Neurocritical Care.2018 - Lori A. Shutter

EOITEO tY
LORI SHUTTER
BRADLEY). MOLYNEAUX
Neurocritical
Care
SERIES EOITOk: JOHN A. KELLUM

Neurocritical Care
Pittsburgh Critical Care Medicine Series
Published and Forthcoming Titles

in the Pittsburgh Critical Care Medicine Series
Continuous Renal Replacement Therapy
edited by John A. Kellum, Rinaldo Bellomo, and Claudio Ronco
Renal and Metabolic Disorders
edited by John A. Kellum and Jorge Cerdá
Emergency Department Critical Care
edited by Donald Yealy and Clifton Callaway
Trauma Intensive Care
edited by Samuel Tisherman and Racquel Forsythe
Abdominal Organ Transplant Patients
edited by Ali Al-Khafaji
Infection and Sepsis
edited by Peter Linden
Pediatric Intensive Care
edited by Scott Watson and Ann Thompson
Mechanical Ventilation
by John W. Kreit
Rapid Response System
edited by Raghavan Murugan and Joseph M. Darby
Neurocritical Care
edited by Lori A. Shutter and Bradley J. Molyneaux
Cardiac Problems
edited by Thomas Smitherman
ICU Procedures
by Scott Gunn and Holt Murray
Neurocritical Care
Edited by
Lori A. Shutter, MD, FNCS, FCCM

Vice Chair of Education, Critical Care Medicine
Professor, Critical Care Medicine, Neurology &
Neurosurgery
Director, Neurocritical Care Fellowship Training Program
Medical Director, Neurovascular & Neurotrauma ICUs
Pittsburgh, PA
Bradley J. Molyneaux, MD, PhD

Assistant Professor, Neurology and Critical Care Medicine
University of Pittsburgh
Pittsburgh, PA
1
1
Oxford University Press is a department of the University of Oxford. It furthers
the University’s objective of excellence in research, scholarship, and education
by publishing worldwide. Oxford is a registered trade mark of Oxford University
Press in the UK and certain other countries.
Published in the United States of America by Oxford University Press

198 Madison Avenue, New York, NY 10016, United States of America.
© Oxford University Press 2018
All rights reserved. No part of this publication may be reproduced, stored in

a retrieval system, or transmitted, in any form or by any means, without the
prior permission in writing of Oxford University Press, or as expressly permitted
by law, by license, or under terms agreed with the appropriate reproduction
rights organization. Inquiries concerning reproduction outside the scope of the
above should be sent to the Rights Department, Oxford University Press, at the
address above.
You must not circulate this work in any other form

and you must impose this same condition on any acquirer.
Library of Congress Cataloging-in-Publication Data

Names: Shutter, Lori, editor. | Molyneaux, Bradley J., editor.
Title: Neurocritical care / edited by Lori Shutter, Bradley J. Molyneaux.
Other titles: Neurocritical care (Shutter) | Pittsburgh critical care medicine.
Description: Oxford; New York: Oxford University Press, [2018] |
Series: Pittsburgh critical care medicine series | Includes bibliographical references.
Identifiers: LCCN 2018003488 | ISBN 9780199375349 (pbk.)
Subjects: | MESH: Central Nervous System Diseases–therapy | Critical Care–methods |
Trauma, Nervous System—therapy
Classification: LCC RC350.N49 | NLM WL 301 | DDC 616.8/0428–dc23
LC record available at https://lccn.loc.gov/2018003488
This material is not intended to be, and should not be considered, a substitute for medical
or other professional advice. Treatment for the conditions described in this material is highly
dependent on the individual circumstances. And, while this material is designed to offer
accurate information with respect to the subject matter covered and to be current as of the
time it was written, research and knowledge about medical and health issues is constantly
evolving and dose schedules for medications are being revised continually, with new side
effects recognized and accounted for regularly. Readers must therefore always check the
product information and clinical procedures with the most up-to-date published product
information and data sheets provided by the manufacturers and the most recent codes
of conduct and safety regulation. The publisher and the authors make no representations
or warranties to readers, express or implied, as to the accuracy or completeness of
this material. Without limiting the foregoing, the publisher and the authors make no
representations or warranties as to the accuracy or efficacy of the drug dosages mentioned
in the material. The authors and the publisher do not accept, and expressly disclaim, any
responsibility for any liability, loss or risk that may be claimed or incurred as a consequence
of the use and/or application of any of the contents of this material.
9 8 7 6 5 4 3 2 1
Printed by WebCom, Inc., Canada
We dedicate this book to our patients and their families—they are the reason we
come to work every day, stay at work later than our families like, and constantly
strive to learn more.
We also must acknowledge the advance practice providers, nurses, pharmacists,

and respiratory therapists who actually run every ICU—nothing gets accomplished
without you.
Contents
Preface xi
Contributors xiii
Section 1: Neurological Conditions

1 Coma and Low Arousal States
Namir Khandker and Lori A. Shutter 3
2 Encephalopathy and Delirium
Joshua Keegan and Colleen Moran 15
3 Acute Ischemic Stroke
Cynthia L. Kenmuir and Tudor G. Jovin 23
4 Intracerebral Hemorrhage
Opeolu Adeoye 37
5 Subarachnoid Hemorrhage
Sherry Hsiang-Yi Chou 44
6 Reversible Cerebral Vasoconstriction Syndrome
Krystle Shafer and Bradley J. Molyneaux 54
7 Anoxic Brain Injury
Jonathan Elmer and Jon C. Rittenberger 58
8 Seizures and Status Epilepticus
Michael E. Reznik and Jan Claassen 68
9 The Management of Traumatic Brain Injury
Jeremy G. Stone, David M. Panczykowski,
and David O. Okonkwo 83
10 The Management of Traumatic Spinal Cord Injury
David M. Panczykowski, Jeremy G. Stone,
and David O. Okonkwo 97
11 Meningitis and Encephalitis
Ruchira Jha 105
12 Inflammation and Demyelination
Daniel B. Rubin and Henrikas Vaitkevicius 119
13 Neuromuscular Conditions
Contents
Deepa Malaiyandi and Saša A. Živković 131

14 Movement Disorder Emergencies and Movement
Disorders in the ICU
Mihai C. Sandulescu and Edward A. Burton 141
Section 2: Interventions and Monitoring

15 Multimodality Monitoring
Maranatha Ayodele and Kristine O’Phelan 155
16 External Ventricular Drainage: Clinical Indications,
Surgical Technique, and Management
Nitin Agarwal and Andrew F. Ducruet 165
17 Neuroimaging and Neurointerventional Procedures
Cynthia L. Kenmuir and Ashutosh P. Jadhav 173
18 Hypothermia and Fever Control in Neurocritical Care
Kees H. Polderman 185
viii
19 Neuropharmacotherapy
Gretchen M. Brophy and Theresa Human 195
Section 3: General Management and Special Populations

20 General ICU Care of the Neurological Patient
Samuel A. Tisherman and Sara Hefton 213
21 Managing the Postoperative Neurosurgical Patient
Daniel Ripepi and Colleen Moran 224
22 Pediatric Neurocritical Care
Dennis W. Simon and Ericka L. Fink 233
23 Neurocritical Care in Pregnancy
Krystle Shafer and Marie R. Baldisseri 247
24 Management of Intracranial Hypertension in Fulminant
Hepatic Failure
Sajid Kadir and Raghavan Murugan 256
25 Prognostication and Ethics
David Y. Hwang and Douglas B. White 265
26 Brain Death, Organ Donation, and Transplantation
Hilary H. Wang and David M. Greer 276
27 Rehabilitation in Neurocritical Care
Michael E. Reznik and Amy K. Wagner 286
Index 297
Preface
We are pleased to provide this book to those health care providers involved
in the care of critically ill patients with neurological conditions. Our goal in
developing this book is to provide a foundation of knowledge to help guide
the identification, understanding, medical decision-making, and management
of this unique group of patients.
The critically ill neurology patient poses a challenge to many providers as
standard critical care management may not be applicable. In addition, the field
of neurocritical care has grown significantly in the past few years. We have
more knowledge and understanding of the physiology of patients with neur
ological conditions presenting in the intensive care unit (ICU), diagnostic and
monitoring technology has advanced, and treatment options have expanded.
These changes allow us to impact significantly on the care of these patients.
This text strives to provide a simple, straightforward guide to these com-
plex patients. This is meant to be a quick reference that provides focused in-
formation regarding the presentation and management of specific neurological
conditions often seen in the ICU. The authors of these chapters are experts
xi
in this field, and we have been privileged to work with them in completion
of this handbook. We hope you will find it to be a useful tool in your care of
these patients.
Lori A. Shutter and Bradley J. Molyneaux
Contributors
Opeolu Adeoye, MD, MS, Gretchen M. Brophy,

FACEP, FAHA PharmD, BCPS, FCCP,
Vice Chair, Research FCCM, FNCS
Co-Director, UC Stroke Team Professor of Pharmacotherapy
Associate Professor & Outcomes Science and
Department of Emergency Medicine Neurosurgery
University of Cincinnati Virginia Commonwealth University
Cincinnati, OH Medical College of Virginia Campus
Richmond, VA
Nitin Agarwal, MD
Resident Edward A. Burton, MD,
Department of Neurological DPhil, FRCP
Surgery Associate Professor of Neurology
University of Pittsburgh and UPMC Endowed Chair in
School of Medicine Movement Disorders
Pittsburgh, PA University of Pittsburgh School
xiii
of Medicine
Maranatha Ayodele, MD Pittsburgh, PA
Assistant Professor of Clinical
Neurology Jan Claassen, MD, PhD, FNCS
Neurocritical Care Division Associate Professor of Neurology
University of Miami Miller School Head of Neurocritical Care
of Medicine Medical Director, Neurological
Miami, FL Intensive Care Unit
Columbia University College
Marie R. Baldisseri, MD, of Physicians & Surgeons
MPH, FCCM New York, NY
Professor of Critical Care Medicine,
Medicine, and Health Promotion Andrew F. Ducruet, MD
& Development Endovascular Neurosurgeon
Department of Critical Care Barrow Neurological Institute
Medicine Phoenix, AZ
University of Pittsburgh
School of Medicine
Pittsburgh, PA
Jonathan Elmer, MD, MS Sherry Hsiang-Yi Chou, MD,
Contributors
Assistant Professor MSc, FNCS, FCCM

Departments of Emergency Associate Professor
Medicine and Critical Department of Critical Care
Care Medicine Medicine, Neurology, and
University of Pittsburgh Neurosurgery

School of Medicine University of Pittsburgh School

Pittsburgh, PA of Medicine
Pittsburgh, PA
Ericka L. Fink, MD, MS
Associate Professor, Critical Care Theresa Human, PharmD,
Medicine, Pediatrics, and Clinical BCPS, FNCS
& Translational Science Institute Neuroscience Clinical Specialist
Department of Critical Care Barnes-Jewish Hospital
Medicine Washington University
Children’s Hospital of St.Louis, MO
Pittsburgh
Safar Center for Resuscitation David Y. Hwang, MD,
Research FCCM, FNCS
University of Pittsburgh Assistant Professor of Neurology
School of Medicine
xiv
Division of Neurocritical Care and

Pittsburgh, PA Emergency Neurology
Yale School of Medicine
David M. Greer, MD, MA, New Haven, CT
FCCM, FAHA, FNCS,
FAAN, FANA Ashutosh P. Jadhav, MD, PhD
Professor and Chairman Assistant Professor of Neurology
Department of Neurology and Neurological Surgery
Boston University School Director, Vascular Neurology
of Medicine Fellowship
Boston, MA University of Pittsburgh School
of Medicine
Sara Hefton, MD Pittsburgh, PA
Assistant Professor
Departments of Neurology and Ruchira Jha, MD, MS
Neurological Surgery Assistant Professor
Division of Neurotrauma and Departments of Critical Care
Critical Care Medicine, Neurology, Neurological
Thomas Jefferson University Surgery, and Clinical &
Philadelphia, PA Translational Science Institute
Scientist-Safar Center for
Resuscitation Research
University of Pittsburgh School
of Medicine
Pittsburgh, PA
Tudor G. Jovin, MD Colleen Moran, MD
Contributors
Professor of Neurology and Assistant Professor
Neurosurgery Department of Anesthesiology
Director, Stroke Institute Duke University School of Medicine
Co-Director, Center for Durham, NC
Endovascular Therapy
University of Pittsburgh School Raghavan Murugan, MD,
of Medicine MS, FRCP
Pittsburgh, PA Associate Professor of Critical
Care Medicine and Clinical &
Sajid Kadir, MD Translational Science
Intensivist, Geisinger Health System Center for Critical Care Nephrology
Department of Medicine Clinical Research Investigation and
Danville, PA Systems Modeling of Acute Illness
Department of Critical Care
Joshua Keegan, MD Medicine
Clinical Assistant Professor University of Pittsburgh
Co-Director, Neurovascular School of Medicine
ICU—UPMC Altoona Pittsburgh, PA
Department of Critical Care Medicine
Kristine O’Phelan, MD
xv
of Medicine Associate Professor of Clinical
Pittsburgh, PA Neurology
Department of Neurology
Cynthia L. Kenmuir, MD, PhD University of Miami Miller School
Fellow, Departments of Neurology of Medicine
and Neurological Surgery Miami, FL
of Medicine David O. Okonkwo, MD, PhD
Pittsburgh, PA Executive Vice Chair, Clinical
Operations
Namir Khandker, MD Clinical Director, Brain Trauma
Adult Fellow, Neurocritical Care Research Center
Department of Critical Care Medicine Department of Neurological Surgery
University of Pittsburgh School University of Pittsburgh
of Medicine School of Medicine
Pittsburgh, PA Pittsburgh, PA
Deepa Malaiyandi, MD David M. Panczykowski, MD

Assistant Professor Chief Resident
Co-Director, Neurovascular Department of Neurological
ICU—UPMC Altoona Surgery
Department of Critical Care Medicine University of Pittsburgh
University of Pittsburgh School of Medicine
School of Medicine Pittsburgh, PA
Pittsburgh, PA
Kees H. Polderman, MD, PhD Krystle Shafer, MD
Contributors
Professor of Critical Care Medicine Chief Fellow, Adult Critical Care

Department of Critical Care Medicine Medicine—Emergency Medicine
University of Pittsburgh Department of Critical Care
School of Medicine Medicine
Pittsburgh, PA University of Pittsburgh School of

Medicine
Michael E. Reznik, MD Pittsburgh, PA
Assistant Professor of Neurology
and Neurosurgery Dennis W. Simon, MD
Division of Neurocritical Care Assistant Professor
Alpert Medical School, Brown Director, Pediatric
University Neurocritical Care
Providence, RI Department of Critical Care
Medicine
Daniel Ripepi, MD Children’s Hospital of Pittsburgh
Resident Safar Center for Resuscitation
Department of Anesthesiology Research
University of Pittsburgh School of University of Pittsburgh
Medicine School of Medicine
Pittsburgh, PA
xvi
Pittsburgh, PA
Jon C. Rittenberger, MD, MS Jeremy G. Stone, MD

Associate Professor Resident
Department of Emergency Department of Neurological
Medicine, Occupational Therapy, Surgery
and Clinical & Translational University of Pittsburgh
Science Institute School of Medicine
University of Pittsburgh Pittsburgh, PA
School of Medicine
Pittsburgh, PA Samuel A. Tisherman, MD,
FACS, FCCM
Daniel B. Rubin, MD, PhD Professor of Surgery, Program
Fellow, Neurocritical Care in Trauma
Massachusetts General Hospital University of Maryland School
Brigham and Women’s Hospital of Medicine
Harvard Medical School Director, Center for Critical Care
Boston, MA and Trauma Education
Director, Surgical Intensive
Mihai C. Sandulescu, MD Care Unit
Clinical Assistant Professor of University of Maryland
Neurology Medical Center
Geisinger Commonwealth School of RA Cowley Shock Trauma Center
Medicine Baltimore, MD
Scranton, PA
Henrikas Vaitkevicius, MD Douglas B. White, MD, MAS
Contributors
Assistant Professor Neurology Professor of Critical Care Medicine
Brigham and Women’s Hospital UPMC Endowed Chair of Ethics in
Harvard Medical School Critical Care Medicine
Boston, MA Director, Program on Ethics and
Decision Making in Critical Illness
Amy K. Wagner, MD University of Pittsburgh School of
Professor, Physical Medicine & Medicine
Rehabilitation and Neuroscience Pittsburgh, PA
UPMC Endowed Chair Translational
Research Saša A. Živković, MD, PhD
Associate Director Safar Center for Associate Professor of Neurology
Resuscitation Research Department of Neurology
University of Pittsburgh University of Pittsburgh
School of Medicine School of Medicine
Pittsburgh, PA Pittsburgh, PA
Hilary H. Wang, MD, MBA

Resident
Department of Neurology &
xvii
Neurological Sciences
Stanford Medicine
Stanford, CA
Neurocritical Care
Section 1
Neurological
Conditions
Chapter 1
Coma and Low Arousal States

Namir Khandker and Lori A. Shutter
Introduction
In the intensive care unit (ICU), coma and other low arousal states are
common enough that physicians of every specialty will encounter them. They
can present as a medical emergency, and all cases require a thoughtful and or-
ganized approach. The differential diagnosis is broad, with prognosis and man-
agement varying widely depending on etiology. Rapid initial assessment with a
thorough history and physical examination is crucial for timely and appropriate
live-saving management.
3
Definitions and Physiology of Consciousness
Plum and Posner presented the most accepted definition of consciousness
as “the state of full awareness of the self and one’s relationship to the en-
vironment.” Control of consciousness is complex and involves a myriad of
structures, pathways, and neurotransmitters. A basic understanding of these
mechanisms helps clarify the etiology of altered consciousness and guide di-
rected assessment and management.
Consciousness
A simple schematic divides consciousness into two components: content and
arousal. Content refers to all cognitive and affective mental functions of the
brain and is generated by the cortex. Arousal refers to behavior consistent
with alertness and responsiveness to the environment and is regulated by
connections between the brainstem (upper pons and midbrain), diencephalon
(thalamus and hypothalamus), and cortex. Some level of arousal is required
for cognition. Dysfunction of these structures or connections will manifest as
an alteration of mental status.
Consciousness represents a continuum from coma to normal arousal
with sleep being the only normal alteration. Imprecise terms such as leth-
argy, obtundation, and stupor represent different points on the continuum of
arousal. Use of these terms are easily misinterpreted. Better communication
between care teams can be achieved using behavioral descriptions (“patient
will not open eyes to voice but will localize with vigorous sternal rub”) or using
coma scales (described later in this chapter).
Coma
Neurological Conditions
Coma is a failure of neuronal systems responsible for arousal and awareness

resulting in complete unresponsiveness to environmental stimuli. Eyes remain
closed and will not open spontaneously or reactively. Reflexive movements
may remain intact, but often brainstem reflexes will be depressed. Coma is
typically transient, and patients will progress to different arousal states, each
of which has a different prognosis for recovery.
Vegetative State
After a coma, patients may begin to exhibit normal sleep/wake cycles while
Section 1
otherwise remaining unresponsive, which defines a vegetative state. A patient

may open his or her eyes spontaneously but will not track or fixate. Reflexive
movements may become prominent, and random movements of limbs, trunk,
and even vocalizations with tears or smiles may occur. No reproducible pur-
poseful responses will be seen. These findings are consistent with diffuse or
multifocal grey and white matter injury. Lower brainstem function typically
remains intact.
Minimally Conscious State
Patients who show evidence of awareness of external stimuli have entered a
minimally conscious state. Patients continue to have significantly altered con-
4
sciousness with variable levels of arousal and inconsistent responsiveness but

demonstrate one or more reproducible behaviors consistent with awareness.
Examples of responsive behaviors include following simple commands, appro-
priate yes/no (verbal or gesture) responses, or other purposeful behavior.
Patients who consistently demonstrate awareness have progressed out of a
low arousal state.
Etiology of Coma
All etiologies of coma result in neuronal dysfunction involving bilateral cerebral
hemispheres or the arousal pathways. During the evaluation, it is crucial to iden-
tify readily reversible causes before permanent damage occurs. Localization of
the structures involved can give clues to the underlying etiology, which can
be divided into structural and nonstructural. Some etiologies can encompass
features of both structural and nonstructural mechanisms, and mechanisms can
evolve over time. An example would be a tumor with cerebral edema and mass
effect causing generalized nonconvulsive seizures. More than one process can
be active simultaneously, leading to multifactorial etiology of coma. Table 1.1
summarizes common causes of coma and coma mimics.
Structural Etiologies of Coma
Structural etiologies of coma cause neuronal dysfunction by mechanical force.
Causes can include but are not limited to bihemispheric or brainstem injury,
mass effect, hematoma, interruption of cerebral blood flow, and trauma.
Table 1.1 Common Etiologies of Coma and Associated Findings

Etiology Associated Findings
Structural
Trauma
subdural hematoma, epidural hematoma, falls, multiple injuries
contusion, diffuse axonal injury
Neurovascular
intracerebral hemorrhage, subarachnoid Headache, sudden onset,
hemorrhage, basilar artery occlusion, hypertension, progressive neurologic
cerebral venous thrombosis, posterior symptoms, cranial nerve deficits,
reversible encephalopathy syndrome papilledema
Neuroinflammatory Disorders
Chapter 1
acute disseminated encephalomyelitis, Subacute onset, preceding viral illness,
autoimmune encephalitis comorbid autoimmune disease,
preceding new psychiatric diagnoses
Hydrocephalus
intraventricular hemorrhage, Chiari Lethargy, headache, nausea/vomiting
malformation, colloid cyst, ependymal
inflammation
Central nervous system infection
meningitis, viral encephalitis, cerebral Subacute onset, fever, altered mental
5
abscess status, nuchal rigidity, petechial rash,
photophobia/phonophobia
Oncologic process
neoplasia with edema and mass effect, Cachexia, history of malignancy,
cerebral metastases seizure
Metabolic
non-convulsive status epilepticus history of epilepsy, abolished motor
activity, sepsis
post-anoxic injury from cardiac or respiratory hemodynamic instability at
arrest presentation
septic encephalopathy sepsis
fever
electrolyte derangement
hypercapnea
endocrine derangement (dysglycemia, cushingoid appearance, hypotension,
Addisonian crisis, thyroid dysfunction) pendular reflexes
hyperammonemia liver failure, valproic acid use, urea
cycle metabolic disorders
uremia kidney failure
nutritional deficiency substance abuse, alcohol abuse,
history of gastric bypass
(continued)
Table 1.1 Continued
Etiology Associated Findings

