship Disclosure: Dr Carhuapoma reports no disclosure.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Carhuapoma discusses the unlabeled/investigational

use of alteplase for the treatment of intracerebral hemorrhage and discusses the use of several devices, including
ultrasound microcatheters, sonothrombolysis for minimally invasive subcortical parafascicular transsulcal access for
clot evacuation (MiSPACE), and the stereotactic mechanical (suction/vibration/aspiration) thrombolytic technique
for minimally invasive evacuation of intracerebral hemorrhage.

Diana Greene-Chandos, MD, FNCS
Director of Neuroscience Critical Care for Education; Program Director, Neuroscience Critical
Care Fellowship, Wexner Medical Center, Arthur G. James Cancer Hospital and Richard J.
Solove Research Institute; Assistant Professor, Departments of Neurology and Neurosurgery,
The Ohio State University College of Medicine, Columbus, Ohio
Relationship Disclosure: Dr Greene-Chandos has received personal compensation as a speaker for the Ecuador
Neurocritical Care Society Annual Meeting and the Neurocritical Care Society Annual Meeting and has received
research/grant support from Biogen, the Neuroemergency Treatment Trials Grant from the National Institutes of
Health, and the Stroke Network Grant from the National Institute of Neurological Disorders and Stroke. Dr Greene-
Chandos has worked as a consultant for court cases.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Greene-Chandos reports no disclosure.
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 ethical and 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 Medical Group and has received personal compensation as a medicolegal consultant in legal cases
involving criminal cases, malpractice, and personal injury.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Kass reports no disclosure.
Matthew A. Koenig, MD, FNCS

Medical Director of Telemedicine, The Queen’s Health Systems; Associate Professor of
Medicine, John A. Burns School of Medicine, University of Hawaii, Honolulu, Hawaii
Relationship Disclosure: Dr Koenig has received research/grant support as principal investigator of a study for the
Hawaii Department of Health Neurotrauma Special Fund and receives publishing royalties from Rutgers University
Press.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Koenig reports no disclosure.
Christos Lazaridis, MD, EDIC

Associate Professor, Division of Neurocritical Care, Departments of Neurology and
Neurosurgery, University of Chicago, Chicago, Illinois
Relationship Disclosure: Dr Lazaridis serves on the editorial board of Neurocritical Care and has received
research/grant support from Cheetah Medical, Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Lazaridis reports no disclosure.

Ariane Lewis, MD
Associate Professor, Departments of Neurology and Neurosurgery, New York University
Langone Medical Center, New York, New York
Relationship Disclosure: Dr Lewis has received personal compensation as a workshop lecturer for the American
Academy of Neurology and for grand rounds lectures from Mount Sinai Hospital, Mount Sinai Beth Israel Hospital,
and Newark Beth Israel Medical Center, and for work as a legal consultant for David A. Axelrod & Associates, PC.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Lewis reports no disclosure.
Susanne Muehlschlegel, MD, MPH, FNCS, FCCM

Associate Professor, Director of Neurocritical Care Research, Departments of Neurology,
Anesthesiology, and Surgery, University of Massachusetts Medical School, Worcester,
Massachusetts
Relationship Disclosure: Dr Muehlschlegel has received research/grant support from the National Institutes of
Health/National Institute of Child Health and Human Development and the Prize for Academic Collaboration and
Excellence (PACE) from the University of Massachusetts Memorial Medical Group. Dr Muehlschlegel receives
partial research salary support as the site principal investigator for the INTREPID (Impact of Fever Prevention in
Brain Injured Patients) trial sponsored by C. R. Bard Inc. Dr Muehlschlegel has received compensation for serving
as a course director for the American Academy of Neurology.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Muehlschlegel discusses the unlabeled/investigational