Toxins
ethanol, toxic alcohols osmolal gap, metabolic acidosis, renal
failure
tricyclic antidepressants wide QRS complex, cholinergic
toxidrome
SSRI, antipsychotics rigidity, fever, myoclonus
benzodiazepines, opioids bradypnea, hypotension, miosis
Section 1
cyanide acidosis, high cardiac output, almond odor

antiepileptic drugs nystagmus, hyperammonemia
carbon monoxide headache, hypoxia, flushed skin
aspirin metabolic acidosis, respiratory
alkalosis
acetaminophen hepatic failure
SSRI = selective serotonin reuptake inhibitors.
Several exam features can help identify a structural lesion. Often, the depth
6
of coma will be more severe, particularly if the change is abrupt. Focal neuro-
logic symptoms, such as anisocoria, cranial nerve deficits, or asymmetric limb
movements, are highly suggestive of a structural lesion. Funduscopic examina-
tion can reveal signs of increased intracranial pressure from cerebral hernia-
tion, acute hydrocephalus, or subhyaloid hemorrhages which can be seen with
acute subarachnoid hemorrhage. No focal findings need to be present to have
a structurally mediated coma.
Toxic/Metabolic Etiologies of Coma
Nonstructural etiologies of coma cause neuronal dysfunction by derangements
at the cellular level. The variety of metabolic causes is vast and commonly in-
clude all manner of electrolyte derangements, endocrine dysfunction including
thyroid and adrenal glands, and vitamin deficiencies. Energy crises from hy-
poglycemia, hypoxia, fever, or seizure can precipitate coma, particularly in
patients with abnormal cerebral function at baseline. Exogenous substances
including opioids, alcohol, and other drugs can often cause coma via a sec-
ondary metabolic derangement and organ failure (hypoglycemia, hypoxia,
seizure, hyperammonemia, uremia), or structural insult (intracerebral hemor-
rhage or diffuse leukoencephalopathy).
Onset is typically gradual and examination shows diffuse findings, roving eye
movements, and fluctuating symptoms. Myoclonic jerking is typically metabolic
in origin and most commonly seen with anoxic brain injury. Muscle rigidity and
fever may indicate toxic effects of neuroleptics or selective serotonin release
inhibitors. A thorough general exam including vital signs is also paramount to
identifying features of infection, liver failure, chronic endocrine dysfunction,
stigma of chronic heavy metal poisoning or vitamin deficiency, and cardiopul-

monary collapse among other causes.
Coma Mimics
Pathophysiology affecting behavioral responses outside of the mechanisms
for arousal can be frequently mistaken for coma. It is important to distinguish
these states as management and prognosis can vary drastically.
Locked-In Syndrome
Damage to the ventral pons affecting motor pathways can lead to a state
where cognition and awareness are completely intact with extremely limited
Chapter 1
motor function, typically with only vertical eye movements and blinking pre-
served. The most common cause is occlusion of the basilar artery. Other
etiologies are rare and include brainstem hemorrhage, botulism and neuro-
muscular diseases such as end stage amyotrophic lateral sclerosis, fulminant
acute demyelinating inflammatory polyneuropathy, tick paralysis, and fulmi-
nant critical illness polyneuropathy/myopathy.
Prognosis for motor recovery depends on etiology. However, several
studies suggest that quality of life can be preserved despite severe motor im-
7
pairment. Physicians should not limit care solely on the idea that severe motor
impairment would represent an unacceptable quality of life.
Catatonia
Catatonia has a wide variety of presentations with both organic and psychi-
atric causes. Immobility, mutism, and refusal to eat are the most common
signs, and severe forms may be confused with coma. Patients retain aware-
ness and will often report remembering a catatonic episode in its entirety. As
catatonia can be a manifestation of delirium from a general medical cause, a
thorough evaluation for etiology, including medicine reconciliation, electro-
encephalogram (EEG), neuroimaging, and laboratory investigations should be
undertaken prior to diagnosis of psychiatric disease.
Abulia and Akinetic Mutism
Bilateral medial frontal lesions can affect motivation. In severe forms, apathy
can be so extreme that the patient will not speak or move spontaneously
even to eat or drink water. Patients remain alert and aware and may take
several minutes to respond to an examiner. Etiologies can be varied and in-
clude stroke, inflammatory white matter disease, and infections. A slow and
thorough neurologic exam can usually differentiate this syndrome from a
true coma.
Psychogenic Unresponsiveness
Patients with psychogenic coma may exhibit a profound unresponsive state
but have no identifiable structural, metabolic, or toxic disorder. These
episodes are often precipitated by stress and when witnessed may show
efforts to avoid injury while suddenly becoming comatose. Neurologic exam

of reflexes and cranial nerves should be normal, while other aspects may be
inconsistent with repeated exam. Several exam maneuvers can help distin-
guish poor effort and unmask awareness. Examiners may note forcible eye
closing when attempting to examine pupils with a Bell’s phenomenon (eyes
rolling up with intentional eye closure). Behaviors that avoid injury, such as
not allowing a hand to drop on the face or bracing for a fall, are usually pre-
sent. These features are not pathognomonic for psychogenic coma, and an
initial thorough workup should be performed to rule out a true coma or
Section 1
other low arousal state.
Coma Scales
Coma scales are useful tools for communicating depth of coma. The most
common scales in use are the Glasgow Coma Scale (GCS) and the Full
Outline of Unresponsiveness (FOUR) score. See Table 1.2 for grading of
these scales.
Glasgow Coma Scale
8
The GCS was developed to improve the care of traumatic brain injury
in 1974 and remains an easy-to-use adjunct to the physical exam that has
been incorporated into many clinical guidelines. The GCS is the sum of three
assessments: motor, verbal, and eye movement. Although it is tempting to
simplify the GCS into a single summed score, doing so may mask changes
as the same score can be attained even with a significant change in individual
components of the exam. Components of the GCS may be difficult to test
when a patient is intubated or sedated, which also limits the value of GCS
as a stand-alone assessment. Reliability among GCS scores in a patient can
suffer when performed by inexperienced members of care team, particularly
in regards to the motor score. Observing the motion of the elbow to central
stimulation is crucial. Elbows typically move away from the body with pur-
poseful movements and withdrawal compared to the elbow moving toward
the trunk in abnormal flexion or extension responses.
Full Outline of Unresponsiveness
The FOUR score was developed at Mayo Clinic and addresses some of the
issues encountered with GCS scoring. The scale has been validated for use in
emergency departments and ICUs with nurses and physicians alike. There are
four testable components: eye movement, motor response, respiration, and
brainstem reflexes. The FOUR score is distinguished from the GCS by the in-
clusion of items that assess nonverbal cues of consciousness such as voluntary
eye movements and assessments for the intubated patient.
Table 1.2 Coma Scales
Glasgow Coma Scale Full Outline of Unresponsiveness
Eye opening Eye response
1 None 0 eyelids remain closed with pain
2 To pressure 1 eyelids closed but open to pain
3 To speech 2 eyelids closed but open to loud voice
4 Spontaneous 3 eyelids open but not tracing
4 eyelids open, tracking, or blinking to command
Verbal Response Motor response
1 None 0 no response to pain or myoclonic status
2 Sounds 1 extension response to pain
Chapter 1
3 Words 2 flexion response to pain
4 Confused speech 3 localizing to pain
5 Oriented 4 thumbs-up, fist, or peace sign
Best Motor Response Respiration
1 None 0 breathes at ventilator rate or apnea
2 Extension 1 breathes above ventilator rate
3 Abnormal flexion 2 not intubated, irregular breathing
4 Withdrawal 3 not intubated, Cheyne-Stokes breathing
5 Localizing 4 not intubated, regular breathing
9
6 Obeying commands
Brainstem reflexes
0 absent pupil, corneal, and cough
1 absent pupil and corneal
2 absent pupil or corneal
3 one pupil wide and fixed
4 pupil and corneal reflexes present
Initial Evaluation of Coma

Coma is a medical emergency. As such, steps for assessment and management
should occur simultaneously rather than in a step-wise fashion (Figure 1.1).
Airway, Breathing, Circulation, and Cervical Spine
Precautions
As with any other medical emergencies, rapid assessment of a patient’s circula-
tory system, airway, and respiration is a crucial first step to preventing long-term
morbidity. Obtaining adequate intravenous access, appropriate resuscitation,
and correction of other readily reversible electrolyte derangements are the
initial steps of caring for the comatose patient. Implementation of Advanced
Cardiac Life Support (ACLS) protocols take priority.
Coma
Resuscitation, Focused
history/exam, Stat labs
Structural or Unclear Nonstructural

Section 1
CT
Diagnostic Nondiagnostic
Vascular Imaging
MRI
10
EEG
Figure 1.1 Suggested algorithm for initial diagnosis of coma etiology.
Normalization of vital signs is crucial to ensure adequate perfusion of the

brain. Hypoxia should be treated with supplemental oxygen. If airway protec-
tion is a concern and initial measures do not reverse coma, intubation should be
performed without delay with in-line cervical spine immobilization. Hypotension
should be treated with Trendelenburg positioning (if elevated intracranial pres-
sure is not suspected), fluids, and vasopressors. Extreme hypertension (systolic
blood pressure > 250 mmHg or mean arterial pressure > 130 mmHg) should
be treated with intravenous medications such as labetalol or hydralazine or use
of antihypertensive drips such as nicardipine or clevidipine. In the setting of pos-
sible elevated intracranial pressure, care should be taken when treating blood
pressure to avoid lowering of cerebral perfusion pressure below 60 mmHg. In
general, a target mean arterial pressure of 85 mmHg or a decrease of approxi-
mately 15% is acceptable. Hyperthermia should be immediately corrected with
cooling blankets or pads, cold saline bolus, or intravenous cooling catheters.
For most patients, cervical spine precautions are appropriate in the imme-
diate phase. If there is any possibility of traumatic injury, cervical immobiliza-
tion should be part of the initial resuscitation effort.
An EKG early in assessment is helpful for evaluating a primary cardiopul-

monary event and is almost always abnormal in a significant toxic exposure or
electrolyte derangement.
General Physical and Neurologic Assessment
Once the patient has been stabilized, assessment for structural or toxic/met-
abolic etiology of coma is important. Gather as much information as possible
regarding the onset, evolution, and associated symptoms as well as pertinent
past medical history, drug abuse history, and medication list.
A good general physical exam, including analysis of vital signs, can help
narrow a differential diagnosis by suggesting a specific toxidrome, metabolic/
endocrine derangement, or organ failure. The neurologic exam is focused on
identifying a structural lesion versus diffuse neuronal dysfunction. This can usu-
Chapter 1
ally be accomplished by assessing level of consciousness, orientation, cranial
nerves, and motor responses. All assessments should be done off sedation
when possible. Coma scales should be used to reliably communicate deficits
and help identify changes on repeat exams.
Laboratory Studies
Initial laboratory studies are aimed at discovering readily correctable met-
abolic and electrolyte derangements that could possibly reverse coma.
11
A point-of-care glucose should be checked immediately. If a glucose meas
urement is not available, consider empiric treatment with dextrose. In alco-
holics or malnourished patients such as those with gastric bypass, empiric
thiamine is warranted.
Other initial laboratory tests should include a complete blood count and com-
prehensive metabolic panel (sodium, potassium, urea, creatinine). Depending
on the history, additional studies such as calcium, magnesium, liver function
panel, ammonia, urine and blood toxin screen (note: drug levels are more infor-
mative than a positive screen), serum osmolality, and arterial blood gases may
be obtained. Derangements in these values should be relatively acute to cause a
significant change in arousal. Chronic changes should prompt evaluation for an-
other cause of coma. See Table 1.3 for laboratory values compatible with coma.
Table 1.3 Laboratory Values Compatible with Coma

Hyponatremia ≤ 110 mmol/L
Hypernatremia ≥160 mmol/L
Hypercalcemia ≥ 3.4 mmol/L
Hypermagnesemia ≥ 5 µg/L
Hypercapnia ≥ 70 mmHg
Hypoglycemia ≤ 40 mg/dL
Hyperglycemia ≥ 800 mg/dL
Note. Changes should be acute.
A history of panhypopituitarism, thyroid dysfunction, or chronic steroid use
may alert examiners to the possibility of an endocrine derangement. Checking

thyroid function evaluation for Hashimoto’s encephalitis may be warranted. In
cases of suspected steroid deficiency, hypotension may not always be present,
so assessing response to a stress dose of corticosteroids may be diagnostic.
Autoimmune encephalopathies and central nervous system (CNS) malignan-
cies (particularly CNS lymphoma) may also respond to steroids.
Imaging
Cerebral imaging should be pursued unless there is rapid improvement in con-
Section 1
sciousness. To avoid additional injury, this should be obtained after stabiliza-

tion of vital signs and correction of acute metabolic derangements.
Computed tomography (CT) of the head is widely available and most useful
for evaluating intracranial hemorrhage. CT angiography should be considered
in all patients if there is any possibility that the patient’s exam is consistent with
basilar occlusion. Nondiagnostic studies should then prompt consideration for
vascular imaging or magnetic resonance imaging.
Electroencephalogram
Patients with a history of epilepsy, seizure on presentation with treatment
abolished motor symptoms, or coma otherwise unexplained after ini-
12
tial evaluation should be evaluated with EEG. Patients outside of the ICU
should be initially evaluated with 20-minute EEG to avoid delaying other
evaluations and treatments. Any clinically apparent seizure or any patient
with high pretest probability should be empirically treated if EEG is not
readily available.
Status epilepticus is common in critically ill patients and is often comorbid
in the elderly with sepsis. Up to 20% of unresponsive patients in a general
medical ICU may have an abnormal EEG that should be treated, and the ma-
jority of those patients have no clinical correlate to alert caregivers of seizure
activity. Any patient who continues to be comatose despite treatment of pre-
sumed cause in the ICU should be monitored on continuous EEG for up to
48 hours.
Prognosis and Treatments for Coma and

Low Arousal States
Prognosis should not be attempted or commented on in the first 72 hours
of coma. This timeline should be extended to 72 hours post-rewarming for
any patient undergoing therapeutic hypothermia. Aside from brain death
and cardiac arrest (considered separately), no reliable metric is available for
predicting if a patient will recover or how much recovery could be expected.
The cause of coma clearly impacts the prognosis and natural history and can
be broadly divided into traumatic and nontraumatic.
Drug-induced coma generally has a good outcome if supportive care is in-

stituted early and reversible derangements are addressed quickly. Mortality
is relatively low for coma induced by epilepsy or poisoning (up to 7%) versus
patients suffering from a neurovascular catastrophe or post-anoxic injury
(54%– 95%). The most common nontraumatic etiologies are stroke and
post-anoxic injury. Patients without recovery at one month are unlikely to
demonstrate meaningful recovery and may be left with severe disability if un-
consciousness improves.
Traumatic coma is equally difficult to prognosticate early after injury. In con-
trast to nontraumatic causes, one month of coma does not suggest an inability
to recover. Age, neurologic exam at one week, and initial postresuscitation
GCS are the most prognostic factors. For severe traumatic brain injury that
Chapter 1
survives to the ICU, mortality still approaches 30%, with 22% achieving inde-
pendence at six months.
Emerging Treatments
With increasing knowledge of pathways governing arousal, several pharma-
cologic treatments have been studied and shown to improve recovery in se-
lect patients. Generally, patients in persistent vegetative states are left with
variable levels of disability despite treatment due to disruption of the arousal
pathways. Patients in a minimally conscious state may respond better, although
13
the recovery process can be prolonged. See Table 1.4 for commonly used
pharmacological agents.
Deep brain stimulation of thalamic nuclei and epidural spinal cord stimula-
tion has limited data but in a minority of selected patients has been shown to
improve outcomes. More study is required to elucidate pathways involved and
identify patients who could benefit.
Table 1.4 Commonly used Pharmacologic Interventions

for Coma and Low-Arousal States
Medication Mechanism of Action
Amantadine NMDA receptor antagonist, dopamine agonist
Methylphenidate Dopamine and norepinephrine receptor agonist
Modafinil Inhibits norepinephrine and dopamine transport;
glutamate, serotonin, and histamine release
Carbidopa/levodopa Dopamine receptor saturation
Further Reading
Berger J. Stupor and Coma, 7th ed. Daroff R, Jankovic J, Mazziotta J, Pomeroy S, eds.
Philadelphia: Elsevier; 2012.
Bruno MA, Ledoux D, Lambermont B, et al. Comparison of the full outline of unre-
sponsiveness and Glasgow Liege Scale/Glasgow Coma Scale in an intensive care unit
population. Neurocrit Care. 2011;15(3):447–453.
Edlow JA, Figaji A, Samuels O. Emergency neurological life support: subarachnoid hem-
orrhage. Neurocrit Care. 2017;27(1):116–123.
Edlow JA, Rabinstein A, Traub SJ, Wijdicks EFM. Diagnosis of reversible causes of coma.
Section 1
Lancet. 2014;384(9959):2064–2076.
Georgia M De, Raad B. Prognosis of coma after cardiac arrest in the era of hypo-
thermia. Continuum. 2012;18(3):515–531.
Horsting MWB, Franken MD, Meulenbelt J, van Klei WA, de Lange DW. The etiology
and outcome of non-traumatic coma in critical care: a systematic review. BMC
Anesthesiol. 2015;15:65.
Jennett B, Teasdale G, Braakman R, Minderhoud J, Heiden J, Kurze T. Prognosis of
patients with severe head injury. Neurosurgery. 1979;4(4):283–289.
Lulé D, Zickler C, Häcker S, et al. Life can be worth living in locked-in syndrome. Prog
Brain Res. 2009;177:339–351.
Oddo M, Carrera E, Claassen J, Mayer SA, Hirsch LJ. Continuous electroencephalog-
14
raphy in the medical intensive care unit. Crit Care Med. 2009;37(6):2051–2056.
Posner J, Saper C, Schiff N, Plum F. Plum and Posner’s Diagnosis of Stupor and Coma, 4th
ed. Oxford: Oxford University Press; 2007.
Rasmussen SA, Mazurek MF, Rosebush PI. Catatonia: our current understanding of its
diagnosis, treatment and pathophysiology. World J Psychiatry. 2016;6(4):391–398.
Rousseau M-C, Pietra S, Nadji M, Billette De Villemeur T. Evaluation of quality of life in
complete locked-in syndrome patients. Ann Phys Rehabil Med. 2012;55(Suppl. 1):e363.
Shaibani A, Sabbagh MN. Pseudoneurologic syndromes: recognition and diagnosis. Am
Fam Physician. 1998;57(10):2485–2494.
Teasdale G, Maas A, Lecky F, Manley G, Stocchetti N, Murray G. The Glasgow Coma
Scale at 40 years: standing the test of time. Lancet Neurol. 2016;13(8):844–854.
Traub SJ, Wijdicks EF. Initial diagnosis and management of coma. Emerg Med Clin North
Am. 2016;34(4):777–793.
Wijdicks EF. The Practice of Emergency and Critical Care Neurology. New York: Oxford
University Press; 2010.
Chapter 2
Encephalopathy and Delirium

Joshua Keegan and Colleen Moran
Definitions
Although sometimes confused or used interchangeably, delirium and encepha-
lopathy are distinct disorders. Delirium is defined as an acute, often reversible
disturbance in attention (ability to direct, focus, sustain, and shift attention)
that tends to fluctuate, is associated with additional changes in cognition, and
occurs acutely over hours to days. This change must be a direct physiolog-
ical consequence of another medical condition and not better explained as
part of an evolving dementia. Encephalopathy, on the other hand, is a much
broader term encompassing any brain damage or dysfunction and therefore
does not necessitate changes in attention or significant acute fluctuations.
15
Encephalopathy is extremely heterogeneous in nature and has numerous
underlying causes, including a variety of metabolic causes, structural lesions,
traumatic injuries, infection, hypoxic- ischemic injury, and medications.
Consequently, risk factors, detection, and management vary with the under-
lying cause, and a comprehensive review is beyond the scope of this chapter.
While delirious patients often have associated hallucinations, delusions,
sleep disturbances, and abnormal psychomotor activity, none of these are
necessary to make the diagnosis. Of note, delirium may be either hyperac-
tive or hypoactive; hypoactive delirium in particular may be easily missed or
misdiagnosed in intensive care unit (ICU) patients. Much of the knowledge
base regarding delirium has been generated from medical and surgical ICUs,
and thus extrapolation to neurological ICU (NICU) patients requires caution.
Prevalence and Relevance

Delirium is quite common, in some studies affecting up to 80% of mechan-
ically ventilated ICU patients and costing $4 billion to $16 billion annually.
Delirium has been identified as an independent predictor of multiple negative
outcomes, including increased mortality, cost, prolonged ICU and hospital
length of stay, and long-term cognitive impairment (although to some extent
delirium may unmask or identify patients with pre-existing limited cognitive
reserve).
Risk Factors and Causes