short-term use of antifibrinolytics (ε-aminocaproic acid and tranexamic acid) for the treatment of early aneurysm
bleeding, the use of fludrocortisone for the treatment of cerebral salt wasting syndrome after subarachnoid
hemorrhage, the use of levetiracetam for seizure prophylaxis, and the use of milrinone, nicardipine, and verapamil as
endovascular therapy using intraarterial vasodilators for the treatment of subarachnoid hemorrhage.
Sarah E. Nelson, MD
Assistant Professor, Departments of Neurology and Anesthesiology and Critical Care Medicine,
Johns Hopkins University School of Medicine, Baltimore, Maryland
Relationship Disclosure: Dr Nelson receives grant support from the Johns Hopkins Anesthesiology and Critical Care
Medicine (ACCM) Stimulating and Advancing ACCM Research (StAAR) program.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Nelson discusses the unlabeled/investigational use of
allopregnanolone, deep brain stimulation, desflurane, diazepam, electroconvulsive therapy, fosphenytoin,
gabapentin, isoflurane, IV immunoglobulin, ketamine, ketogenic diet, lacosamide, levetiracetam, lidocaine,
lorazepam, methylprednisolone, midazolam, pentobarbital, phenobarbital, phenytoin, plasma exchange, propofol,
pyridoxine, thiopental, topiramate, vagal nerve stimulation, and valproic acid for the treatment of refractory status
epilepticus.
Marc R. Nuwer, MD, PhD, FAAN

Professor, Vice Chair, Department of Neurology, David Geffen School of Medicine; Department
Head, Department of Clinical Neurophysiology, Ronald Reagan University of California Los
Angeles Medical Center, University of California Los Angeles, Los Angeles, California
Relationship Disclosure: Dr Nuwer has received personal compensation for serving as an honorary editor for
Clinical Neurophysiology; for serving on the board of directors of CortiCare, Inc; and for serving on the editorial
boards of the Journal of Clinical Neurophysiology and Neurology Clinical Practice. Dr Nuwer receives royalties
from Cambridge University Press and has received research/grant funding from the National Institutes of Health,
Second Sight, and the United States Army. Dr Nuwer has given expert medicolegal testimony in a court deposition
and a court hearing.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Nuwer reports no disclosure.

Hans Adrian Püttgen, MD
Assistant Professor, Department of Neurology, Johns Hopkins University School of Medicine,
Baltimore, Maryland
Relationship Disclosure: Dr Püttgen reports no disclosure.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Püttgen discusses the unlabeled/investigational use of
near-infrared spectroscopy as a surrogate of cerebral blood flow to measure cerebral autoregulation.
Alejandro A. Rabinstein, MD, FAAN

Professor of Neurology, Mayo Clinic, Rochester, Minnesota
Relationship Disclosure: Dr Rabinstein receives royalties from Elsevier and Oxford University Press and has
received research support from DJO Global, Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Rabinstein reports no disclosure.
Lucia Rivera Lara, MD, MPH

Assistant Professor, Departments of Neurology and Anesthesiology and Critical Care Medicine,
Johns Hopkins University School of Medicine, Baltimore, Maryland
Relationship Disclosure: Dr Rivera Lara has received research/grant support from Covidien Ltd/Medtronic; the
Johns Hopkins Anesthesiology and Critical Care Medicine (ACCM) Stimulating and Advancing ACCM Research
(StAAR) award; and Ornim, Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Rivera Lara discusses the unlabeled/investigational
use of near-infrared spectroscopy as a surrogate of cerebral blood flow to measure cerebral autoregulation.

Assistant Professor, Departments of Neurology and Neurotherapeutics, University of Texas
Southwestern Medical Center, Dallas, Texas
Relationship Disclosure: Dr Rubin reports no disclosure.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Rubin reports no disclosure.
David B. Seder, MD, FCCP, FCCM, FNCS

Chairman, Department of Critical Care Services, Maine Medical Center, Portland, Maine;
Associate Professor of Medicine, Tufts University School of Medicine, Boston, Massachusetts
Relationship Disclosure: Dr Seder receives grant support from the Patient-Centered Outcomes Research Institute
(CER-1602-34137).
Unlabeled Use of Products/Investigational Use Disclosure: Dr Seder discusses the unlabeled/investigational use of
the bispectral index in determining the severity of brain injury after cardiac arrest.