Although the pathophysiology of delirium is not well understood, multiple
risk factors and causes have been identified, allowing practitioners to iden-
tify patients at high risk for development of delirium during their ICU stay.
Significant risk factors identified include pre-existing dementia, alcoholism, hy-
pertension, coma, and high severity of medical illness at admission. Notably,
although age is an independent delirium risk factor for non-ICU patients, only
two of six studies found it to be significant in ICU populations. Regardless
Section 1
of underlying risk factors, delirium may be caused or worsened by multiple

contributors, including
• Underlying or complicating disease
o Sepsis
o Organ dysfunction
o Alcohol/drug withdrawal
o Withdrawal from analgesia and sedatives, including opiates and
dexmedetomidine
• Medication exposures
o Opiates (most studies demonstrate no risk, but some do show an
increase)
16
o Benzodiazepines (several studies report strong effects while others

show none)
• Environmental factors
o Physical restraints
o Prolonged immobilization
o Sleep disruption
o Visual impairment
o Hearing impairment
Whenever possible, therapeutic interventions should be chosen so as to limit
their delirium-inducing potential.
Screening
ICU personnel often underestimate the prevalence of delirium, particularly in
its hypoactive form. Delirium detection is improved with the routine use of
screening tools, and a variety of tools exist for this purpose. The Confusion
Assessment Method for the ICU (CAM-ICU) and Intensive Care Delirium
Screening Checklist (ICDSC) have been found to be the most accurate (high
sensitivity and specificity) and repeatable (high interrater reliability), and as
such are the screening tools recommended by the American College of
Critical Care Medicine. The CAM-ICU is performed every shift, with the
ICDSC performed every 24 hours; several studies exist demonstrating the
feasibility of such screening. One important caveat specific to the neurolog-

ically impaired patient is that it can be difficult or impossible to perform the
CAM-ICU in aphasic patients.
Prevention
Given the significant morbidity and mortality associated with delirium, increasing
interest in pharmacologic and nonpharmacologic methods of prevention has
developed. Early mobilization has been found to decrease the prevalence and
duration of delirium in the ICU by as much as 50%. Multicomponent protocols
targeted toward reducing modifiable environmental risk factors for delirium
Chapter 2
have been shown to be helpful in non-ICU settings, although they have not yet
been studied in the ICU. Studies examining pharmacologic preventative strat-
egies, including medications to maintain sleepwake cycles, use of haloperidol,
and use of atypical antipsychotics have been fairly limited, and consequently
no current recommendation regarding such approaches has been made. One
large study in medical and surgical ICUs found that interruption of a previously
prescribed statin also increased risk for delirium, and consequently continuing
such medications may have preventative value.
17
Treatment
Often, treatment of ICU delirium with either typical or atypical antipsychotics
occurs. Despite the frequency of this practice, data supporting it are limited,
with no published evidence supporting treatment with haloperidol and very
limited data suggesting that quetiapine may shorten duration of delirium, al-
though at the expense of increased somnolence. Given the likelihood of poly-
pharmacy and concurrent electrolyte imbalances, ruling out QTc prolongation
via electrocardiogram prior to initiation of these medications should be un-
dertaken. Statins may also have some effect on delirium, particularly in sepsis,
via limitation of production of inflammatory mediators thought to play a role
in delirium pathogenesis. One study found that statin use in the ICU on day
one of sepsis was shown to decrease delirium; however, this study was not a
randomized trial, and the criteria for initiation/continuation of a statin were
not reported.
Sedation and Delirium

Sedation to minimize agitation and anxiety in the ICU setting is often desir-
able from a patient comfort perspective. However, sedatives may also cause
or worsen delirium and result in multiple other complications. The type and
amount of sedative used should be carefully considered to minimize these
negative effects. Of note, when a clear cause for agitation is identified (i.e.,
pain, urinary retention, hypoxia, ventilator dyssynchrony), correcting the un-

derlying cause is often more effective and with fewer adverse effects. ICU
patients frequently experience pain, and most sedatives have minimal anal-
gesic effects. Consequently, a trial of analgesics, potentially including aceta
minophen and nonsteroidal anti-inflammatory agents, should be considered
prior to administration of sedation. In a NICU-specific population, using a
protocolized analgesia-first regimen did not change durations of ventilation,
NICU, or hospital stay but improved patient alertness and decreased patient
pain. A sample of such a protocol can be found in Figure 2.1. Medications
Section 1
commonly used for sedation in the ICU include propofol, benzodiazepines

(e.g., midazolam and lorazepam), dexmedetomidine, and, less frequently,
ketamine and barbiturates. In recent years, there has been a greater shift to-
ward avoiding benzodiazepines as they can exacerbate delirium.
As previously mentioned, both sedative agent and amount should be chosen
to minimize adverse effects. Ideally, the minimum dose necessary for the de-
sired effect should be used, with frequent reassessment of this requirement
Yes Tylenol 650 mg q4 PO/PR/IV PRN

18
Fentanyl 25–50 mcg IV PRN

NeuroICU
>3 boluses/hr Protocol
Pain
Fentanyl infusion 25–50 mcg/hr
Titrate up to 2 mcg/kg/hr (max 200 mcg/hr)
No
Monitor closely
Hold sedation until Riker = 4 Riker < 4 Riker = 4 Daily Sedation Interruption
Restart sedation 50% lower with SBT
Sedation Restart Sedation 25% lower
CAM-ICU qshift
Riker > 4 Minimize restraints
Maximize daytime activity
Propofol infusion 25 mcg/kg/hr Consider Seroquel 25–100 mg
Increase every 15 min by 20 mcg/kg/hr q12 PRN or Haldol 1–2 mg IV
to Riker = 4 if CAM-ICU+
Triglycerides checked q72 hrs
Dexmedetomidine indications:
Delirium
Wean to extubate in agitated or delirious patient
Wean to extubate with difficult airway
Figure 2.1 Sample neurologic ICU sedation protocol.

as a patient’s underlying condition evolves. Two methods of achieving this

goal are daily sedation interruptions and targeting light or minimal, rather than
heavy, sedation. It should be noted that sedation interruptions may occur
more often in the NICU, due to the paramount importance of reliable neuro-
logic exams to guide patient care. However, particularly among patients with
decreased cerebral compliance, sedation interruption may cause intracranial
pressure (ICP) crises, significant swings in cerebral perfusion pressure, and
critical decreases in brain tissue oxygenation, and therefore harm may out-
weigh benefit in certain patients. Maintaining light levels of sedation decreases
the duration of mechanical ventilation and ICU stay. While this results in in-
creased levels of physiological stress (i.e., elevated heart rate, blood pressure,
increased catecholamines), no negative clinical outcomes (such as myocardial
Chapter 2
infarction) have been demonstrated.
Studies of decreased sedation on psychological outcomes have varied, with
some studies showing no or minimal negative psychological outcomes for
sedation interruptions versus light sedation, respectively. One study demon-
strated that light sedation was associated with greater recall and risk of per-
ceiving the ICU experience negatively; however, another demonstrated more
disturbing memories with deep sedation.
Multiple studies comparing sedative agents exist, primarily comparing a
benzodiazepine to either propofol or dexmedetomidine. A review of two
19
studies comparing propofol to benzodiazepines did not find a significant dif-
ference in delirium rates, and five studies comparing dexmedetomidine to
benzodiazepines generally favored dexmedetomidine, although results were
mixed. One multinational randomized controlled trial found a 22.6% absolute
reduction in delirium rates with dexmedetomidine versus midazolam.
In addition to their general effects on ICU patients, sedative medications
have specific effects pertinent to neurological ICU patients. Sedatives often
decrease the cerebral metabolic rate and may thereby limit ischemic damage
in low-flow regions as well as limiting ICP. Additionally, many sedatives in-
cluding benzodiazepines, propofol, and ketamine have antiepileptic properties
which may be desirable in patients who are seizure-prone from a variety of
neurological insults.
Ultimately, many different options exist for the management of agitation.
There is no ideal agent for all situations; specific agents are best chosen based on
the underlying cause of agitation, with consideration of their pharmacology and
side effect profiles. These effects particularly include the potential of these agents
to cause or worsen delirium, a side effect which must be explicitly considered.
Specific Medications
Benzodiazepines
Benzodiazepines are centrally acting GABA agonists with anxiolytic, sedative,
and amnestic effects; undesirable side effects include respiratory depression,
hypotension, and increased delirium. They are primarily metabolized by the
liver, and many have active metabolites prolonging their duration of action.
When used for short-term sedation, midazolam has a faster offset than lor-
azepam; however, studies regarding prolonged use have demonstrated a
longer and more variable duration of action with midazolam, possibly from ac-
cumulation of these metabolites in peripheral tissues. Propylene glycol is a dil-
uent for lorazepam and may result in toxicity with prolonged use at high doses.
Propofol
Propofol acts on multiple central receptors including GABA, resulting in anx
Section 1
iolytic, sedative, amnestic, and hypnotic effects (the amnestic effects at light
doses are less than with benzodiazepines). Propofol is very lipid-soluble and
rapidly redistributes to peripheral tissues, resulting in a rapid offset of effect
with short-term use, which may facilitate frequent neurologic exams in NICU
patients. However, accumulation in these tissues may lead to prolonged du-
ration with heavier use. Propofol is also associated with numerous adverse
effects, including respiratory depression, vasodilation, negative inotropy,
triglyceridemia, and pancreatitis. Triglyceride levels should be checked every
72 hours in patients treated with propofol infusions, and alternate treatments
should be sought if levels become elevated. One of the most concerning ad-
verse effects is propofol infusion syndrome, which has up to a 33% mortality
20
rate. Risk of propofol infusion syndrome is elevated in younger patients on

high rates of infusion (>60 mcg/kg/min). Earliest laboratory abnormalities
include elevated CK and lactic acidosis.
Dexmedetomidine
Dexmedetomidine is a hepatically metabolized centrally acting alpha-2 agonist
with primarily sedative effects; side effects include hypotension and brady-
cardia. As compared to other sedatives, it often results in a much more arous-
able patient and causes less respiratory depression; consequently, it is the only
sedative infusion approved for use in nonintubated ICU patients. Additionally,
dexmedetomidine may decrease opiate requirements.
Opioids
Although not sedatives, opioid medications are very helpful for managing ag-
itation secondary to pain. All opioids act via similar mechanisms, resulting in
fairly similar benefits and side effects. However, patients may respond better
to certain agents, particularly given incomplete cross-tolerance with chronic
use of other opioids. The majority of opioids cause hypotension through a
histamine-mediated pathway which is therefore not reversible with naloxone
(which counteracts somnolence and respiratory depression from opioid tox-
icity). A specific opioid deserving additional discussion is fentanyl. Fentanyl is
fast-acting and generally fairly quick in offset, although tissue accumulation may
occur with prolonged use due to its lipid-soluble structure. As a completely
synthetic opioid, fentanyl does not cause histamine release and therefore does
not intrinsically generate in hypotension. However, hypotension after adminis-

tration may still result in certain patients due to attenuation of a pain-induced
catecholamine response.
Conclusion
Delirium is common in the ICU, is often underrecognized, and is associated
with significant morbidity and mortality. Providers should be aware of the
variety of baseline and modifiable risk factors that have been elucidated, and,
particularly in high-risk populations, screening tools should be utilized to limit
underdiagnosis. In accordance with available evidence , providers
Chapter 2
should attempt delirium prevention through environmental modifications
such as sleep promotion, consideration of side effects when choosing
medications, using the minimal levels of sedation necessary, and institu-
tion of early mobility. Limited data are currently available to guide effective
pharmacologic treatments after delirium occurs, but quetiapine should be
considered. Future research will hopefully further clarify safe and effective
treatments for ICU delirium.
Further Reading
21
American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders,
5th ed. Washington, DC: Author; 2013.
Barr J, Fraser GL, Puntillo K, et al. Clinical practice guidelines for the management of
pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med.
2013;41:263–306.
Beretta L, De Vitis A, Grandi E. Sedation in neurocritical patients: is it useful? Minerva
Anestesiol. 2011;77:828–834.
Devlin JW, Fong JJ, Fraser GL, Riker RR. Delirium assessment in the critically ill. Intensive
Care Med. 2007;33:929–940.
Devlin JW, Roberts RJ, Fong JJ, et al. Efficacy and safety of quetiapine in critically ill
patients with delirium: a prospective, multicenter, randomized, double- blind,
placebo-controlled pilot study. Crit Care Med. 2010;38:419–427.
Egerod I, Jensen MB, Herling SF, Welling KL. Effect of an analgo-sedation protocol for
neurointensive patients: a two-phase interventional non-randomized pilot study. Crit
Care. 2010;14:R71.
Helbok R, Kurtz P, Schmidt MJ, et al. Effects of the neurological wake-up test on clinical
examination, intracranial pressure, brain metabolism and brain tissue oxygenation in
severely brain-injured patients. Crit Care. 2012;16:R226.
Inouye SK, Bogardus ST Jr, Charpentier PA, et al. A multicomponent intervention to
prevent delirium in hospitalized older patients. N Engl J Med. 1999;340:669–676.
Kress JP, Gehlbach B, Lacy M, Pliskin N, Pohlman AS, Hall JB. The long-term psycholog-
ical effects of daily sedative interruption on critically ill patients. Am J Respir Crit Care
Med. 2003;168:1457–1461.
Morandi A, Hughes CG, Thompson JL, et al. Statins and delirium during critical illness: a
multicenter, prospective cohort study. Crit Care Med. 2014;42:1899–1909.

Needham DM, Korupolu R, Zanni JM, et al. Early physical medicine and rehabilitation
for patients with acute respiratory failure: a quality improvement project. Arch Phys
Med Rehabil. 2010;91:536–542.
Neto AS, Nassar AP Jr, Cardoso SO, et al. Delirium screening in critically ill patients: a
systematic review and meta-analysis. Crit Care Med. 2012;40:1946–1951.
Riker RR, Shehabi Y, Bokesch PM, et al. Dexmedetomidine vs midazolam for sedation of
critically ill patients: a randomized trial. JAMA. 2009;301:489–499.
Samuelson KA, Lundberg D, Fridlund B. Stressful experiences in relation to depth of se-
Section 1
dation in mechanically ventilated patients. Nurs Crit Care. 2007;12:93–104.

Schweickert WD, Pohlman MC, Pohlman AS, et al. Early physical and occupational
therapy in mechanically ventilated, critically ill patients: a randomised controlled trial.
Lancet. 2009;373:1874–1882.
Skoglund K, Enblad P, Marklund N. Effects of the neurological wake-up test on intracra-
nial pressure and cerebral perfusion pressure in brain-injured patients. Neurocrit Care.
2009;11:135–142.
Spronk PE, Riekerk B, Hofhuis J, Rommes JH. Occurrence of delirium is severely
underestimated in the ICU during daily care. Intensive Care Med. 2009;35:1276–1280.
Treggiari MM, Romand JA, Yanez ND, et al. Randomized trial of light versus deep seda-
tion on mental health after critical illness. Crit Care Med. 2009;37:2527–2534.
22
Vasilevskis EE, Morandi A, Boehm L, et al. Delirium and sedation recognition using vali-
dated instruments: reliability of bedside intensive care unit nursing assessments from
2007 to 2010. J Am Geriatr Soc. 2011;59(Suppl. 2):S249–255.
Chapter 3
Acute Ischemic Stroke

Cynthia L. Kenmuir and Tudor G. Jovin
Acute Management of Acute

Ischemic Stroke
Early evaluation and treatment is critically important for obtaining good clin-
ical outcomes in acute ischemic stroke (AIS). A brief history should include the
patient’s last known well (LKW) time. Importantly, LKW must be distinguished
from the time that symptoms were first noted. Level 1 evidence of benefit is
available for intravenous (IV) thombolysis administered within four and a half
hours from LKW and for endovascular therapy initiated within twenty-four
hours from LKW. For both treatment modalities a strong time dependency be-
23
tween treatment initiation relative to LKW and likelihood of obtaining a good
outcome has been established. When evaluating the patient, vital signs including
blood pressure (BP) and respiratory status should be monitored regularly.
While BP management in AIS remains somewhat controversial, it is generally
accepted that BP should not be acutely lowered during the initial assessment but
should be maintained below systolic pressure of 220 mmHg and diastolic pres-
sure of 120 mmHg. If a patient qualifies for IV thrombolysis, additional BP low-
ering may be necessary. The National Institutes of Health Stroke Scale (NIHSS)
score (Table 3.1) is a universally used bedside tool that quantifies the severity
of the neurological deficit in a quick and consistent manner. Once cardiopulmo-
nary stability is confirmed, a noncontrasted computed tomography scan of the
head (CTH) is essential in order to rule out intracerebral hemorrhage, deter-
mine extent of infarction, and rule out mass lesions or other potential conditions
(e.g., subdural hematoma) that may render the patient ineligible for IV tissue
plasminogen activator (tPA). The only laboratory study needed emergently
for the purposes of IV tPA administration is serum glucose, with both hyper-
and hypoglycemia addressed. In addition, if the patient has received heparin
within 48 hours; has a known bleeding diathesis; or is taking warfarin, direct
thrombin inhibitors, or other novel anticoagulant drugs, then prothrombin time
(PT)/international normalized ratio (INR), partial thromboplastin time (PTT),
and platelet counts are needed. If eligibility for IV tPA has been confirmed, the
drug should be mixed early to avoid treatment delays. In 2015 the Food and
Drug Administration released an updated version of the prescribing informa-
tion for IV tPA that eliminated many of the previously listed contraindications
Table 3.1 National Institute of Health Stroke Scale Score
Instructions Scale Definition Score
1a. Level of Consciousness: The investigator must choose a response 0 = Alert; keenly responsive. _____
if a full evaluation is prevented by such obstacles as an endotracheal tube, 1 = Not alert; but arousable by minor stimulation to obey, answer, or
language barrier, orotracheal trauma/bandages. A 3 is scored only if the respond.
patient makes no movement (other than reflexive posturing) in response
2 = Not alert; requires repeated stimulation to attend, or is obtunded
to noxious stimulation.
and requires strong or painful stimulation to make movements (not
stereotyped).
3 = Not alert; responds only with reflex motor or autonomic effects or
totally unresponsive, flaccid, and areflexic.
1b. LOC Questions: The patient is asked the month and his/her age. 0 = Answers both questions correctly. _____
The answer must be correct—there is no partial credit for being close. 1 = Answers one question correctly.
Aphasic and stuporous patients who do not comprehend the questions
2 = Answers neither question correctly.
will score 2. Patients unable to speak because of endotracheal intubation,
orotracheal trauma, severe dysarthria from any cause, language barrier, or
any other problem not secondary to aphasia are given a 1. It is important
that only the initial answer be graded and that the examiner not “help” the
patient with verbal or nonverbal cues.
1c. LOC Commands: The patient is asked to open and close the eyes 0 = Performs both tasks correctly. _____
and then to grip and release the nonparetic hand. Substitute another one 1 = Performs one tasks correctly.
step command if the hands cannot be used. Credit is given if an unequivocal
2 = Performs neither tasks correctly.
attempt is made but not completed due to weakness. If the patient does
not respond to command, the task should be demonstrated to him or
her (pantomime), and the result scored (i.e., follows none, one, or two
commands). Patients with trauma, amputation, or other physical impediments
should be given suitable one-step commands. Only the first attempt is scored.
2. Best Gaze: Only horizontal eye movements will be tested. Voluntary or 0 = Normal. _____
reflexive (oculocephalic) eye movements will be scored, but caloric testing 1 = Partial gaze palsy; gaze is abnormal in one or both eyes but forced
is not done. If the patient has a conjugate deviation of the eyes that can be deviation or total gaze paresis is not present.
overcome by voluntary or reflexive activity, the score will be 1. If a patient
2 = Forced deviation; or total gaze paresis not overcome by the
has an isolated peripheral nerve paresis (CN III, IV, or VI), score a 1. Gaze
oculocephalic maneuver.
is testable in all aphasic patients. Patients with ocular trauma, bandages,
pre-existing blindness, or other disorder of visual acuity or fields should be
tested with reflexive movements, and a choice made by the investigator.
Establishing eye contact and then moving about the patient from side to
side will occasionally clarify the presence of a partial gaze palsy.
3. Visual: Visual fields (upper and lower quadrants) are tested by 0 = No visual loss. _____
confrontation, using finger counting or visual threat, as appropriate. 1 = Partial hemianopia.
Patients may be encouraged, but if they look at the side of the moving
2 = Complete hemianopia.
fingers appropriately, this can be scored as normal. If there is unilateral
blindness or enucleation, visual fields in the remaining eye are scored. 3 = Bilateral hemianopia blind including cortical blindness).
Score 1 only if a clear-cut asymmetry, including quadrantanopia, is
found. If patient is blind from any cause, score 3. Double simultaneous
stimulation is performed at this point. If there is extinction, patient
receives a 1, and the results are used to respond to item 11.
4. Facial Palsy: Ask—or use pantomime to encourage—the patient to 0 = Normal symmetrical movements. _____
show teeth or raise eyebrows and close eyes. Score symmetry of grimace in 1 = Minor paralysis (flattened nasolabial fold, asymmetry on smiling).
response to noxious stimuli in the poorly responsive or noncomprehending
2 = Partial paralysis (total or near-total paralysis of lower face).
patient. If facial trauma/bandages, orotracheal tube, tape, or other physical
barriers obscure the face, these should be removed to the extent possible. 3 = Complete paralysis of one or both sides (absence of facial
movement in the upper and lower face).
5. Motor Arm: The limb is placed in the appropriate position: extend 0 = No drift; limb holds 90 (or 45) degrees for full 10 seconds. _____
the arms (palms down) 90 degrees (if sitting) or 45 degrees (if supine). 1 = Drift; limb holds 90 (or 45) degrees, but drifts down before full 10
Drift is scored if the arm falls before 10 seconds. The aphasic patient is seconds; does not hit bed or other support.
encouraged using urgency in the voice and pantomime, but not noxious
2 = Some effort against gravity; limb cannot maintain (if cued) 90 (or
stimulation. Each limb is tested in turn, beginning with the nonparetic
45) degrees, drifts down to bed, but has some effort against gravity.
arm. Only in the case of amputation or joint fusion at the shoulder, the
examiner should record the score as untestable (UN), and clearly write 3 = No effort against gravity; limb falls.
the explanation for this choice. 4 = No movement.
UN = Amputation or joint fusion, explain: ___________
5a. Left Arm
5b. Right Arm
(continued )
Table 3.1 Continued
6. Motor Leg: The limb is placed in the appropriate position: hold the 0 = No drift; leg holds 30 degree position for full 5 seconds. _____
leg at 30 degrees (always tested supine). Drift is scored if the leg falls 1 = Drift; leg falls by the end of the 5-second period but does not hit bed.
before 5 seconds. The aphasic patient is encouraged using urgency in the
2 = Some effort against gravity; leg falls to bed by 5 seconds, but has
voice and pantomime, but not noxious stimulation. Each limb is tested in
some effort against gravity.
turn, beginning with the nonparetic leg. Only in the case of amputation or
joint fusion at the hip, the examiner should record the score as untestable 3 = No effort against gravity; leg falls to bed immediately.
(UN), and clearly write the explanation for this choice. 4 = No movement.
UN = Amputation or joint fusion, explain: ___________
5a. Left Leg
5b. Right Leg
7. Limb Ataxia: This item is aimed at finding evidence of a unilateral 0 = Absent. _____
cerebellar lesion. Test with eyes open. In case of visual defect, ensure 1 = Present in one limb.
testing is done in intact visual field. The finger-nose-finger and heel-shin
2 = Present in two limbs.
tests are performed on both sides, and ataxia is scored only if present out
of proportion to weakness. Ataxia is absent in the patient who cannot UN = Amputation or joint fusion, explain: ____________
understand or is paralyzed. Only in the case of amputation or joint fusion,
the examiner should record the score as untestable (UN), and clearly
write the explanation for this choice. In case of blindness, test by having
the patient touch nose from extended arm position.
8. Sensory: Sensation or grimace to pinprick when tested, or withdrawal 0 = Normal; no sensory loss. _____
from noxious stimulus in the obtunded or aphasic patient. Only sensory 1 = Mild-to-moderate sensory loss; patient feels pinprick is less sharp
loss attributed to stroke is scored as abnormal and the examiner should or is dull on the affected side; or there is a loss of superficial pain with
test as many body areas (arms [not hands], legs, trunk, face) as needed pinprick, but the patient is aware of being touched.
to accurately check for hemisensory loss. A score of 2, “severe or
2 = Severe to total sensory loss; patient is not aware of being touched
total sensory loss,” should only be given when a severe or total loss of
in the face, arm, and leg.
sensation can be clearly demonstrated. Stuporous and aphasic patients
will, therefore, probably score 1 or 0. The patient with brainstem stroke
who has bilateral loss of sensation is scored 2. If the patient does not
respond and is quadriplegic, score 2. Patients in a coma (item 1a = 3) are
automatically given a 2 on this item.
9. Best Language: A great deal of information about comprehension 0 = No aphasia; normal. _____
will be obtained during the preceding sections of the examination. For 1 = Mild-to-moderate aphasia; some obvious loss of fluency or facility
this scale item, the patient is asked to describe what is happening in the of comprehension, without significant limitation on ideas expressed
attached picture, to name the items on the attached naming sheet, and to or form of expression. Reduction of speech and/or comprehension,
read from the attached list of sentences. Comprehension is judged from however, makes conversation about provided materials difficult or
responses here, as well as to all of the commands in the preceding general impossible. For example, in conversation about provided materials,
neurological exam. If visual loss interferes with the tests, ask the patient examiner can identify picture or naming card content from patient’s
to identify objects placed in the hand, repeat, and produce speech. The response.
intubated patient should be asked to write. The patient in a coma (item
2 = Severe aphasia; all communication is through fragmentary expression;
1a = 3) will automatically score 3 on this item. The examiner must choose
great need for inference, questioning, and guessing by the listener.
a score for the patient with stupor or limited cooperation, but a score
Range of information that can be exchanged is limited; listener carries
of 3 should be used only if the patient is mute and follows no one-step
burden of communication. Examiner cannot identify materials provided
commands.
from patient response.
3 = Mute, global aphasia; no usable speech or auditory comprehension.
10. Dysarthria: If patient is thought to be normal, an adequate sample 0 = Normal. _____
of speech must be obtained by asking patient to read or repeat words 1 = Mild-to-moderate dysarthria; patient slurs at least some words
from the attached list. If the patient has severe aphasia, the clarity of and, at worst, can be understood with some difficulty.
articulation of spontaneous speech can be rated. Only if the patient
2 = Severe dysarthria; patient’s speech is so slurred as to be unintelligible
is intubated or has other physical barriers to producing speech, the
in the absence of or out of proportion to any dysphasia, or is mute/
examiner should record the score as untestable (UN), and clearly write an
anarthric.
explanation for this choice. Do not tell the patient why he or she is being
tested. UN = Intubated or other physical barrier, explain: ____________
11. Extinction and Inattention (formerly Neglect): Sufficient 0 = No abnormality. _____
information to identify neglect may be obtained during the prior testing. If 1 = Visual, tactile, auditory, spatial, or personal inattention or
the patient has a severe visual loss preventing visual double simultaneous extinction to bilateral simultaneous stimulation in one of the sensory
stimulation, and the cutaneous stimuli are normal, the score is normal. If modalities.
the patient has aphasia but does appear to attend to both sides, the score
2 = Profound hemi-attention or extinction to more than one
is normal. The presence of visual spatial neglect or anosagnosia may also
modality; does not recognize own hand or orients to only one side of
be taken as evidence of abnormality. Since the abnormality is scored only
space.
if present, the item is never untestable.
https://www.ninds.nih.gov/sites/default/files/NIH_Stroke_Scale.pdf
and warnings that were not considered to be grounded in evidence. In re-
sponse to these changes, in 2016, the American Heart Association released new
guidelines for IV thrombolysis with a thorough evidence-based review of each
potential inclusion and exclusion criterion. The new guidelines have eased many
of the previous exclusion criteria as follows.
IV tPA inclusion criteria:
• Diagnosis of ischemic stroke causing measurable disabling neurological
deficit
• Treatment can begin within four and a half hours of LKW
• Age >18 years
SECTION 1
IV tPA exclusion criteria:

• Sustained elevated BP (systolic >185 mmHg or diastolic >110 mmHg).
Eligible for tPA if BP can be stabilized below 185/110. Goal for first 24
hours after tPA is below 180/105.
• Blood glucose concentration <50 mg/dL.
• Presentation suspicious for subarachnoid hemorrhage or aortic dissection.
• CTH reveals large area of infarction.
• Known/suspected bleeding diathesis including platelets <100,000/mm3,
heparin within 48 hours with elevated PTT, current anticoagulation with
28
INR >1.7 or PT >15 sec, current use of direct thrombin or factor Xa inhib-
itor within the past 48 hours with elevated active PTT, INR, platelet count,
ecarin clotting time, thrombin time, or factor Xa activity assay.
IV tPA additional precautions (determination to be made by a physician with
expertise in stroke care):
• Active internal bleeding or recent major surgery/arterial puncture at a
noncompressible site
• Recent major trauma, stroke, or myocardial infarction
• Previous intracranial hemorrhage not including microhemorrhages, se-
vere head trauma within three months, or intracranial surgery within
three months
• Intracranial neoplasm or unruptured arteriovenous malformation (AVM)
aneurysm
• Pregnancy
• Seizure at onset
The risk of symptomatic intracranial hemorrhage associated with IV tPA treat-
ment in stroke mimics like complicated migraine and seizure has been esti-
mated as <1%. As such, in case the diagnosis of ischemic stroke is not certain,
further ancillary testing to confirm cerebral ischemia (such as magnetic reso-
nance imaging [MRI]) prior to administration of IV tPA for a suspected stroke
is generally not recommended due to inherent time delays. IV tPA at a dose
of 0.9 mg/kg (maximum dose of 90 mg) given as a bolus of 10% followed
by continuous infusion of the remaining dose over one hour has been firmly
established to improve clinical outcome in a time-dependent fashion despite a

6% risk of symptomatic hemorrhage and is currently considered the standard
of care for the treatment of AIS regardless of etiology.
Recently published randomized endovascular trials have established
endovascular therapy as the standard of care for the treatment of AIS in
patients with proximal anterior circulation large vessel occlusion (LVO) who
are ineligible for IV tPA or who have persistent LVO despite having received
IV tPA. The highest benefit from endovascular therapy in AIS is thought to
be obtained in patients with a large mismatch, that is, a relatively small is-
chemic core (area of irreversible ischemic damage) in the presence of a large
Chapter 3
penumbra (area of ischemia that is functionally impaired but structurally intact
and is at imminent risk of infarction if blood flow is not restored). The 2018
American Heart Association/American Stroke Association guidelines have
endorsed endovascular therapy based on level 1 evidence if all of the following
criteria have been met:
• prestroke modified Rankin score 0 to 1
• causative occlusion of the internal carotid artery or proximal middle cere-
bral artery (MCA) (M1)
• age ≥18 years
• NIHSS score of ≥6
29
• treatment can be initiated (groin puncture) within 24 hours of
symptom onset
• for patients presenting within 6 hours of symptom onset, Alberta Stroke
Program Early CT Score (see Figure 3.1) of ≥6
• for patients presenting between 6 and 24 hours of symptom onset, MRI
diffusion weighted imaging or CT perfusion with core-penumbra mismatch
meeting DAWN or DEFUSE3 eligibility criteria
At most endovascular centers, noninvasive vessel imaging is obtained prior
to referral for endovascular therapy to identify the target lesion and for
preprocedure planning. Essential information provided by imaging for the
purposes of patient selection for endovascular therapy in AIS besides pres-
ence of hemorrhage includes presence and location of large vessel occlusion,
size of ischemic core, and to a lesser extent size of tissue at risk. Since timing
is critical for good outcomes in AIS treatment, CT angiography with perfu-
sion imaging is often the preferred modality because it conveys the necessary
information for patient selection faster than MRI/magnetic resonance angi-
ography although MRI is widely accepted to better estimate the core volume
compared to CT-based modalities. Recent data suggests that, in cases of AIS
with a large mismatch between the severity of clinical exam and extent of in-
farction on noncontrasted CTH (with or without a hyperdense vessel sign),
it may be reasonable to proceed directly from CTH to the angio-suite for
diagnostic angiography immediately followed by endovascular treatment if an
appropriate lesion is identified. It is essential to understand that similar to IV
tPA, endovascular therapy for AIS is exquisitely time dependent and the time
30 SECTION 1
Figure 3.1 Alberta Stroke Program Early CT Score (ASPECT) scoring scheme.
The upper row demonstrates axial CT cuts of ganglionic ASPECTS level [M1-M3, insula (I), lentiform
nucleus (L), caudate nucleus (C), posterior limb of internal capsule (IC)]. The lower row demonstrates CT
cuts of supraganglionic ASPECTs level (M4–M6). All axial cuts are reviewed for scoring. Subtract 1 point
from a total of 10 points for any area (M1-6, I, L, C, IC) that is hypodense on CT to obtain ASPECTS.
Image courtesy of Puetz V, Dzialowski I, Hill MD, Demchuk AM. The Alberta Stroke Program.
expenditure necessary for obtaining information from imaging may lower the
likelihood of a good clinical outcome. It has been estimated that for every
minute of ischemia 1.9 million neurons are lost, and endovascular reperfusion
trials have established that that every 30-minute delay in reperfusion is associ-
ated with a 10% reduction in good clinical outcome.
General ICU Care of the AIS Patient

General supportive care of the patient with AIS can lead to reduced
complications and improved outcomes. Airway support and ventilatory
assistance are recommended for patients with decreased consciousness or

bulbar dysfunction that leads to airways compromise. Supplemental oxygen
should be used to maintain an oxygen saturation >94%. A bedside dysphagia
screen should be performed prior to oral intake, and a temporary nasogas-
tric tube may be required. Formal swallowing evaluation by a speech ther-
apist should be performed prior to considering percutaneous gastric tube
placement. Euvolemia should be maintained with isotonic crystalloids, and
any arrhythmia suspected to lower cardiac output should be corrected.
Dextrose-containing fluids are not recommended due to the risk of hyper-
glycemia, which has been associated with poor outcomes. Hypoglycemia (glu-
Chapter 3
cose <60mg/dL) should be corrected, and normoglycemia should be the goal.
Hyperthermia >38ºC should be treated with antipyretic medications and/or
surface cooling. Sources of hyperthermia should be investigated and suspected
infections treated with appropriate antibiotics. Anemia should be avoided, as
should transfusions unless hemoglobin is less than 7 mg/100 ml3. While clinical
seizures should be treated with appropriate medications, prophylactic antisei-
zure medications are not recommended. Pharmacologic prophylaxis for deep
venous thrombosis of immobilized patients should be initiated on admission.
Aspirin should be administered within 48 hours of symptom onset if not oth-
erwise contraindicated.
Hypertension should be cautiously managed in the patient with AIS.
31
Systemic hypertension in AIS is a compensatory mechanism to preserve ce-
rebral perfusion, and aggressive BP reduction has been shown to worsen
outcomes. For patients who have not received IV or intra- arterial (IA)
thrombolysis, the current guideline is to allow permissive hypertension if tol-
erated up to systolic BP 220 mmHg and diastolic BP 120 mmHg, and it may
be reasonable to consider gradual reduction of BP by up to 15% over the
first 24 hours. In the case of large vessel occlusive disease not amenable
to endovascular therapy, BP augmentation may be considered beginning
with conservative treatments like an IV fluid bolus and supine body position.
Although there is limited evidence to support BP augmentation in AIS, in
some centers more aggressive BP support with IV vasopressors is tempo-
rarily used to improve cerebral perfusion pressure and thus prevent progres-
sion of ischemia.
ICU Care after IV-tPA Administration

In addition to the general supportive care guidelines discussed previously, care
of patients who have undergone IV tPA treatment follows well-established
protocols. BP and neurologic monitoring are required every 15 minutes for two
hours following administration of tPA, then every 30 minutes for the next six
hours, and then once every hour for 16 hours. BP should be strictly maintained
below a systolic pressure of 180 mmHg and a diastolic pressure of 105 mmHg
using continuous infusions of easily titratable antihypertensive medications
Table 3.2 Radiographic Classification of the Spectrum

of Hemorrhagic Transformation
Hemorrhagic Classification Radiographic Appearance
Hemorrhagic infarction type 1 (HI1) Small hyperdense petechiae
Hemorrhagic infarction type 2 (HI2) More confluent hyperdensity throughout the
infarct zone; without mass effect
Parenchymal hematoma type 1 (PH1) Homogenous hyperdensity occupying <30%
of the infarct zone; some mass effect
Parenchymal hematoma type 2 (PH2) Homogenous hyperdensity occupying >30%
of the infarct zone; significant mass effect. Or,
SECTION 1
any homogenous hyperdensity located beyond

the borders of the infarct zone
if needed (e.g., nicardipine or clevidipine). An emergent noncontrasted CTH

should be obtained with any change in neurologic exam, new headache, acute
increase in BP, or vomiting. Repeat head imaging is recommended 24 hours
after IV tPA administration to assess for the presence of intracranial hemor-
rhage (Table 3.2) or large territorial infarction prior to initiating antiplatelet or
anticoagulation therapy.
Symptomatic intracranial hemorrhage (ICH; defined as NIHSS in-
32
crease of 4 points or higher) is estimated to occur in up to 6% of patients

treated with IV tPA and usually occurs within the first 24 hours. If ICH
is suspected, emergent blood samples for complete blood count, PT/
PTT/INR, type and screen, and fibrinogen should be collected in addition
to obtaining an emergent CTH. Although there is no current universal
guideline for reversal of IV tPA in the setting of acute hemorrhage, a rea-
sonable approach includes aggressive BP reduction below 140/90 and the
use of cryoprecipitate for fibrinogen <100. Some institutional protocols
may include the administration of epsilon aminocaproic acid, tranex-
amic acid, or prothrombin complex concentrate. Overall, thrombolysis-
associated symptomatic hemorrhages carry a poor prognosis. Orolingual
angioedema has been reported in up to 5% of patients treated with IV
tPA. It usually occurs contralateral to the ischemic hemisphere and is most
commonly associated with concomitant angiotension converting enzyme
inhibitor use and insular or frontal lobe infarctions. Though angioedema is
usually mild and transient, it can be treated with IV ranitidine, diphenhy-
dramine, or methylprednisolone.
ICU Care after IA Thrombectomy

Although there are no specific guidelines for postprocedural care following
endovascular thrombectomy for AIS, most institutions follow similar principles
as for post-IV thrombolysis including a BP cap to prevent reperfusion injury.
A reasonable systolic BP cap in the absence of evidence could be between 140

and 180 mmHg depending on the resulting angiographic reperfusion and/or
the presence of any untreated large vessel stenosis. In general, if reperfusion
is complete, a lower BP cap is recommended.
Additionally, the puncture site should be inspected for signs of developing
hematoma or frank hemorrhage, and peripheral pulses distal to the puncture
site should be assessed. BP and heart rate should be closely monitored for
signs of instability. Relative hypotension, pallor, lightheadedness, chest pain,
or shortness of breath can be early signs of blood loss from a retroperito-
neal hemorrhage after a high femoral puncture. If a retroperitoneal hemor-
Chapter 3
rhage is suspected, pressure should be held at the arterial site and the patient
should be hemodynamically stabilized, including crystalloid and/ or blood
cell transfusion, before obtaining CT imaging of the abdomen/pelvis with
CT angiography of the femoral vessels. Occasionally, severe retroperitoneal
hemorrhages and/or pseudoaneurysm formation may require acute vascular
surgery intervention.
Similar to the care of AIS patients treated with IV thrombolytics, close neu-
rological monitoring of the post–endovascular treatment patient is required
to allow early detection of post–endovascular treatment complications such
as parenchymal hematoma, vessel reocclusion, and in cases of large infarcts
malignant edema with secondary deterioration due to brain swelling and
33
herniation. Hemorrhagic complications may be managed with reversal of
antithrombotic agents (if any were used), strict BP control, and neurosurgical
evacuation. Reocclusion may at times require reintervention if the size of the
infarct is acceptably small.
ICU Care of Large Territorial Infarction

and Malignant Edema
Cerebral edema after a large territorial infarction can cause significant clin-
ical deterioration and in some cases may become life-threatening. Cytotoxic
edema typically peaks three to four days after the initial injury, though reperfu-
sion of infarcted tissue can speed the swelling considerably. Malignant edema
typically refers to a MCA large territorial infarction with swelling that results in
significant early midline shift or clinical deterioration. It is estimated to occur
in 5% to 10% of all ischemic strokes and up to 30% of MCA hemispheric
infarctions. Patients with large posterior fossa infarctions are also at increased
risk of developing life-threatening edema and must be monitored closely.
Monitoring for Edema Progression
Reduced level of arousal in the absence of other reversible medical conditions
remains the most predictive sign of malignant edema and death. In addition to
higher scores on the NIHSS, the development of nausea/vomiting within the
first 24 hours and systolic BP >180 mmHg within the first 12 hours have been
associated with malignant edema. Although serial CTH scans are commonly
used to monitor edema progression after large territorial infraction, currently

there is no data to support this practice. Within the first six hours, an infarct
volume of >82cc on MRI diffusion weighted imaging (DWI) and within 14
hours after stroke onset an infarct volume of >145cc on DWI are strongly
predictive of malignant edema. Early CTH scan with large hypodensity or ab-
sence of perfusion in two-thirds of the MCA territory have also been associ-
ated with increased risk of herniation.
Management of Edema
SECTION 1
In addition to avoiding hyperglycemia, hyperthermia, anemia, and hypoxemia

as described here, avoiding hypercarbia and consequently cerebral vasodila-
tion is critical for patients at increased risk of developing malignant edema.
If intubation and mechanical ventilation are required, sedation should be
minimized when possible to maintain the neurologic examination. Minimally
sedating agents like dexmedetomidine are preferred if needed for patient
comfort. The head of the bed can be elevated to 20º to 30º to encourage
venous drainage. Pharmacologic deep venous thrombosis prophylaxis is en-
couraged, though anticoagulation should be discontinued with an INR goal of
<1.4. Although osmotic therapy is not indicated as prophylactic treatment for
malignant edema, it is reasonable to institute hyperosmotic therapy for treat-
34
ment of progressive malignant edema with suspected increased intracranial

pressure (ICP). However, it is important to recognize that the clinical efficacy
of osmotic therapy for the management of complications related to edema
associated with large ischemic stroke has not been established by random-
ized trials. Mannitol administered IV at doses of 0.25 to 1.0 g/kg over 20
minutes does lower ICP and can be redosed after four to six hours if needed
while avoiding hyperosmolarity >320 mOsm/kg and a persistent elevation
in the osmolar gap. The use of mannitol is generally reserved as a tempo-
rary ICP-lowering measure prior to surgical decompression. Hypertonic sa-
line has also been shown to reduce ICP in patients with clinically suspected
transtentorial herniatiation. Hypertonic saline is commonly administered for
ICP reduction as a 30 cc IV bolus of 23.4% saline, a 250 cc bolus of 3% sa-
line, or as a continuous infusion of 3% saline. A goal sodium range should
be established (typically 140–150 mEq/L), and treatment titrated to maintain
that goal. Care should be taken to maintain euvolemia when administering
osmotic agents as mannitol is a potent diurectic and can cause hypovolemia
while hypertonic saline is a volume expander and can cause hypervolemia and
exacerbated heart failure. Notably, no data exists comparing dosing regimens
of these osmotic agents, so institutional protocols should be followed when
available. Hyperventilation with a goal PaCO2 of 30 to 35 mmHg can also be
used in the intubated patient as a temporizing measure to reduce ICP prior
to decompressive surgery. Intravenous glyburide may improve outcomes and
reduce mortality when administered to patients with large ischemic infarcts
based on preliminary results from phase 2 randomized controlled trial (the

phase 3 study is in advanced planning stages).
Decompressive Surgery
Decompressive craniectomy can be life saving for patients with malignant ce-
rebral edema. A 2007 pooled analysis of three randomized controlled trials
showed a significant mortality benefit (78% medical management, 29% surgical
management—absolute risk reduction 50%) of decompressive surgery for
patients 60 years old or younger with large hemispheric infarction who devel-
oped a decrease in their level of arousal within 48 hours. The number needed
Chapter 3
to treat for survival irrespective of functional outcome was two (four for sur-
vival with modified Rankin Scale <3 and two for survival with modified Rankin
Scale <4). Similarly, in 2014 DESTINY II results were published reporting a
significant reduction in mortality for patients 60 years of age or older who
were treated with decompressive craniectomy, although at one-year follow-
up all of the patients remained dependent. Decompressive craniectomy is
recommended for patients 60 years of age or younger with large territo-
rial infarctions and may be appropriate for patients older than 60, although
presurgical counseling concerning expected postsurgical functional status and
quality of life is needed.
Large cerebellar infarcts can be fatal due to edema resulting in obstructive
35
hydrocephalus or brainstem compression by upward transtentorial hernia-
tion or downward tonsillar herniation. To date there are no large randomized
trials comparing the efficacy of suboccipital decompression versus conserv
ative management. The current consensus is that radiographic evidence of
cerebellar edema accompanied by clinical deterioration justifies immediate
suboccipital craniectomy with dural expansion. Given the additional risk of
upward transtentorial herniation associated with ventriculostomy, it is also
recommended that ventriculostomy to relieve obstructive hydrocephalus
should always be accompanied by decompressive suboccipital craniectomy.
Decompressive surgery after a cerebellar infarct in the absence of brainstem
infarction leads to acceptable functional outcome in many patients.
Further Reading
Broderick JP, Palesch YY, Demchuk AM, et al. Endovascular therapy after intravenous
t-PA versus t-PA alone for stroke. N Engl J Med. 2013;368:893–903.
CAST (Chinese Acute Stroke Trial) Collaborative Group. CAST: randomised placebo-
controlled trial of early aspirin use in 20,000 patients with acute ischaemic stroke.
Lancet. 1997;349:1641–1649.
Demaerschalk BM, Kleindorfer DO, Adeoye OM, et al. Scientific rationale for the inclu-
sion and exclusion criteria for intravenous Alteplase in acute ischemic stroke: a state-
ment for healthcare professionals from the American Heart Association/American
Stroke Association. Stroke. 2016;47:581–641.
Fiorelli M, Bastianello S, Kummer von R, et al. Hemorrhagic transformation within 36
hours of a cerebral infarct: relationships with early clinical deterioration and 3-month
outcome in the European Cooperative Acute Stroke Study I (ECASS I) cohort.
Stroke. 1999;30:2280–2284.
Goyal M, Menon BK, vanZwam WH, et al. Endovascular thrombectomy after large-
vessel ischemic stroke: a meta-analysis of individual patient data from five random-
ized trials. Lancet. 2015;387:1723–1731.
Hacke W, Kaste M, Bluhmki E, et al. Thrombolysis with alteplase 3 to 4.5 hours after
acute ischemic stroke. N Engl J Med. 2008;359:317–329.
International Stroke Trial Collaborative Group. The International Stroke Trial (IST): a
randomised trial of aspirin, subcutaneous heparin, both, or neither among 19435
SECTION 1
patients with acute ischaemic stroke. Lancet. 1997;349:1569–1581.

Jauch EC, Saver JL, Adams HP, et al. Guideline for the early management of patients
with acute ischemic stroke: a guideline for healthcare professional from the American
Heart Association/American Stroke Association. Stroke. 2013;44:870–947.
Juttler E, Unterberg A, Woitzik J, et al. Hemicraniectomy in older patients with exten-
sive middle-cerebral-artery stroke. N Engl J Med. 2014;370:1091–1100.
Mozaffarian D, Benjamin EJ, Go AS, et al.; on behalf of the American Heart Association
Statistics Committee and Stroke Statistics Subcommittee. Heart disease and stroke
statistics—2016 update: a report from the American Heart Association. Circulation.
2015;133:e38–360.
National Institute of Neurological Disorders and Stroke and Stroke rt- PA Stroke
36
Study Group. Tissue plasminogen activator for acute ischemic stroke. N Engl J Med.
1995;333:1581–1587.
Oppenheim C, Samson Y, Manai R, et al. Prediction of malignant middle cerebral artery
infarction by diffusion-weighted imaging. Stroke. 2000;31:2175–2181.
Pfefferkorn T, Eppinger U, Linn J, et al. Long- term outcome after suboccipital
decompressive craniectomy for malignant cerebellar infarction. Stroke.
2009;40:3045–3050.
Powers WJ, Derdeyn CP, Biller J, et al. 2015 AHA/ASA focused update of the 2013
guidelines for the early management of patients with acute ischemic stroke regarding
endovascular treatment. Stroke. 2015;46:3020–3035.
Saver JL. Time is brain—quantified. Stroke. 2006;37:263–266.
Sheth K, Elm J, Beslow L, et al. Glyburide Advantage in Malignant Edema and Stroke
(GAMES-RP) trial: rationale and design. Neurocrit Care. 2016;24:132–139.
Thomalla GJ, Kucinski T, Schoder V, et al. Prediction of malignant middle cerebral artery
infarction by early perfusion-and diffusion-weighted magnetic resonance imaging.
Stroke. 2003;34:1892–1899.
Tsivgoulis G, Zand R, Katsanos A, et al. Safety of intravenous thrombolysis in stroke
mimics: prospective 5- year study and comprehensive meta- analysis. Stroke.
2015;46:1281–1287.
Vahedi K, Hofmeijer J, Juettler E, et al. Early decompressive surgery in malignant infarc-
tion of the middle cerebral artery: a pooled analysis of three randomized controlled
trials. Lancet Neurol. 2007;6:215–222.
Wijdicks EF, Sheth KN, Carter BS, et al. Recommendations for the management of ce-
rebral and cerebellar infarction with swelling: a statement from the American Heart
Association/American Stroke Association. Stroke. 2014;45:1222–1238.
Chapter 4
Intracerebral Hemorrhage
Opeolu Adeoye
Introduction
Spontaneous intracerebral hemorrhage (ICH) accounts for 10% to 15% of all
strokes but 50% of overall stroke mortality. While there remain no clinical trial
proven treatments for improving functional outcomes after ICH, trials in the
past decade have contributed significantly to knowledge and clinical manage-
ment of the disease. In this chapter, we first briefly discuss the natural history
and clinical course of ICH. We then discuss the ways in which the trials in the
past decade have attempted to mitigate secondary injury given the expected
clinical course. Lastly, we discuss ongoing clinical trials and potential future
directions in the management of ICH.
37
Natural History and Clinical Course of ICH
ICH typically presents with sudden onset neurological deficits that may be
similar to that observed in ischemic stroke. Brain imaging with computed to-
mography (CT) or magnetic resonance imaging is necessary to distinguish is-
chemic stroke from ICH. Early deterioration occurs in up to 25% of ICH cases
and is associated with increased morbidity and mortality. Predictive scores
exist (ICH; FUNC), but should be used with extreme caution if at all due to
the potential for labeling a patient inappropriately and contributing to a sense
of nihilism. Goals of care discussions in this patient population should use
shared decision-making measures that focus on the patient’s own values and
preferences related to quality of life.
Hemorrhage Expansion
Historically, the hemorrhage component of the ICH was thought to be
static and deterioration was thought secondary to cerebral edema and mass
effect. However, a salient report on early and serial CT scanning following
ICH presenting within six hours of symptom onset showed that 26% of ICH
patients experience hemorrhage expansion within an hour of the baseline CT
with an additional 12% expanding by 24 hours. Hemorrhage expansion was
associated with early neurological deterioration and worse clinical outcomes.
Hemorrhage expansion may occur within the brain parenchyma or egress
into the ventricular system. Intraventricular hemorrhage (IVH) occurs in up

to 40% of all ICH cases and is associated with poorer outcome. The contri-
bution of IVH to poor outcome may be due to acute hydrocephalus and po-
tential herniation from intracranial pressure (ICP) crisis. Patient selection and
current use of EVDs to relieve hydrocephalus and elevated ICP after ICH/
IVH is highly variable. Given the contributions of hemorrhage volume and IVH
to overall ICH mortality, multiple trials in the past decade have investigated
interventions to reduce the impact of ICH volume and IVH on outcomes
following ICH.
SECTION 1
Cerebral Edema
Cerebral edema also contributes to ICH morbidity and mortality. Early base-
line relative (to the ICH volume) edema volume has been found to be a strong
predictor of three-month functional outcome, with higher relative edema
volumes associated with poorer outcomes. However, another study found
that edema volume did not predict outcome after adjusting for baseline ICH
volume. The contribution of early perihematomal edema to ICH outcomes
remains unclear.
Delayed cerebral edema occurs three days to two weeks after ICH onset
in up to 10% of cases and is associated with mass effect and poor outcome.
38
Edema volume is estimated to peak at five to six days after symptom onset
and may be mediated by red blood cell lysis, iron deposition, or slow thrombin
release. The clinical data on the contribution of cerebral edema to ICH
outcomes and the pathophysiologic understanding of edema after ICH have
led to multiple completed and ongoing clinical trials.
Clinical Trials
Hemostasis
Total ICH volume is by far the strongest predictor of outcomes following ICH.
Thus, a strong area of focus of ICH research in the past decade has been
hemostasis and prevention of hemorrhage expansion. Two prospective ran-
domized trials of recombinant factor VIIa (rFVIIa) for treatment of ICH have
been completed that showed conflicting results. The first was a phase IIb study
that randomized patients to placebo or two different doses of rFVIIa. While
underpowered for efficacy ascertainment, the study did demonstrate a signif-
icant reduction in the percentage change in ICH volume and improvement in
the rates of both mortality and severe disability with rFVIIa treatment com-
pared with placebo. Unfortunately, the follow-up phase III trial did not find
a functional outcome benefit with rFVIIa treatment although a reduction in
hemorrhage expansion at 24 hours was confirmed.
Possible explanations for the conflicting results of the two rFVIIs trials in-
clude randomization imbalances (lower IVH rates in the treatment arms in
the phase IIb trial and higher rates in the phase III trial) and higher rates of
stroke and myocardial infarction in the treatment arms in the phase III trial.
Since patients who are not destined to experience hemorrhage expansion
were exposed to the potential thrombotic complications of rFVIIa without
any potential for benefit, ongoing ICH studies are attempting to use contrast
extravasation on CT angiography or the “spot sign” to identify patients at risk
for hemorrhage expansion and who thus may benefit from administration of
a hemostatic agent (Figure 4.1). Ongoing or recently completed trials include
the United States–based STOP-IT study, the Canadian SPOTLIGHT trial, and
the Australian STOP-AUST trial.
Anticoagulant Associated ICH
Chapter 4
Anticoagulant associated ICH is conservatively estimated to account for 20%
of all ICH cases. Anticoagulant use is associated with larger ICH volumes and
increased risk of hemorrhage expansion with consequent poor outcomes.
Although warfarin remains the most commonly used oral anticoagulant,
the introduction of the direct oral anticoagulants (DOAC) for which there
are limited reversal agents further complicate the clinical management of
anticoagulant ICH.
For warfarin associated ICH, the faster the international normalized ratio
(INR) is normalized, the better the chances of preventing hemorrhage expan-
39
sion. A recently published study of 50 patients randomized to fresh frozen
plasma (FFP) versus four factor prothrombin complex concentrates (PCC) as
first-line therapy for reversing warfarin coagulopathy. The trial was terminated
early based on efficacy with 67% of 27 in the PCC group achieving an INR
1.2 or less within three hours of treatment (vs. 9% of 23 patients in the FFP
(a) (b) (c)
Figure 4.1 Demonstration of a “spot sign” and hematoma expansion. Panel A: Baseline
CT, Panel B: Baseline CTA with spot sig (arrow), Panel C: Follow-up CT five hours later
performed after neurological deterioration.
Early CT Score in clinical practice: what have we learned? Int J Stroke. 2009 Oct;4(5):354–364.
group). For warfarin-associated ICH, four-factor PCCs should be adminis-
tered as soon as possible for reversal of coagulopathy.

DOACs include factor II inhibitors (dabigatran) and factor Xa inhibitors
(apixaban, rivaroxaban, and edoxaban). Late in 2015, the Food and Drug
Administration approved use of idarucizumab for emergent reversal of
dabigatran. ICH patients on dabigatran should be treated rapidly with this
novel reversal agent. No reversal agents are approved for factor Xa inhibitors;
andexanet alpha is a promising agent that has recently been granted FDA
approval. Many centers administer PCCs to factor Xa inhibitor treated
patients with life-threatening hemorrhage based on in vitro improvement in
SECTION 1
coagulation studies (see also c hapter 20, Table 20.2).

Platelet Transfusion in ICH
About 30% of ICH patients are on antiplatelet therapy pre-ICH, and pre-ICH
antiplatelet use is associated with a 27% greater odds of mortality after ICH.
Thus, platelet transfusion in ICH patients on antiplatelet therapy pre-ICH has
been an area of interest. The platelet transfusion versus standard care after acute
stroke due to spontaneous cerebral hemorrhage associated with antiplatelet
therapy trial was recently published. In this study, platelet transfusion was associ-
ated with increased risk of death and dependency at three months. Thus, routine
platelet transfusion for ICH patients on antiplatelet therapy should be avoided in
40
spontaneous ICH. In the event of a planned surgical intervention, the known risks
versus potential benefit of platelet transfusions should be carefully considered.
Blood Pressure Management
A decade ago, whether or not elevated blood pressure in the setting of acute
ICH was associated with hemorrhage volume and hemorrhage expansion was
unclear. However, there was also concern about the safety of aggressively re-
ducing blood pressure in patients who may have elevated ICP, thereby precipi-
tating ischemia. Thus, recommendations were based on “suspicion of elevated
ICP.” Two phase III randomized clinical trials of blood pressure management in
ICH have now been completed.
While neither study (INTERACT-2, ATACH-II) found a significant func-
tional outcome benefit of blood pressure reduction to less than 140 mm Hg,
both trials have established there are also no safety concerns with reducing
blood pressure to a target of 140 mm Hg.
Surgical Management of ICH
The debate on surgical management of ICH has focused primarily on
supratentorial ICH since surgical decompression and hemorrhage evacuation
can be life-saving in cerebellar ICH. The trials discussed here thus focus on
surgical management of deep/lobar ICH and not cerebellar ICH.
Craniotomy has been investigated in two phase III randomized clinical trials
(Surgical Trial of ICH, STICH, and STICH-II), and neither showed any overall
benefit of surgery over medical management. Minimally invasive techniques
have been suggested as a means of avoiding secondary tissue damage that
may occur with craniotomy with hemorrhage evacuation. The largest trial
published to date randomized 377 patients with basal ganglionic ICH to min-
imally invasive craniopuncture with clot aspiration versus medical therapy
alone. While there was no significant difference in mortality rates at 90 days,
patients in the medical arm were significantly more likely to be dependent at
90 days. A confirmatory trial is ongoing. An independent phase III trial, the
Minimally Invasive Surgery Plus Recombinant Tissue Plasminogen Activator
(rt-PA) For ICH Evacuation (MISTIE III) is ongoing to compare medical
therapy alone versus minimally invasive surgery plus intermittent rt-PA for
clot evacuation.
The Clot Lysis Evaluation of Accelerated Resolution of Intraventricular
Chapter 4
Hemorrhage III trial randomized ICH patients with occluded third or fourth
ventricles due to IVH who had routine external ventricular drains (EVDs)
placed to alteplase versus saline irrigation for removal of IVH. While more
patients treated with alteplase achieved IVH clearance, alteplase treatment did
not improve functional outcome at 180 days. Therefore routine use of intra-
ventricular alteplase cannot be recommended at this time.
Decompressive hemicraniectomy without tissue excision is a well-established
treatment for improving mortality and functional outcomes in selected patients
with severe ischemic stroke. While there are no data to support its use in ICH,
decompressive hemicraniectomy is occasionally used in the clinical management
41
of patients with large, primarily deep ICH, and multiple case series have been
published. A small randomized trial is ongoing, the SWITCH trial (Swiss Trial of
Decompressive Craniectomy versus Best Medical Treatment of Spontaneous
Supratentorial Intracerebral Hemorrhage), and should contribute to future
knowledge on the potential utility of this approach in selected ICH patients.
Neuroprotection
The contribution of red blood cell lysis, iron deposition, slow thrombin re-
lease, and other processes to cerebral edema and potentially secondary brain
injury after ICH has been termed “neurohemoinflammation.” Deferoxamine
mesylate is an iron chelator that has been found to reduce perihematomal
edema and neuronal damage in animal models of ICH, and phase II testing is in
progress (iDEF). Other early phase research with neuroprotective strategies
using hypothermia, statins, and erythropoietin is also underway.
Conclusions and Future Directions

A great deal of progress has been made in providing definitive answers to
clinical questions in ICH in the past decade. While the efficacy of a variety
of interventions remain unproven, the safety of many of these interventions
has been established. Thus, while widespread adoption of the interventions
cannot be recommended, clinicians may be confident in making a choice to
pursue a particular course of clinical action on a patient-by-patient basis.
The next decade of ICH research and treatment will be further informed by
ongoing, recently completed, and future clinical trials. STOP-IT, SPOTLIGHT,

and STOP-AUST are investigating the potential utility of imaging to select ICH
patients for hemostatic therapy. The MISTIE III trial will establish whether
minimally invasive surgery is beneficial. The iDEF trial will determine if ad-
ministering deferoxamine is promising as a neuroprotective and may improve
functional outcomes. Other potentially neuroprotective interventions will
move from early to late stage clinical trial testing.
Lastly, the chapter has broadly discussed medical and surgical management
of ICH agnostic to the physical location of the patient. It is important to em-
SECTION 1
phasize that the critical care management and duration of intensive care unit
(ICU) stay for some ICH patients represents a pivotal period in the course
toward recovery. Much more work needs to be done to define optimal ICU
care and the contribution of such care to ICH outcomes.
Further Reading
Adeoye O, Broderick JP. Advances in the management of intracerebral hemorrhage. Nat
Rev Neurol. 2010;6(11):593–601.
Anderson CS, Huang Y, Wang JG, et al. Intensive Blood Pressure Reduction in Acute
42
Cerebral Haemorrhage Trial (INTERACT): a randomised pilot trial. Lancet Neurol.

2008;7(5):391–399.
Badenes R, Bilotta F. Neurocritical care for intracranial haemorrhage: a systematic re-
view of recent studies. Br J Anaesth. 2015;115(Suppl. 2):ii68–74.
Baharoglu MI, Cordonnier C, Salman RA, et al. Platelet Transfusion versus Standard
Care after Acute Stroke due to Spontaneous Cerebral Haemorrhage Associated
with Antiplatelet Therapy (PATCH): a randomised, open-label, phase 3 trial. Lancet.
2016; 387:2605–2613.
Brott T, Broderick J, Kothari R, et al. Early hemorrhage growth in patients with intrace-
rebral hemorrhage. Stroke. 1997;28(1):1–5.
Flaherty ML, Kissela B, Woo D, et al. The increasing incidence of anticoagulant-
associated intracerebral hemorrhage. Neurology. 2007;68(2):116–121.
Flaherty ML, Tao H, Haverbusch M, et al. Warfarin use leads to larger intracerebral he-
matomas. Neurology. 2008;71(14):1084–1089.
Fung C, Murek M, Z’Graggen WJ, et al. Decompressive hemicraniectomy in patients
with supratentorial intracerebral hemorrhage. Stroke. 2012;43(12):3207–3211.
Hemphill JC, 3rd, Bonovich DC, Besmertis L, Manley GT, Johnston SC. The ICH
score: a simple, reliable grading scale for intracerebral hemorrhage. Stroke.
2001;32(4):891–897.
Hemphill JC 3rd, Farrant M, Neill TA, Jr. Prospective validation of the ICH score for 12-
month functional outcome. Neurology. 2009;73(14):1088–1094.
Hemphill JC, 3rd, Greenberg SM, Anderson CS, et al. Guidelines for the management
of spontaneous intracerebral hemorrhage: a guideline for healthcare professionals
from the American Heart Association/ American Stroke Association. Stroke.
2015;46(7):2032–2060.
Heuts SG, Bruce SS, Zacharia BE, et al. Decompressive hemicraniectomy without clot
evacuation in dominant-sided intracerebral hemorrhage with ICP crisis. Neurosurg
Focus. 2013;34(5):E4.
Krishnan K, Scutt P, Woodhouse L, et al. Glyceryl trinitrate for acute intracerebral hem-
orrhage: results from the Efficacy of Nitric Oxide in Stroke (ENOS) trial, a subgroup
analysis. Stroke. 2016;47(1):44–52.
Leira R, Davalos A, Silva Y, et al. Early neurologic deterioration in intracerebral hemor-
rhage: predictors and associated factors. Neurology. 2004;63(3):461–467.
Mayer SA, Brun NC, Begtrup K, et al. Recombinant activated factor VII for acute intra-
cerebral hemorrhage. N Engl J Med. 2005;352(8):777–785.
Mayer SA, Brun NC, Begtrup K, et al. Efficacy and safety of recombinant activated factor
Chapter 4
VII for acute intracerebral hemorrhage. N Engl J Med. 2008;358(20):2127–2137.
Mendelow AD, Gregson BA, Fernandes HM, et al. Early surgery versus initial conserva-
tive treatment in patients with spontaneous supratentorial intracerebral haematomas
in the International Surgical Trial in Intracerebral Haemorrhage (STICH): a
randomised trial. Lancet. 2005;365(9457):387–397.
Mendelow AD, Gregson BA, Rowan EN, et al. Early surgery versus initial conserva-
tive treatment in patients with spontaneous supratentorial lobar intracerebral
haematomas (STICH II): a randomised trial. Lancet. 2013;382(9890):397–408.
Pollack CV, Jr., Reilly PA, Eikelboom J, et al. Idarucizumab for dabigatran reversal. N Engl
J Med. 2015;373(6):511–520.
Qureshi AI, Palesch YY, Barsan WG, et al. Intensive blood-pressure lowering in patients
43
with acute cerebral hemorrhage. N Engl J Med. 2016; 375(11):1033–1043.
Siegal DM, Curnutte JT, Connolly SJ, et al. Andexanet alfa for the reversal of factor Xa
inhibitor activity. N Engl J Med. 2015;373(25):2413–2424.
Steiner T, Diringer MN, Schneider D, et al. Dynamics of intraventricular hemorrhage
in patients with spontaneous intracerebral hemorrhage: risk factors, clinical impact,
and effect of hemostatic therapy with recombinant activated factor VII. Neurosurgery.
2006;59(4):767–773; discussion 773–764.
Steiner T, Poli S, Griebe M, et al. Fresh frozen plasma versus prothrombin complex con-
centrate in patients with intracranial haemorrhage related to vitamin K antagonists
(INCH): a randomised trial. Lancet Neurol. 2016;15(6):566–573.
Ziai WC, Tuhrim S, Lane K, et al. A multicenter, randomized, double-blinded, placebo-
controlled phase III study of Clot Lysis Evaluation of Accelerated Resolution of
Intraventricular Hemorrhage (CLEAR III). Int J Stroke. 2014;9(4):536–542.
Chapter 5
Subarachnoid Hemorrhage
Sherry Hsiang-Yi Chou
Epidemiology and Risk Factors