Michel Torbey, MD, MPH, FCCM, FAHA, FNCS, FAAN
Professor and Chairman, Department of Neurology; University of New Mexico, Albuquerque,
New Mexico
Relationship Disclosure: Dr Torbey has received personal compensation as the past president of the Neurocritical
Care Society and has received grant support from the National Institute of Neurological Disorders and Stroke.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Torbey reports no disclosure.
Panayiotis N. Varelas, MD, PhD, FNCS, FAAN

Professor of Neurology, Wayne State University; Division Head, Neurosciences Critical Care
Services, Director Neuro-Intensive Care Unit, Henry Ford Hospital, Detroit, Michigan
Relationship Disclosure: Dr Varelas serves on the board of directors of the Neurocritical Care Society, on the
editorial board of Neurocritical Care, and on an advisory board of Portola Pharmaceuticals, Inc. Dr Varelas has
received personal compensation for speaking engagements for Portola Pharmaceuticals, Inc and UCB SA and
receives publishing royalties from Springer. Dr Varelas receives research/grant support from Bard Pharmaceuticals
Limited; Edge Therapeutics, Inc; Marinus Pharmaceuticals, Inc; the National Institutes of Health; the Patient-
Centered Outcomes Research Institute; and Portola Pharmaceuticals, Inc.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Varelas discusses the unlabeled/investigational use of
allopregnanolone, deep brain stimulation, desflurane, diazepam, electroconvulsive therapy, fosphenytoin,
gabapentin, isoflurane, IV immunoglobulin, ketamine, ketogenic diet, lacosamide, levetiracetam, lidocaine,
lorazepam, methylprednisolone, midazolam, pentobarbital, phenobarbital, phenytoin, plasma exchange, propofol,
pyridoxine, thiopental, topiramate, vagal nerve stimulation, and valproic acid for the treatment of refractory status
epilepticus.
Chethan P. Venkatasubba Rao, MD, FNCS

Section Head, Vascular Neurology and Neurocritical Care; Program Director, Neurocritical Care
Fellowship; Assistant Professor, Baylor College of Medicine; Medical Director, Neurocritical
Care, Baylor St. Luke’s Medical Center, Houston, Texas
Relationship Disclosure: Dr Venkatasubba Rao has received personal compensation as an editorial board member of
Brain Disorders & Therapy.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Venkatasubba Rao reports no disclosure.
Paul M. Vespa, MD, FAAN

Assistant Dean of Critical Care Medicine, Gary L. Brinderson Family Chair in Neurocritical
Care, Neurology Director of Neurocritical Care, Professor of Neurosurgery and Neurology,
David Geffen School of Medicine, University of California Los Angeles, Los Angeles,
California
Relationship Disclosure: Dr Vespa has received personal compensation as an editor for Critical Care Medicine and
Neurocritical Care. Dr Vespa has received personal compensation as a consultant for Sage Therapeutics, has
received research/grant support from the National Institutes of Health, and holds stock options in InTouch Health.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Vespa reports no disclosure.
Wendy C. Ziai, MD, MPH, FAHA, FNCS, FESO

Associate Professor, Departments of Neurology, Neurosurgery, and Anesthesiology and Critical
Care Medicine, Johns Hopkins University School of Medicine, Baltimore, Maryland