Aneurysmal subarachnoid hemorrhage (SAH) affects more women than men
with a ratio of approximately 1.24:1, and its incidence varies greatly world-
wide, with the highest incidences in Finland and Japan. Within the United
States, SAH affects 14.5/100,000. People of black and Hispanic races have a
higher incidence than white Americans. Overall, SAH incidence increases with
age, peaking at age 40 to 60 years.
SAH risk factors include hypertension, smoking, heavy alcohol use, sym-
pathomimetic drug use, female sex, presence of cerebral aneurysm, history
of prior SAH, and family history of cerebral aneurysms and SAH. While no
44
specific genes have been identified to confirm risk for SAH, certain genetic
syndromes such as autosomal dominant polycystic kidney disease and type
IV Ehlers-Danlos syndrome do confer a higher risk. Hypertension control,
smoking cessation, and reduced alcohol consumption are potentially modifi-
able SAH risk factors and are generally recommended. The presence of ce-
rebral aneurysms >7 mm in size is associated with a higher risk of aneurysm
rupture and SAH, and pre-emptive treatment of high-risk cerebral aneurysms
is a known preventative strategy for SAH.
Clinical Presentation and Diagnosis

The most common presentation of SAH is a sudden onset of “the worse
headache of my life” as described by about 80% of patients who are capable
of giving a history. About 20% patients may present after a sentinel, less se-
vere headache. Headaches may be associated with nausea/vomiting, stiff
neck, brief or persistent loss of consciousness, and focal neurological deficits.
SAH can present with sudden onset of loss of consciousness, coma, collapse,
or seizure-like symptoms. Despite recent advances in medical diagnostics, ap-
proximately 12% of SAH are misdiagnosed, and those misdiagnosed are four
times more likely to die or have severe disability one year after SAH. The most
common error in SAH misdiagnosis is the failure to obtain a non-contrast
head computed tomography (CT) scan.
Diagnosis of SAH is often made with a non-contrast CT scan or a lumbar
puncture (LP). Sensitivity of non-contrast CT for detection of SAH is highest
in the first 12 hours (98%–100%) but decays over time: 93% at 24 hours and
only 57% to 85% at six days post-SAH. For patients with a negative CT scan
but high clinical suspicion of SAH, LP should be performed. Cerebral spinal
fluid (CSF) analyses include relative quantities of red and white blood cell
counts, protein and glucose content, and the presence of xanthrochromia
and bilirubin. CSF analysis interpretation should be based on guidelines and
take into account the potential time lapse between LP and the onset of SAH.
Advances in brain magnetic resonance imaging (MRI) sequences such as gra-
dient echo, proton density, diffusion-weighted imaging, and fluid-attenuated
Chapter 5
inversion recovery have improved the sensitivity of MRI to detect SAH in the
setting of negative CT and high clinical suspicion. However, MRI should not
replace LP as a diagnostic choice for SAH.
Detection of potential cerebral aneurysm associated with SAH is most
commonly achieved by CT angiogram (CTA) and conventional cerebral angiog-
raphy, which is the gold standard test. Magnetic resonance angiogram has been
used, but its sensitivity and specificity is inferior to CTA. Depending on aneurysm
size, location, and study quality, CTA has a sensitivity of 77% to 100% and speci-
ficity of 79% to 100% for detecting cerebral aneurysms. Small aneurysms may not
be reliably detected by CTA. In cases where the index of suspicion of aneurysmal
45
SAH is high based on clinical presentation or a diffuse pattern of SAH on the CT,
a negative CTA should be followed by cerebral angiography. Approximately 20%
to 25% of all SAH cases undergoing cerebral angiogram will not have an identifi-
able bleeding source. Repeating cerebral angiography approximately seven days
later may increase aneurysm detection by 1% to 2%. One particular variant of
SAH, perimesencephalic SAH, presents with a stereotypical SAH pattern, has no
detectable aneurysm on cerebral angiography, and typically has a benign course
with a good outcome compared to aneurysmal SAH.
There are both clinical and radiographic scales to help risk-stratify SAH
patients for the likelihood of developing secondary complications and outcome
prognosis. SAH clinical grading is defined based on the severity of neurologic
dysfunction at the time of presentation. Multiple clinical studies have consist
ently shown that SAH clinical grade is one of the strongest predictors of out-
come. Two well-validated SAH clinical grading systems are the Hunt and Hess
score and the World Federation of Neurological Surgeons classification (Table
5.1). Radiographic scales are used to quantify the amount of blood and predict
risk for secondary vasospasm. The most commonly used radiographic scale in
SAH is the Fisher grade, originally developed in the 1980s. The Fisher grade was
assigned based on an initial head CT upon SAH presentation within five days of
hemorrhage and found that the presence of blood >1 mm in the subarachnoid
space was predictive of vasospasm. However, Fisher grade 4, which denotes
the presence of intraparenchymal or intraventricular hemorrhage without thick
SAH, appears to have lower vasospasm risk. As a result, a modified Fisher scale
was later developed and is now more commonly used (Table 5.2).
Table 5.1 SAH Clinical Brading Scales

Grade Hunt and Hess World Federation of
Neurological Surgeons
GCS Motor Deficit
I Asymptomatic, mild headache, slight 15 Absent

nuchal rigidity
II Moderate to severe headache, nuchal 14–13 Absent
rigidity, no neurological deficit except
cranial nerve palsy
SECTION 1
III Drowsiness, confusion, mild focal deficit 14–13 Present

IV Stupor, moderate to severe hemiparesis 12–7 Present or absent
V Coma, decerebrate posturing 6–3 Present or absent
GCS = Glasgow Coma Scale.
Acute SAH Management

Immediate SAH management includes delivery of any emergent critical care
necessary to ensure the patient has adequate airway, breathing, and circula-
46
tion, and the prevention or treatment of any impending cardiopulmonary col-

lapse. It is equally important to diagnose and treat any ongoing or impending
hyperacute neurological deterioration due to acute hydrocephalus and/or
aneurysm rebleeding.
Hydrocephalus
Acute hydrocephalus is a well-known complication of SAH. Hydrocephalus
is a clinical diagnosis and should be considered in any SAH patient with un-
explained impaired consciousness. Treatment of acute symptomatic hydro-
cephalus with urgent insertion of external ventricular drain or CSF diversion is
Table 5.2 SAH Radiographic Grading Scales

Grade Fisher Modified Fisher
0 n/a No SAH or IVH
1 No blood Focal or diffuse SAH <1 mm, no IVH
2 Diffuse deposition or vertical layer Focal or diffuse SAH <1 mm, with IVH
of SAH <1 mm thick
3 Localized clot or vertical layers of Focal or diffuse SAH >1 mm, no IVH
SAH >1 mm thick
4 IPH or IVH with diffuse or no Focal or diffuse SAH >1 mm, with IVH
subarachnoid blood
SAH = subarachnoid hemorrhage; IVH = intraventricular hemorrhage; IPH = intraparenchymal
hemorrhage.
generally associated with neurological improvement. Lumbar drainage of CSF
in SAH-related hydrocephalus is reported to be safe and may be associated
with early improvement but does not improve long-term SAH outcome.
Re-bleeding Risk
Aneurysm rebleeding is associated with high mortality and morbidity. Risk for
aneurysm rebleeding is greatest in the first 2 to 12 hours after SAH and may
be as high as 13.6% within the first 24 hours. Following the first 24 hours, if
the aneurysm is not obliterated, the SAH rebleeding risk goes up cumulatively
over time, increasing by 1% to 2% per day up to 20% to 30% for the first
month. Urgent treatment and obliteration of the bleeding cerebral aneurysm
Chapter 5
is the best way to decrease rebleeding risk following SAH.
Prior to definitive treatment of the bleeding cerebral aneurysm, possible
medical interventions to minimize the risk of rebleeding include strict blood
pressure control, bed rest, and antifibrinolytic therapies such as aminocaproic
acid or tranexamic acid infusion. Bed rest alone is not sufficient to prevent
aneurysm rebleeding. While studies show conflicting results on blood pressure
control and risk of rebleeding, it is generally recommended to maintain good
control to avoid both hypertension and hypotension. There are conflicting
data on the utility of antifibrinolytic therapy. While there is some evidence that
short-term use may reduce risk of early aneurysm rebleeding, prolonged use
47
is not recommended due to increased risk of thrombotic events.
Aneurysm Treatment
Definitive treatment with complete obliteration of the bleeding aneurysm after
SAH should be performed as early as possible. Typical treatments include open
surgical clipping and endovascular coil embolization of the aneurysm. Choice of
the optimal modality of aneurysm treatment depends on aneurysm factors such
as size, morphology, location, as well as patient factors such as periprocedural risk
and comorbidities. One large prospective randomized trial compared aneurysm
clipping to coil embolization in patients clinically amendable to either modality
and showed no significant difference in mortality rate at one year (8.1%–10.1%)
but greater disability with open surgical approach (21.6% vs. 15.6%). In con-
trast, the coil embolization group had higher rebleeding rate (2.9% vs. 0.9%)
and a higher incidence of needing additional aneurysm treatments. At this time,
aneurysm management is determined based on institutional preferences and
feasibility with increasing use of endovascular techniques.
Early Brain Injury, Cerebral Vasospasm,

and Delayed Cerebral Ischemia
Angiographic vasospasm occurs in 30% to 70% of all SAH patients,
typically starting three to five days post-SAH ictus, achieving maximal
narrowing at 5 to 14 days post-SAH, and resolves over two to four weeks.
Approximately half of all patients develop delayed neurological ischemic
deficit, which may resolve or progress to cerebral infarction. Despite max-

imal therapy, 15% to 20% of SAH patients will develop ischemic infarcts or
die from vasospasm.
The pathogenesis and potential therapy for cerebral vasospasm has been
the focus of intense research for decades. For many years, clinicians and
scientists focused on the hypothesis that narrowing of cerebral vessels, or
cerebral vasospasm, leads to cerebral ischemia and that is the main patho-
physiologic process in SAH-associated brain injury. Recent data and clinical
trial results now suggest that while severe vasospasm may lead to ischemic in-
SECTION 1
farction in SAH, the pathogenesis of SAH-associated brain injury is multifacto-

rial. Cohort studies and clinical trials have not shown a consistent association
between angiographic vasospasm and SAH outcome. In fact, only 50% of
patients with vasospasm may show corresponding symptoms of cerebral is-
chemia, while some patients with severe angiographic vasospasm may not
demonstrate any symptoms of ischemia.
Diagnosis of delayed cerebral ischemia (DCI) can be problematic for sev-
eral reasons. By definition, DCI is a decline in neurologic function attrib-
utable to ischemia from cerebral vasospasm with other etiologies excluded.
There is inherent ambiguity in such a definition in a heterogeneous and crit-
ically ill patient population. Neurologic examinations are of limited sensi-
48
tivity to detect DCI in poor clinical grade SAH patients due to their poor
baseline neurologic function to begin with, and yet these patients are in fact
the highest risk population for DCI and poor outcome after SAH. Current
practice and treatment protocols for vasospasm and DCI surveillance vary
between individuals and institutions; this lack of a standardized approach is
partly due to limited sensitivity and specificity in detecting impending vaso-
spasm related ischemic brain injury. More advanced monitoring and surveil-
lance methods such as the use of biomarkers and multimodal monitoring are
under investigation, but there is yet limited data on the efficacy and utility of
these novel methods.
While DCI is consistently shown to predict death and disability in SAH
clinical trials, there is no clear evidence that treatment and resolution of
angiographic vasospasm is associated with incidence of DCI or with SAH
outcome. Prophylactic hypervolemia and cerebral angioplasty aimed to pre-
vent vasospasm do not improve outcome and may cause increased mor-
bidity. To date, oral nimodipine 60 mg administered every four hours for
21 days is the only therapy that has been shown to improve SAH functional
outcome in a prospective randomized clinical trial. However, the same trial
showed that nimodipine did not reduce the incidence of angiographic va-
sospasm, suggesting that the beneficial effects of nimodipine may be me-
diated mechanisms unrelated to vasospasm. Similarly, many drugs that are
efficacious in reducing angiographic cerebral vasospasm did not improve SAH
outcome in clinical trials. These include L-type calcium channel blockers such
as nicardipine and endothelin receptor 1A antagonists such as clozosentan.
Potentially neuroprotective agents such as statins and high dose magne-
sium started within 96 hours of SAH also did not improve SAH outcome
in large phase III clinical trials. Oral nimodipine remains the only medication
recommended in SAH for potential reduction of DCI.
Historically, clinicians treated vasospasm and DCI with “triple- H”
therapy—h ypertension, hypervolemia, and hemodilution—with or without
endovascular rescue therapy (endovascular injection of vasodilators and/or
angioplasty). No large randomized trials have been conducted to evaluate
the efficacy of triple-H therapy. Two prospective randomized trials showed
that prophylactic hypervolemic therapy did not improve cerebral blood
flow or outcome of SAH, but it was associated with increased incidence
Chapter 5
of pulmonary edema. Most recent guidelines from the American Heart
Association and the Neurocritical Care Society recommend that treatment
of DCI includes maintenance of euvolemia, prevention hypovolemia, and
hemodynamic augmentation by inducing hypertension. For patients who
do not respond to hemodynamic augmentation or with sudden focal neu
rological deficits attributable to the vascular territory with visible vasospasm,
endovascular injection of vasodilators and/or angioplasty may lead to clin-
ical improvement.
49
Systemic Complications and Critical Care
Management Considerations
Seizures and Seizure Prevention
The diagnosis of seizures in SAH can be complicated because patients may
manifest seizure-like motor movements or posturing at the time of the an-
eurysm rupture that may be secondary to a sudden increase in intracranial
pressure and/or direct brain compression from acute aneurysm rupture.
In a patient with known SAH who has not yet had the bleeding cerebral
aneurysm(s) secured, a clinical seizure event often represents aneurysm re-
rupture and may require emergent management. Furthermore, up to 20% of
comatose SAH patients may experience nonconvulsive seizures that have no
or very subtle clinical manifestation and can only be reliably detected using
continuous EEG monitoring.
Clinical seizures are uncommon in SAH, affecting only 1% to 7% of
patients. Risk factors for developing seizures in SAH include thick subarach-
noid clot, open surgical aneurysm repair in patients >65 years of age, and
presence of intraparenchymal hematoma or cerebral infarction. Whether
anticonvulsants should be used for seizure prophylaxis in SAH remains con-
troversial. Some studies found that anticonvulsant use in SAH patients with
secure aneurysms is associated with worse outcome. It should be noted that
most of the SAH patients in these studies were exposed to phenytoin and
there is limited data on the use of other anticonvulsants and SAH outcome.
In patients with no history of seizure, 72 hours of anticonvulsant prophy-
laxis appears as effective in preventing seizures as a more prolonged course.

While there is no data, expert consensus states that SAH patients who
have had a clinical seizure should be treated with anticonvulsants and that
anticonvulsants should be discontinued if patients remain seizure free for
three to six months. The Neurocritical Care Society SAH treatment guide-
line recommends against the routine use of phenytoin for seizure prophy-
laxis in SAH.
Fever and Systemic Inflammatory Response
SECTION 1
Fever is highly prevalent after SAH and may occur in up to 72% of patients.
Risk factors for fevers include poor-grade SAH, more blood in the subarach-
noid space, and intraventricular blood. While several retrospective studies in
SAH found that fever is independently associated with more cerebral infarc-
tion and poor outcome there is currently no clear data to suggest a beneficial
effect of fever suppression on SAH outcome. Despite this, many centers do
practice fever suppression in SAH given the connection between fever and
poor outcome. Acetaminophen and ibuprofen alone are not very effective
in SAH-related fever suppression. Continuous infusions of nonsteroidal anti-
inflammatory drugs (NSAIDS) may be more effective than oral agents, but
there is a concern that the antiplatelet effects of NSAIDs may worsen intra-
50
cranial bleeding in patients who undergo craniotomy. Surface and intravascular

cooling devices are often used in temperature management in SAH patients,
and they are more effective than oral antipyretics in fever suppression. There
is no data on the effect of antipyretic use or cooling device use and clinical
outcome in SAH.
Systemic inflammatory response syndrome (SIRS), which includes
fever/h ypothermia, leukocytosis/leukopenia, tachycardia, and tachypnea,
is prevalent in SAH. In addition to fever, leukocytosis is common in SAH
and is associated with increased mortality, vasospasm, and poor outcome.
Spontaneous hyperventilation is also common in SAH, found in up to 55%
of patients, and is associated with DCI and poor SAH outcome. One large
prospective randomized trial found up to 63% of SAH patients met SIRS
criteria within four days of presentation. Several studies showed that SIRS
burden is associated with vasospasm and poor SAH outcome. SIRS in SAH is
mostly of noninfectious etiology. There is limited data to determine whether
SIRS is simply a symptom of more severe SAH or if it in fact worsens brain
injury in SAH.
Neurogenic Cardiac and Pulmonary Injuries
Commonly referred to as “neurogenic stress cardiomyopathy” or “neuro-
genic stunned myocardium,” cardiac dysfunction and injury is common fol-
lowing SAH. Approximately 35% SAH patients have troponin I elevation, 35%
suffer arrhythmias, and 25% have cardiac wall motion abnormalities. Cardiac
injury following SAH is generally not due to coronary ischemia but rather
thought to result from catecholamine surge secondary to SAH. Clinically, neu-
rogenic stunned myocardium demonstrates a wide spectrum of severity, from
elevated cardiac biomarkers, dyspnea, hypoxemia, and pulmonary edema to
cardiogenic shock and even sudden death. This syndrome may develop within
hours of SAH onset and typically lasts one to three days, after which cardiac
function generally returns to pre-SAH baseline. Treatment of choice is sup-
portive critical care to maintain adequate blood pressure, cerebral perfusion,
and cerebral oxygen delivery.
Symptomatic pulmonary dysfunction including pulmonary edema, acute
lung injury, and acute respiratory distress syndrome occur in over 20% of SAH
patients. Pulmonary edema can occur in the absence of cardiac dysfunction
Chapter 5
and is referred to as “neurogenic pulmonary edema.” The pathophysiology
of neurogenic pulmonary edema remains controversial, and treatment is sup-
portive critical care to maintain adequate oxygen delivery to the brain and
other end organs while avoiding hypovolemia, which may worsen secondary
brain injury following SAH.
Hyponatremia and Intravascular Volume

Hyponatremia occurs in up to 50% of all SAH patients. The key to managing
hyponatremia in SAH lies in the accurate determination of the patient’s in-
travascular volume and careful management to avoid hypovolemia. Both
51
syndrome of inappropriate secretion of antidiuretic hormones (SIADH) and
cerebral salt wasting may occur in SAH, and sometimes the two syndromes may
coexist. Diagnosis of SIADH can only be made if the patient is intravascularly
euvolemic or mildly hypervolemic. Cerebral salt wasting, on the other hand,
requires hypovolemia in addition to excessive urine output. Hyponatremia
treated with fluid restriction is associated with poor SAH outcome. This likely
reflects the detrimental effect of hypovolemia on SAH outcome. With ade-
quate fluid resuscitation, there is no evidence that hyponatremia alone confers
poor SAH prognosis.
Fludrocortisone and hydrocortisone may mitigate excessive natriuresis and
hyponatremia and reduce the volume of fluids needed to maintain euvolemia
in SAH. Clinicians should anticipate and treat potential side effects such as
hypokalemia related to mineralocorticoid use and hyperglycemia related to
corticosteroid use. Clinical studies did not suggest any increase in incidence
of congestive heart failure associated with hydrocortisone or fludrocortisone
use in SAH with hyponatremia and excessive natriuresis. Use of a 3% hy-
pertonic sodium chloride solution in SAH-associated hyponatremia may be
safe, but there is not enough data to determine its efficacy. Vasopressin-
receptor antagonists should be used with extreme caution in treating SAH-
related hyponatremia. While they are effective in treating euvolemic and
hypervolemic hyponatremia, their use can be associated with significant di-
uresis that may lead to intravascular hypovolemia and worsen SAH outcome,
particularly in the presence of vasospasm and DCI.
Anemia and Transfusion
SAH-associated anemia occurs in over half of all SAH patients, and 80% of
all anemic SAH patients have hemoglobin below 11 g/dl. The pathophysio-
logic mechanism of SAH-associated anemia is poorly understood. The trans-
fusion threshold in SAH patients, particularly those with vasospasm and DCI,
has been an issue of debate because cerebral oxygen delivery is determined
by cerebral blood flow and arterial oxygen content, which is dependent on
hemoglobin concentration. Retrospective studies have demonstrated that an
increase in hemoglobin from 8 to 10 g/dl increases cerebral oxygen delivery
and that higher hemoglobin concentration is associated with good SAH out-
SECTION 1
come. However, transfusion in critically ill patients is associated with delete-

rious effects such as infection, and small studies in SAH showed red blood cell
transfusion may be associated with infection, vasospasm, and poor outcome.
Currently, the appropriate target hemoglobin for SAH patients is undeter-
mined. A pilot randomized trial to test liberal versus conservative transfusion
strategies is ongoing.
Further Reading
Allen GS, Ahn HS, Preziosi TJ, et al. Cerebral arterial spasm— a controlled
52
trial of nimodipine in patients with subarachnoid hemorrhage. N Engl J Med.

1983;308(11):619–624.
Connolly ES, Rabinstein AA, Carhuapoma JR, et al. Guidelines for the management
of aneurysmal subarachnoid hemorrhage: a guideline for healthcare professionals
from the American Heart Association/ American Stroke Association. Stroke.
2012;43(6):1711–1737.
Diringer MN, Bleck TP, Connolly ES, et al. Critical care management of patients
following aneurysmal subarachnoid hemorrhage: recommendations from the
Neurocritical Care Society’s Multidisciplinary Consensus Conference. Neurocrit Care.
2011;15(2):211–240.
Frontera JA, Claassen J, Schmidt JM, et al. Prediction of symptomatic vasospasm after
subarachnoid hemorrhage: the modified Fisher scale. Neurosurgery. 2006;59(1):21–
27; discussion 21–27.
Hasan D, Lindsay KW, Wijdicks EF, et al. Effect of fludrocortisone acetate in patients
with subarachnoid hemorrhage. Stroke. 1989;20(9):1156–1161.
Hunt WE, Hess RM. Surgical risk as related to time of intervention in the repair of intra-
cranial aneurysms. J Neurosurg. 1968;28(1):14–20.
Kassell NF, Torner JC, Haley EC Jr, Jane JA, Adams HP, Kongableet GL. The International
Cooperative Study on the Timing of Aneurysm Surgery. Part 1: overall management
results. J Neurosurg, 1990;73(1):18–36.
Lee VH, Oh JK, Mulvagh SL, Wijdicks EFM. Mechanisms in neurogenic stress cardiomyop-
athy after aneurysmal subarachnoid hemorrhage. Neurocrit Care. 2006;5(3):243–249.
Molyneux AJ, Kerr RS, Yu LM, et al. International Subarachnoid Aneurysm Trial
(ISAT) of neurosurgical clipping versus endovascular coiling in 2143 patients with
ruptured intracranial aneurysms: a randomised comparison of effects on survival,
dependency, seizures, rebleeding, subgroups, and aneurysm occlusion. Lancet,
2005;366(9488):809–817.
Muench E, Horn P, Bauhuf C, et al. Effects of hypervolemia and hypertension on re-
gional cerebral blood flow, intracranial pressure, and brain tissue oxygenation after
subarachnoid hemorrhage. Crit Care Med. 2007;35(8):1844–1851; quiz 1852.
Report of World Federation of Neurological Surgeons Committee on a Universal
Subarachnoid Hemorrhage Grading Scale. J Neurosurg. 1988;68(6): 985–986.
53 Chapter 5
Chapter 6
Reversible Cerebral
Vasoconstriction Syndrome
Krystle Shafer and Bradley J. Molyneaux
Introduction
Reversible cerebral vasoconstriction syndrome (RCVS) is a rare disorder that
often presents with acute thunderclap headache and can result in infarcts as
well as intraparenchymal and subarachnoid hemorrhage. Headache is a well-
known common patient complaint with 90% of these complaints falling under
benign diagnoses. However, this means that 1 out of 10 patients will have a
dangerous diagnosis as the cause of their headache, and oftentimes the symp-
54
tomatology of these dangerous headaches mimics benign conditions. It is im-

portant for providers to appropriately recognize and treat these critically ill
patients. Arguably one of the great masqueraders is RCVS, a rare diagnosis
that has the potential to result in permanent neurologic injury and death.
Definition and Pathophysiology

RCVS refers to the reversible multifocal narrowing of cerebral arteries. The
reversibility suggests an abnormality in control of cerebrovascular tone, but
the pathophysiology of this vasculopathy is not well understood. This is not an
inflammatory process but rather a disorder of the endothelial smooth muscle
tone. Peak incidence for developing this order is reported as between the
ages of 20 and 50 years old. This disorder is more common in women and
has been associated with pregnancy (including the postpartum period). Other
risk factors include exposure to vasoactive medications, pheochromocytoma,
history of migraine headaches, and physical exertion, among many others.
Signs and Symptoms

The overwhelming predominate symptom of RCVS is dramatic and severe
headaches, which are often sudden-onset thunderclap in nature. Associated
symptoms also include nausea and vomiting, photophobia, and visual changes.
Resolution of the headaches and vasoconstriction occurs over days to weeks.
Complications from RCVS in one study occurred in 81% of patients and
included ischemic infarcts, subarachnoid hemorrhage, lobar hemorrhage, and
RCVS
cerebral edema. As these complications develop, additional symptomatology
such as focal neurologic deficits, seizures, coma, and even death (although
Chapter 6
rare) may occur. Seizures, subarachnoid hemorrhage, and intracerebral hem-
orrhage tend to occur within the first week of headache onset while ischemic
events typically present within two weeks.
Diagnosis
Despite having widespread cerebral artery vasoconstriction, up to 70% of
patients will have no abnormalities on their initial brain computed tomography
and/or magnetic resonance imaging. Thus cerebral angiography is arguably
the gold standard for diagnosis. The angiography in these patients typically
displays segmental narrowing and dilation of multiple segmental arteries in a
“string and beads” pattern (Figure 6.1).
Treatment
Management of RCVS can be difficult as there is little data supporting the
55
treatment options, the headaches can be severe despite attempts at treat-
ment, and patients can experience continued clinical decline resulting in
death. Oftentimes the first-line treatment and prevention of the cerebral ar-
terial vasoconstriction induced cerebral ischemia is with oral calcium channel
antagonists such as nimodipine or verapamil. Use of calcium channel blockers
Figure 6.1 Segmental narrowing and dilation of multiple segmental arteries in a “string and
beads” pattern.
is extrapolated from the data supporting nimodipine use in subarachnoid hem-
orrhage patients. Nimodipine is typically administered as 60 mg every four

hours. Due to the need for frequent administration and high cost, nimodipine
as not ideal for transition to hospital discharge. Alternatively, verapamil can be
administered at 80 to 160 mg every 8 hours, dose can be titrated, and it can
more easily be continued at hospital discharge. Calcium channel blockers can
usually be tapered off after resolution of symptoms as an outpatient. These
agents may not shorten the course of vasoconstriction, but they may lessen
the intensity of the headaches.
For patients with persistent headache and vasospasm despite use of cal-
SECTION 1
cium channel blockers, infusion of magnesium in dosing regimens used for

pre-eclampsia and eclampsia patients has highly effective in case reports with
dramatic relief of symptomatology. Typical magnesium dosing would be a
loading dose of 2 to 4 grams over 15 to 20 minutes, which may result in com-
plete relief of headache. Continued treatment with oral magnesium oxide is
an option. Opioid analgesics and acetaminophen may also be used as adjunc-
tive therapy for headache management.
If symptoms continue to progress despite these interventions, directed
intra-arterial therapies with medications such as milrinone or verapamil as
well as cerebral balloon angioplasty have all been reported with successful
results. Intra-arterial intervention is not routinely recommended for all
56
patients due to the risk of reperfusion injury and also because the majority
of patients will have a self-limited disease course with no long-term focal
deficits.
Previously it was thought that treating these patients with permissive hyper-
tension to overcome the flow limited by vasoconstriction would be effective,
but studies have demonstrated that pharmacologically induced hypertension
can induce further cerebral vasoconstriction and/or result in brain hemor-
rhage. However, in patients with severe symptomatology requiring intra-
arterial therapy, occasionally pharmacology-induced hypertension is utilized
as a temporary bridge to improve blood flow to the affected brain regions
while awaiting this directed therapy.
Finally, it is important to recognize that glucocorticoids were previously
given to all RCVS patients more so to treat a suspected angiitis component
of the disease. However, the literature has demonstrated steroid use in RCVS
patients to be an independent predictor of clinical, imaging, and angiographic
worsening as well as poor outcome. As such, steroids should never be given
prophylactically in this particular patient population.
Risk of Recurrence
Five percent of RCVS patients will develop recurrence. Sexual activity as
the precluding trigger to the thunderclap headache is an independent pre-
dictor for recurrent RCVS. Patients who develop this condition during or
immediately following pregnancy only rarely develop this syndrome with
RCVS
recurrent pregnancies, and thus it is not considered a contraindication for
future pregnancies.
Chapter 6
Further Reading
Bouchard M, Verreault S, Gariépy J, Dupré N. Intra-arterial milrinone for reversible ce-
rebral vasoconstriction syndrome. Headache. 2009;49(1):142–145.
Calabrese L, Dodick D, Schwedt T, Singhal A. Narrative review: reversible cerebral vas
oconstriction syndromes. Ann Intern Med. 2007;146(1):34–44.
Chen S, Fuh J, Lirng J, Wang Y, Wang S. Recurrence of reversible cerebral vasocon-
striction syndrome: a long-term follow-up study. Neurology. 2015;84(15):1552–1558.
Ducros A, Boukobza M, Porcher R, Sarov M, Valade D, Bousser M. The clinical and ra-
diological spectrum of reversible cerebral vasoconstriction syndrome. a prospective
series of 67 patients. Brain. 2007;130(Pt. 12):3091–3101.
Katz B, Fugate J, Ameriso S, et al. Clinical worsening in reversible cerebral vasoconstric-
tion syndrome. JAMA Neurol. 2014;71(1):68–73.
Mijalski C, Dakay K, Miller-Patterson C, Saad A, Silver B, Khan M. Magnesium for treat-
ment of reversible cerebral vasoconstriction syndrome: case series. Neurohospitalist.
2016;6(3):111–113.
Rosenbloom M, Singhal A. CT angiography and diffusion-perfusion MR imaging in a
57
patient with ipsilateral reversible cerebral vasoconstriction after carotid endarterec-
tomy. Am J Neuroradiol. 2007;28(5):920–922.
Sattar A, Manousakis G, Jensen M. Systematic review of reversible cerebral vasocon-
striction syndrome. Expert Rev Cardiovasc Ther. 2011; 8(10);1417–1421.
Singhal A, Hajj- Ali R, Topcuoglu M, et al. Reversible cerebral vasoconstriction
syndromes: analysis of 139 cases. Arch Neurol. 2011; 68(8):1005–1012.
Singhal A, Kimberly W, Schaefer P, Hedley-Whtye E. Case 8-2009: a 36-year old
woman with headache, hypertension, and seizure 2 weeks postpartum. N Engl J Med.
2009;360:1126–1137.
Singhal A, Topcuoglu M. Glucocorticoid-associated worsening in reversible cerebral vas
oconstriction syndrome. Neurology. 2017;88(3):228–236.
Chapter 7
Anoxic Brain Injury

Jonathan Elmer and Jon C. Rittenberger
Incidence and Epidemiology

Sudden cardiac arrest is the most common cause of death in North America.
Each year, more than 120,000 patients in the United States are treated in the
hospital after resuscitation from cardiac arrest. Recent advances in post-arrest
care including targeted temperature management, early coronary revasculari-
zation, and delayed neurological prognostication have improved the likelihood
of favorable neurological outcome. Yet, a majority of patients hospitalized
after cardiac arrest still die before discharge. Anoxic-ischemic brain injury
is the most common proximate cause of death in nonsurvivors (Table 7.1).
Fortunately, for those discharged with a favorable neurological outcome, long-
58
term survival is excellent.
Initial Patient Evaluation

Etiology of Cardiac Arrest
Cardiac arrest is the final common pathway for numerous disease processes.
Accurately diagnosing or excluding specific etiologies of an individual patient’s
arrest is critical for acute management (e.g., emergent cardiac catheterization
Table 7.1 Patient Outcomes After Cardiac Arrest

Patient Outcome Percentage of Patients
Survive to discharge 30–50
Favorable neurological outcome 10–20
Unfavorable neurological outcome 20–30
In-hospital death 50–70
Withdrawal of life-sustaining therapy based on 40–60
perceived neurological prognosis
Brain death 5–10
Pre-existing advanced directives or surrogates’ 5–15
representation of patient’s wishes
Rearrest or medically unstable 10–20
in acute coronary syndromes, early antibiotics, and hemodynamic resuscita-
Anoxic Brain Injury

tion for septic shock), secondary prevention (e.g., internal defibrillator place-
ment) and risk stratification.
Historically, consensus guidelines dichotomized arrest etiology as “pre-
sumed cardiac” and “noncardiac.” While potentially useful for clinical re-
search, this definition has little clinical utility, and providers must generate a
more comprehensive differential diagnosis. Fortunately, thoughtful consider-
ation and targeted testing can identify or rule out the underlying etiology of
Chapter 7
cardiac arrest in most patients (Table 7.2).
Severity of Illness
Precise neurological prognostication based on initial evaluation is impossible
(see later discussion). However, there are several validated tools for early risk
stratification after cardiac arrest that can inform discussions with families and
surrogates and help them understand a patient’s anticipated clinical course.
The simplest validated tool is the Pittsburgh Cardiac Arrest Category scale,
which stratifies patients into four categories based on initial neurological ex-
amination and severity of cardiopulmonary dysfunction (Figure 7.1).
In addition to a focused physical examination, brain imaging with computed
tomography (CT) should be part of the initial evaluation of the comatose
post-arrest patient. Baseline CT imaging may alter early post-arrest manage-
59
ment and further inform risk stratification. Approximately 1 in 20 patients will
have intracranial hemorrhage or other acute primary central nervous system
etiologies of arrest identified on initial post-arrest CT scan. More commonly,
in roughly 20% of cases, the initial brain CT will demonstrate at least some de-
gree of early cerebral edema, characterized by loss of gray-white differentia-
tion, as well as mass effect and effacement of the sulci and basal cisterns when
severe. Studies demonstrate a significant increase in mortality when the ratio
of Hounsfield units in gray matter compared to white matter falls below 1.2.
Emergent Cardiac Catheterization
In patients who may have suffered cardiac arrest from acute coronary syn-
drome, emergent cardiac catheterization is strongly associated with favorable
neurological outcomes for both patients with ST-elevation myocardial infarc-
tion and those without ST elevation. However, in the face of severe brain
injury with loss of brainstem reflexes (Pittsburgh Cardiac Arrest Category
4), risk of death from neurological deterioration far outweighs the risk of
death from multiple system organ failure. In these patients, it may be reason-
able to delay coronary angiography in favor of stabilization and aggressive
neurocritical care in the intensive care unit.
Critical Care Management

Anoxic-ischemic brain injury is the major driver of both morbidity and
mortality for these patients, particularly after out-of-hospital cardiac arrest
Table 7.2 Common Causes of Cardiac Arrest and Suggested

Initial Diagnostic Evaluation
Etiology Diagnostic Evaluation
Cardiovascular ECG—ischemia, rhythm, intervals
Acute coronary syndrome Troponin
Congenital arrhythmia (long QT, Brugada, etc) Echocardiogram
Congenital structural heart disease
Arrhythmia from (non-)ischemic cardiomyopathy
Cardiogenic shock
SECTION 1
Pulmonary hypertension/right ventricular failure

Pulmonary Chest x-ray
Large airway obstruction Arterial blood gas
Respiratory failure (e.g., asthma, pneumonia, Physical exam
chronic obstructive pulmonary disease) Bronchoscopy
Catastrophic neurological event CT head
Exsanguination (traumatic or nontraumatic) History
Physical exam
Hemoglobin
CT scan chest/abdomen/pelvis
Obstructive shock Echocardiogram
60
Pulmonary embolism Chest x-ray

Pneumothorax CT chest
Tamponade
Distributive shock Chest x-ray
Sepsis Urinalysis
Anaphylaxis White blood cell count
Cultures
Metabolic Arterial blood gas
Diabetic ketoacidosis Serum chemistries
Hypokalemia or hyperkalemia
Hypomagnesemia
Overdose History
Sedative, hypnotic, opioid ECG
Sympathomimetic Drug screen
Other
Environmental (electrocution, hypothermia, etc.) History
ECG = electrocardiogram; CT = computed tomography.
(OHCA). Post-arrest brain injury can be conceptualized as a primary ischemia-

reperfusion injury occurring during pulselessness and the minutes after return
of spontaneous circulation and secondary brain injury that may develop in the
days after resuscitation when the acutely damaged brain is vulnerable to new
Anoxic Brain Injury
Pittsburgh Cardiac Arrest Categories
Examine wakefulness and brainstem reflexes, shock and pulmonary status
Category 1 Awake
Follows commands or makes purposeful movements (e.g. pulling at tubes and lines)
Category 2 Light coma without severe cardiopulmonary failure
Does not follow commands or make purposeful movement, but brainstem reflexes are present.
Modest vasopressor requirements (norepinephrine <0.1 mcg/kg/min) and reasonable to oxygenate
(e.g. SaO2 90–100% with standard pressure control ventilation)
Chapter 7
Category 3 Light coma with severe cardiopulmonary failure
Does not follow commands or make purposeful movement but brainstem reflexes are present. High
vasopressor requirements and/or hard to oxygenate (SaO2 <90% or requiring special ventilation modes)
Category 4 Deep coma with loss of brainstem reflexes

Does not follow commands or make purposeful movement
Some or all brainstem reflexes are lost (e.g. no pupil response or gag or cough).
Outcome by Pittsburgh Cardiac Arrest Category

Category 1: 80% survival, 60% favorable outcomes
100 Survival with favorable outcome
Survival with poor outcome
Percent of patients

50
Risk of Multiple Organ Failure
61
0 3% 40%
20% 35%
1 2 3 4
Pittsburgh Cardiac Arrest Category Category: 1 2 3 4
Figure 7.1 Defining characteristics and outcomes across the spectrum of Pittsburgh
Cardiac Arrest Category.
injury (Figure 7.2). The major focus of post-arrest critical care is prevention of
secondary brain injury.
Targeted Temperature Management
Preclinical data demonstrate pleiotropic neuroprotective effects of mild hypo-
thermia after anoxic brain injury. Hypothermia reduces free radical formation,
oxidative injury, cerebral edema, intracranial pressure, cerebral metabolism,
and excitotoxicity; attenuates apoptotic signaling pathways and protease acti-
vation; and raises the seizure threshold. After two randomized controlled trials
published in 2002 demonstrated improved outcomes among hypothermia-
treated OHCA patients, clinical use of mild therapeutic hypothermia after
cardiac arrest became common.
More recently, the large Targeted Temperature Management (TTM) trial
demonstrated equivalent long-term neurological outcomes among patients
randomized to 33oC compared to 36oC after OHCA. Several differences from
earlier studies may explain these seemingly discrepant findings. First, both
Primary brain injury: Ischemia/reperfusion Secondary brain injury

Energetic failure Oxidative injury Hypoxia + hypoperfusion
Anoxic depolarization Excitotoxicity Autoregulatory failure
Loss of ion gradients Reactive hyperemia Cerebral edema
Calcium/glutamate No-reflow Seizures
BBB disrupted Cerebral edema Oxidative stress
Fever
Alive Collapse Pulseless CPR ROSC Transport Post-arrest

SECTION 1
High-quality CPR
Rate and depth Neurocritical care
Rapid defibrillation
Correct reversible causes
Figure 7.2 Anoxic-ischemic brain injury associated with cardiac arrest and evidence-based
interventions to mitigate its severity.
BBB = blood brain barrier; CPR = cardiopulmonary resuscitation; ROSC = return of spontaneous
circulation. Adapted from Elmer J, Callaway CW. The brain after cardiac arrest. Semin Neurol. 2017
62
Feb;37(1):19–24.
arms of the TTM trial were treated with mild therapeutic hypothermia; only
the depth of cooling differed. By contrast, control-arm temperatures in earlier
trials ranged from 37oC to 38.5oC (normothermia or fever). It is unknown
whether the beneficial effect of targeting 36oC lies in fever prevention or use
of mild hypothermia, but it is clear that the majority of post-arrest patients
not treated with active temperature management develop fevers, and fever is
strongly associated with worse outcomes in brain-injured populations.
Second, the TTM trial used a detailed protocol for blinded, standard-
ized neurological prognostication, ensuring that exposure to withdrawal of
life-sustaining therapy based on perceived neurological prognosis (the most
common proximate cause of death after OHCA) did not differ across treat-
ment arms, thus eliminating a major potential source of bias present in ear-
lier studies. In multiple prespecified and post hoc subgroup analyses, no
populations were identified that benefitted from treatment at 33oC compared
to 36oC. At present, therefore, data are supportive of actively targeting either
33oC or 36oC in comatose OHCA patients, and many clinicians translate these
findings to treatment of in-hospital cardiac arrest survivors as well.
From a practical perspective, any active temperature management method
is reasonable. Surface cooling pads and endovascular catheters have equivalent
safety and efficacy profiles. Continuous, core temperature measurement and
active feedback to the temperature management device are mandatory, and
reliance on rectal temperature should be avoided since the measured rectal
Anoxic Brain Injury
Table 7.3 Extracerebral Physiological Effects of Mild Hypothermia
Organ System Effects
Cardiovascular • Decreased heart rate
• Increased stroke volume
• Preserved cardiac output
• Cutaneous vasoconstriction
Respiratory • PaO2 measured at 37oC results 5 mm Hg/degree higher
Chapter 7
than in vivo PaO2
• PaCO2 measured at 37oC results 2mmHg/degree higher
than in vivo PaCO2
• Decreased ciliary motility
Gastrointestinal • Decreased hepatic clearance of medications
(~10% decrease/oC)
• Delayed gastric emptying
Renal • Intracellular shift of potassium
• Cold diuresis
Hematologic/Immune • Mild immunosuppression
Coagulation • Mildly decreased platelet function
• Mildly decreased clotting factor activity
Integumentary • With surface cooling, decreased skin blood flow increasing
63
risk of skin breakdown
Endocrine • Insulin resistance, mild hyperglycemia
temperature may lag behind true core temperature by several hours. Goal
temperature should be achieved as soon as feasible after cardiac arrest and is
reasonable to maintain for 24 hours as was tested in the TTM trial. Following
this period, rewarming should be carried out slowly to avoid rebound cere-
bral edema and neurological deterioration. It is our local practice to rewarm
at 0.25oC/hour, so rewarming from 33oC is accomplished over 16 hours.
Mild hypothermia has physiological effects on every organ system (Table 7.3).
Development of institutional protocols for safe and standardized delivery of
TTM and management of the hypothermic patient is recommended.
Ventilator Management
Observational studies associate severe hyperoxemia (PaO2 >300 mm Hg)
with mortality after cardiac arrest, presumably related to increased reactive
oxygen species formation and free radical injury. Hypoxemia is also associ-
ated with mortality. Therefore, it is reasonable to target normoxia or mild
hyperoxia during the early post-arrest period, after temperature correction
of the arterial blood gas (Table 7.3). Data regarding optimal PaCO2 goals
are mixed. Hypocarbia may cause cerebral vasoconstriction and has consist
ently been associated with worse patient outcomes. Hypercarbia is a ce-
rebral vasodilator and may improve cerebral blood flow at the expense of
increased intracranial pressure. Some observational studies have associated
mild hypercarbia with improved outcomes, while others have found increased
mortality in hypercarbic patients. At present, it is reasonable to normalize
PaCO2 (after temperature correction) or permit mild hypercarbia.
Hemodynamic Management
After anoxic brain injury, cerebral pressure autoregulation is frequently right-
shifted, impaired, or absent, and higher mean arterial pressures may be re-
quired to maintain adequate cerebral blood flow. Observational data suggest
improved survival when higher mean arterial pressures are maintained to
SECTION 1
promote cerebral perfusion, regardless of whether patients are vasopressor