Relationship Disclosure: Dr Ziai serves as an associate editor for Neurocritical Care, on an advisory board for C.R.
Bard Inc, and has received personal compensation as a consultant for and research/grant support from HeadSense
Medical Ltd. Dr Ziai has received grant support from the Johns Hopkins Anesthesiology and Critical Care Medicine
(ACCM) Stimulating and Advancing ACCM Research (StAAR) program and from the National Institutes of Health
(5U01NS062851, 1U01NS08082).
Unlabeled Use of Products/Investigational Use Disclosure: Dr Ziai discusses the unlabeled/investigational use of
alteplase for the treatment of intracerebral hemorrhage and discusses the use of several devices, including ultrasound
microcatheters, sonothrombolysis for minimally invasive subcortical parafascicular transsulcal access for clot
evacuation (MiSPACE), and the stereotactic mechanical (suction/vibration/aspiration) thrombolytic technique for
minimally invasive evacuation of intracerebral hemorrhage.
Self-Assessment and CME Test Writers
Adam G. Kelly, MD
Associate Professor of Neurology; Chief, Neurovascular Division, University of Florida,
Gainesville, Florida
Relationship Disclosure: Dr Kelly has received personal compensation as CME editor of Neurology.
Unlabeled Use of Products/Investigational Use Disclosure: Dr Kelly reports no disclosure.
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.

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.

Neurocritical Care.2018 Continuum

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

Neurocritical Care.2018 Continuum

Uploaded by

Copyright:

Available Formats

DECEMBER 2018

VOL. 24 NO. 6 Neurocritical Care

1586 Editor’s Preface

1588 Cerebral Edema and Elevated Intracranial Pressure

1603 Intracerebral Hemorrhage

1623 Subarachnoid Hemorrhage

1658 Management of Stroke in the Neurocritical Care Unit

1683 Status Epilepticus, Refractory Status Epilepticus,

1708 Coma and Brain Death

1732 Management of Comatose Survivors of Cardiac Arrest

1753 Critical Care of Neuromuscular Disorders

DENOTES SUPPLEMENTAL 1776 Multimodality Monitoring in the Neurocritical Care Unit

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

1789 Ethical Considerations in End-of-life Care in the Face of

1794 End-of-life Considerations and Shared Decision Making in

1800 Neurocritical Care Coding for Neurologists

SELF-ASSESSMENT AND CME

1576 Learning Objectives and Core Competencies

1811 Instructions for Completing Postreading Self-Assessment and CME

1813 Postreading Self-Assessment and CME Test

1827 Postreading Self-Assessment and CME Test—Preferred Responses

1838 Appendix A: Summary of Evidence-based Guideline for Clinicians:

1847 Appendix B: AAN Summary of Evidence-based Guideline for

List of Abbreviations (Back Cover)

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

Jose I. Suarez, MD, FNCS, FANA, Diana Greene-Chandos,

J. Ricardo Carhuapoma, Unlabeled Use of Products/Investigational

Unlabeled Use of Products/Investigational

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

Unlabeled Use of Products/Investigational

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

Susanne Muehlschlegel, MD, Sarah E. Nelson, MD

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

Unlabeled Use of Products/Investigational

Unlabeled Use of Products/Investigational

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

David B. Seder, MD, FCCP, Panayiotis N. Varelas, MD, PhD,

Unlabeled Use of Products/Investigational

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

Unlabeled Use of Products/Investigational

Unlabeled Use of Products/Investigational

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

Self-Assessment and CME Test Writers

Adam G. Kelly, MD James W. M. Owens Jr, MD, PhD

Michael J. Bradshaw, MD Aarti Sarwal, MD, FAAN

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

A Critical Look Back and Ahead

1586 DECEMBER 2018

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

neurologic patients. © 2018 American Academy of Neurology.

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

RECENT FINDINGS: While corticosteroids may be effective in reducing vasogenic

1588 DECEMBER 2018

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

Treatment of Cerebral Edema

CORTICOSTEROIDS. Dexamethasone has been a mainstay of treatment for

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

significant symptoms attributable to cerebral edema rather than focal neurologic

OSMOTIC AGENTS. The mainstays of osmotic therapy in the treatment of cerebral

1590 DECEMBER 2018

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

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

As demonstrated in FIGURE 1-1,