dependent to achieve this goal. It is our local practice to maintain a mean ar-
terial pressure of >80 mm Hg during active temperature management, then
liberalize as conditions allow.
Post-arrest Seizures
Electroencephalographic (EEG) findings after resuscitation from cardiac arrest
are prognostic. Seizures and other epileptiform EEG findings develop in ap-
proximately 25% of patients resuscitated from cardiac arrest and are asso-
ciated with in-hospital mortality. Early malignant EEG patterns in the first 48
hours after initial resuscitation are ominous. In this population, most authors
64
consider periodic epileptiform discharges, seizures, and polyspike bursts to

be malignant, although definitions vary in the literature. Observation of myo-
clonic jerks in lock-step with polyspike bursts, variably termed status myo
clonus or myoclonic status epilepticus, portends poor prognosis. By contrast,
favorable neurological outcomes are not uncommon among patients who de-
velop malignant EEG patterns later in their post-arrest course. Other prog-
nostic EEG features include the presence of reactivity and development of a
continuous background by 48 hours post-arrest, both of which are strongly
associated with favorable outcome.
Whether early malignant EEG patterns after cardiac arrest are injurious in
and of themselves or simply a sign of severe brain injury is controversial. It
is our practice to aggressively treat these patients with antiepileptic drugs. In
other critically ill populations, early seizure control has been associated with
improved outcomes (see chapter 8). However, data supporting this practice
after cardiac arrest are lacking.
Neurological Prognostication
Accurate neurological prognostication after cardiac arrest is challenging, and
withdrawal of life-sustaining therapy based on anticipated neurological prog-
nosis is the most common proximate cause of death after cardiac arrest.
Evidence- based guidelines advocate delaying neurological prognostication
until at least 72 hours after cardiac arrest, and this timeline should be extended
to 72 hours after rewarming to normothermia for patients treated with
hypothermia. Before this time, no clinical sign, test result, or combination of
Anoxic Brain Injury

findings precludes a favorable neurological outcome. Importantly, awakening
from coma may occur days to weeks after cardiac arrest, and patients who
remain comatose at 72 hours may go on to have favorable recoveries. In the
absence of awakening or progression to brain death, prognostication depends
on integration of results from multiple diagnostic modalities (Table 7.4).
It is important to note that many prognostic findings once thought to uni-
formly predict poor outcome are imprecise and observational studies of
Chapter 7
prognostic modalities have a high risk of bias. To date, no blinded studies of
neurological prognostication after cardiac arrest have been conducted. Thus,
clinicians caring for patients included in observational studies are at liberty to
withdraw life-sustaining therapy based on the results of prognostic tests. Since
withdrawal of life-sustaining therapy after cardiac arrest invariably results
in death, self-fulfilling prophecies are created that have biased the literature
toward inappropriately pessimistic point estimates and narrow confidence
intervals when reporting false positive rates for predicting poor outcome.
Table 7.4 Prognostic Modalities after Cardiac Arrest

Modality Timing Notes
Clinical examination Daily • Initial exam findings can be summarized by
65
the Pittsburgh Cardiac Arrest Category
classification system (Figure 7.1)
• Persistent coma on Day 3 does not preclude a
favorable outcome
• Persistent absence of pupillary and corneal
reflexes on Day 3 is ominous
Brain CT Baseline • Gray matter to white matter ratio of Hounsfield
units <1.2 on initial CT scan is ominous
EEG Days 1–3 • Early development of malignant EEG findings is
ominous
• Presence of reactivity is associated with
favorable outcomes
• Early burst suppression has little prognostic
significance, but failure to develop a continuous
EEG background by Day 2 is ominous
Somatosensory Day 3 • Bilaterally absent N20 cortical responses to
evoked potentials median nerve stimulation is ominous
NSE Day 3 • Higher NSE is associated with worse
outcomes, but cutoffs vary by laboratory assay
• Use of NSE requires knowledge of the specific
lab’s operating characteristics
Brain MRI Days 3–5 • Greater burden of injury on apparent diffusion
coefficient and diffusion weighted imaging are
associated with worse prognosis
CT = computed tomography; EEG= electroencephalography; ECG = electrocardiogram; NSE = neuron
specific enolase; MRI = magnetic resonance imaging.
Provided that ongoing intensive care is compatible with an individual patient’s
values and preferences, there is little downside to delaying prognostication to

allow an opportunity for signs of recovery to manifest.
Conclusion
Outcomes after hospital treatment for cardiac arrest are steadily improving
over time. Provision of a comprehensive bundle of care including active tem-
perature management, coronary revascularization, delayed neurological prog-
SECTION 1
nostication, and best practice neurocritical care has resulted in a significant

increase in favorable outcomes from patterns of injury once thought to be
incompatible with survival.
Further Reading
Beylin ME, Perman SM, Abella BS, et al. Higher mean arterial pressure with or without vas
oactive agents is associated with increased survival and better neurological outcomes
in comatose survivors of cardiac arrest. Intensive Care Med. 2013;39:1981–1988.
Callaway CW, Donnino MW, Fink EL, et al. Part 8: post-cardiac arrest care: 2015
American Heart Association guidelines update for cardiopulmonary resuscitation and
66
emergency cardiovascular care. Circulation. 2015;132:S465–S482.

Callaway CW, Schmicker RH, Brown SP, et al. Early coronary angiography and in-
duced hypothermia are associated with survival and functional recovery after out-of-
hospital cardiac arrest. Resuscitation. 2014;85:657–663.
Coppler PJ, Elmer J, Calderon L, et al. Validation of the Pittsburgh Cardiac Arrest
Category illness severity score. Resuscitation. 2015;89C:86–92.
Elmer J, Scutella M, Pullalarevu1 R, et al. The association between hyperoxia and patient
outcomes after cardiac arrest: analysis of a high-resolution database. Intensive Care
Med. 2015;41:49–57.
Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to im-
prove the neurologic outcome after cardiac arrest. N Engl J Med. 2002;346:549–556.
Kilgannon JH, Jones AE, Shapiro NI, et al. Association between arterial hyperoxia
following resuscitation from cardiac arrest and in- hospital mortality. JAMA.
2010;303:2165–2171.
Kilgannon JH, Roberts BW, Jones AE, et al. Arterial blood pressure and neurologic out-
come after resuscitation from cardiac arrest. Crit Care Med. 2014;42:2083–2091.
Laver S, Farrow C, Turner D, Nolan J. Mode of death after admission to an intensive
care unit following cardiac arrest. Intensive Care Med. 2004;30:2126–2128.
Metter RB, Rittenberger JC, Guyette FX, Callaway CW. Association between a quantita-
tive CT scan measure of brain edema and outcome after cardiac arrest. Resuscitation.
2011;82:1180–1185.
Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33
degrees C versus 36 degrees C after cardiac arrest. N Engl J Med. 2013;369:2197–2206.
Oh SH, Oh JS, Kim YM, et al. An observational study of surface versus endovascular
Anoxic Brain Injury

cooling techniques in cardiac arrest patients: a propensity-matched analysis. Crit Care.
2015;19:85.
Phelps R, Dumas F, Maynard C, Silver J, Rea T. Cerebral performance category and
long-
term prognosis following out- of-
hospital cardiac arrest. Crit Care Med.
2013;41:1252–1257.
Polderman KH. Mechanisms of action, physiological effects, and complications of hypo-
thermia. Crit Care Med. 2009;37:S186–202.
Rittenberger JC, Popescu A, Brenner RP, Guyette FX, Callaway CW. Frequency and
Chapter 7
timing of nonconvulsive status epilepticus in comatose post-cardiac arrest subjects
treated with hypothermia. Neurocrit Care. 2012;16:114–122.
Roberts BW, Kilgannon JH, Chansky ME, Mittal N, Wooden J, Trzeciak S. Association
between postresuscitation partial pressure of arterial carbon dioxide and neu-
rological outcome in patients with post- cardiac arrest syndrome. Circulation.
2013;127:2107–2113.
Sandroni C, Cavallaro F, Callaway CW, et al. Predictors of poor neurological out-
come in adult comatose survivors of cardiac arrest: a systematic review and
meta-analysis. Part 2: patients treated with therapeutic hypothermia. Resuscitation.
2013;84:1324–1338.
67
Chapter 8
Seizures and Status

Epilepticus
Michael E. Reznik and Jan Claassen
Introduction
Seizures are a common reason for presenting to the hospital, representing 1%
of all visits to the emergency department. However, they may also arise as a
result of other neurologic and medical conditions, especially in the critically ill.
Medical providers should therefore be familiar with their diagnosis and sub-
sequent management, also recognizing that prolonged or repetitive seizure
activity, called status epilepticus (SE), represents a medical emergency.
68
Diagnosis and Classification

Seizures may manifest in a multitude of ways depending on the area of the
brain affected. The most recognizable type is the generalized convulsive sei-
zure, which involves a tonic phase followed by bilateral clonic movements. In
adults, these typically have a focal onset, which may or may not be clinically
recognizable; meanwhile, other seizures may remain focal without secondary
generalization. The semiology of focal seizures can range from motor activity
such as facial twitching, automatisms, or unilateral convulsions, to eye devi-
ation, aphasia, echolalia, or aura. They may also be accompanied by altered
consciousness ranging from confusion to coma, and occasionally this may be
the only clinical manifestation of ongoing seizures, in which case they are gen-
erally termed nonconvulsive seizures.
Most seizures are self-limiting and typically terminate in one to two minutes
or less. However, if a seizure has not terminated after five minutes, it is un-
likely to do so spontaneously. Seizures that involve generalization to both
hemispheres are usually followed by a postictal period that may last from
minutes to hours, wherein patients may be lethargic, hypopneic, rhonchorous,
and confused. In such situations, patients are often amnestic to the events im-
mediately preceding their seizure, up until the time that their postictal state
resolves.
It is important to note that a number of conditions may also mimic seizures,
Seizures and Status Epilepticus

ranging from transient ischemic attacks, migraine auras, movement disorders,
convulsive syncope, rapid eye movement (REM) sleep behavioral disorders,
and nonepileptic behavioral spells of psychogenic etiology. For this reason,
an electroencephalogram (EEG) is a helpful diagnostic tool, particularly if re-
corded during a clinical event. However, even an interictal EEG can be useful
if it shows epileptiform activity such as spike-wave discharges or sharp waves,
as these are indicative of a seizure tendency. Seizures arising from a very small
or deep focus in the brain may also not have a clear electrographic corre-
late given the relatively poor spatial resolution of surface EEG, as well as its
dependence on orientation of electrical dipoles. In such scenarios, diagnosis
rests on purely clinical grounds.
Chapter 8
Causes and Initial Workup
From a pathophysiologic standpoint, seizures result from abnormal excessive
or synchronous neuronal activity in the brain. This may stem from dysfunc-
tion of various ion channels leading to upregulation of excitatory circuits,
downregulation of inhibitory circuits, or both. Most often, such dysfunction
is due to underlying idiopathic epilepsy, an acquired injury to the brain, or a
69
systemic inflammatory state. Common causes of seizures, and appropriate
initial diagnostic workup, can be found in Box 8.1. Note that it is important to
keep the differential diagnosis for etiology in mind as it can significantly alter
early management (e.g., hypoglycemia) (Box 8.1).
Status Epilepticus
Status epilepticus (SE) is defined as continuous or recurrent seizure activity
lasting for longer than five minutes without a return to baseline between
episodes. It should be considered a medical emergency warranting immediate
treatment, as the excitotoxicity induced by SE is thought to trigger a cascade
leading to neuronal cell death starting as early as 30 to 60 minutes after onset.
As with single seizures, SE may be classified as focal SE, generalized
convulsive SE (GCSE), or non-convulsive SE (NCSE). There is significant
overlap, however, as patients may transition from one type to another. Non-
convulsive seizures can persist in 20% to 48% of patients with GCSE after
cessation of convulsive seizures, and 14% may remain in NCSE. Another type
of SE, myoclonic status epilepticus, can occur after anoxic brain injury and has
historically been considered a harbinger of poor outcomes (although recent
reports have suggested that the outcomes may be less bleak than previously
thought). This involves repetitive brief myoclonic jerks or eyelid opening with
an associated polyspike discharge on EEG. In contrast, myoclonus without
Box 8.1 Possible Causes of Seizures and Diagnostic Workup
Possible causes of seizures

For patients previously known to have epilepsy:
• Medication noncompliance
• Medications that lower seizure threshold
• Synthetic opiates (e.g., tramadol, fentanyl)
• Antibiotics (e.g., beta-lactams, metronidazole, isoniazid)
• Antidepressants (e.g., bupropion, tricyclic antidepressants)
• Other psychotropics (e.g., lithium, clozapine)
• Psychostimulants (e.g., amphetamines)
• Anticholinergics
• Infection
• Sleep deprivation
• Other physical or emotional stressors
For patients without previous epilepsy:
• Traumatic brain injury
• Ischemic or hemorrhagic stroke
• Brain neoplasm
• Anoxia
• Central nervous system infections (e.g., encephalitis or meningitis)
• Hypertensive encephalopathy (including Posterior Reversible Encephalopathy Syndrome [PRES]
and eclampsia)
• Central nervous system inflammatory or autoimmune conditions (e.g., limbic encephalitis)
• Metabolic derangements (e.g., hypo-or hyperglycemia, hypo-or hypernatremia, hypocalcemia,
hypomagnesemia, uremia, hyperammonemia)
• Critical illness and systemic inflammatory states
• Toxic etiologies
• Medication overdoses (e.g., aspirin, antidepressants, phenytoin)
• Illicit drug use (e.g., cocaine, ecstasy, amphetamines)
• Toxic exposures (e.g., carbon monoxide, heavy metals)
• Poisonings (e.g., organophosphates, strychnine, hydrocarbons)
• Withdrawal states (e.g., chronic alcohol, benzodiazepine, barbiturate, or baclofen use)
Diagnostic workup
• History and physical examination
• Laboratory testing
• Complete blood count
• Electrolytes
• Toxicology screen
• Pregnancy test
• AED levels
In patients without a previous history of seizures:
• Imaging
• Head computed tomography (CT) scan
• Brain magnetic resonance imaging (MRI) +/–contrast
• Angiogram +/–venogram (MR, CT, or conventional)
• EEG
• Lumbar puncture (should not delay the administration of empiric antimicrobial coverage if men-
ingitis is suspected)
EEG correlate is also common after anoxic brain injury and does not repre-

sent seizure activity.
SE has been shown to become more refractory in animal studies within
minutes of seizure onset. This is hypothesized to be caused by receptor
trafficking of GABA receptors from the synaptic membrane into endosomes,
while glutamatergic AMPA and NMDA receptors are mobilized from the cell
interior to the synaptic membrane. As a result, there is a loss of inhibition (and
decreased efficacy of GABA agonists) coupled with an increase in excitation.
Other hypothesized mechanisms include a reduced sensitivity to antiepileptic
drugs (AEDs) in GABA receptors in the hippocampal dentate granule cell, as
well as overexpression of AED efflux transporters.
Clinically, refractory status epilepticus (RSE) is defined as ongoing seizures
that fail to respond to first-and second-line drug therapy and can occur in up
Chapter 8
to 30% to 43% of patients with SE. A further 10% to 15% of patients with
RSE fail to respond to third-line therapy and are considered to have super-
refractory status epilepticus (SRSE). Both of these are associated with pro-
gressively increasing morbidity and mortality.
The immediate complications of SE are many. At the onset of seizures,
convulsions pose a risk of inflicting traumatic injury, while airway com-
promise may lead to aspiration and respiratory failure. An initial cate-
cholamine surge may also cause further cardiopulmonary complications
71
including pulmonary edema and hypoxia, cardiac arrhythmias, and car-
diomyopathy. Meanwhile, other complications may stem from a pro-
longed state of relative immunocompromise (in part induced by prolonged
anesthetics, especially pentobarbital), immobility, and overall critical illness.
These lead to an increased risk for infections, fever, thrombosis, and met-
abolic derangements (especially metabolic acidosis, which is common after
seizures, and hyperkalemia, which can occur if prolonged convulsions result
in rhabdomyolysis).
The risks of prolonged anesthetics are also becoming more recognized.
Recent studies reviewed by Sutter and Kaplan in 2015 have raised some
concern that medical therapy with anesthetic infusions may contribute to
the poor cognitive outcomes after RSE and SRSE. However, this remains a
topic of considerable debate given the likely selection bias inherent in these
studies.
Management of Status Epilepticus

The body of work investigating SE over the past several decades has es-
tablished the importance of starting treatment as early as possible, with
the goal of rapid seizure cessation. In agreement with the findings in animal
models indicating early changes leading to RSE mentioned earlier, clinical
studies have shown that SE is more likely to be successfully terminated
when treatment is started early. Such studies have also validated an aggres-
sive approach to seizure termination in the prehospital setting.

As delineated in the study by Treiman et al. in 1998, benzodiazepines
should always be the emergent first-line treatment of choice given their su-
perior efficacy. It is also crucial that they be dosed appropriately since when
they fail to control SE, it is most often because of underdosing or there
being too long of an interval before the first dose or in between doses.
A bolus infusion of a maintenance AED is generally recommended as urgent
second-line therapy to prevent further seizures from occurring, although ev-
idence for choosing between phenytoin, valproate, levetiracetam, and phe-
SECTION 1
nobarbital is lacking. The choice between these agents often comes down
to immediate availability and patient-specific contraindications. However, if
seizures have not terminated with benzodiazepines, they are also unlikely to
do so with typical second-line agents. In such situations one should consider
expediting the initiation of a continuous anesthetic infusion and continuous
EEG monitoring.
Meanwhile, as with any other medical emergency, concurrent evaluation
and management of the patient’s airway, respiratory status, and hemody-
namics should also be undertaken, along with obtaining adequate intravenous
(IV) access and evaluating for any immediately reversible causes, especially
hypoglycemia. Critically, when correcting hypoglycemia, IV thiamine should
72
be given before or concurrently with glucose to avoid precipitating an

acute Wernicke encephalopathy. Ongoing investigation into the etiology of
the seizures while it remains unknown should continue, as in some cases
AEDs alone will not be sufficient to control SE until the underlying cause is
treated (e.g., metabolic derangements, autoimmune limbic encephalitis; see
Figure 8.1).
For RSE, there is no consensus as to intensity and duration of treatment.
Certainly, a minimum goal of treatment intensity is seizure suppression on
EEG, but this may not be possible for some patients without a higher depth
of anesthesia—therefore many experts advocate aiming for a goal of burst
suppression on EEG (typically with a goal of 3 to 10 seconds of suppression
between bursts) for anywhere from 24 to 72 hours while other treatment is
optimized. Note that in some cases seizures may even persist while in burst
suppression, with some experts advocating complete background suppression
in such situations.
Once it is thought that treatment has been sufficiently optimized (or a re-
versible etiology has resolved), it is appropriate to begin weaning anesthetic
agents. However, withdrawal seizures are common, necessitating careful
weaning and vigilant EEG monitoring. In cases of failure to wean, it is typical to
add another maintenance AED prior to attempting to do so again. The various
therapeutic options for SE can be found in Table 8.1.
First-Line Management
• Recognize prolonged seizure • Lorazepam IV 0.1 mg/kg (max 4 mg per dose) and can
>5 min repeat at 5–10 min, or
• Airway, Breathing, Circulation • Diazepam IV 0.15 mg/kg (max 10 mg per dose) and can
• Ensure adequate IV access repeat in 5 min, or
• Assess for hypoglycemia • Diazepam PR 0.2 mg/kg (>12 yr), 0.3 mg/kg for (6–11 yr),
0.5 mg/kg (2–5 yr), or
• Midazolam IM 0.2 mg/kg (max 10 mg) or intranasal
0.2 mg/kg
Chapter 8
Second-Line Management
• Phenytoin 20 mg/kg at 50 mg/min, or • Monitor hemodynamics and respiratory status
• Fosphenytoin 20 mg/kg at 150 mg/min, or during treatment
• Valproic acid 20–40 mg/kg IV over 10 min • Order cEEG if patient does not return to
baseline
• Order level post-loading dose, then start
maintenance dose (and/or re-bolus • Prepare for transfer to ICU, consider
73
if needed) intubation
• Consider adjunctive levetiracetam 1–3 g IV
over 2–5 mg/kg/min
Refractory Status Epilepticus

• Intubate and begin IV anesthetics, titrate to burst suppression on EGG. Options include:
• Midazolam 0.2 mg/kg boluses (up to max 2 mg/kg), then maintenance infusion
0.05–2.9 mg/kg/hr
• Propofol 1–2 mg/kg boluses over 3–5 min, then maintenance infusion 30–100 µg/kg/min
• Ketamine 1.5 mg/kg boluses (up to max 4.5 mg/kg), then maintenance infusion
0.9–7.5 mg/kg/hr
• Pentobarbital 5–15 mg/kg boluses (up to 50 mg/min), then maintenance infusion
1–10 mg/kg/hr
• Maintain burst and seizure suppression on cEEG for 24–48 hours
• Optimize maintenance AEDs, consider adding adjunct agents as indicated
Figure 8.1 Overview of status epilepticus management. Note that in some cases one may
opt to begin treatment for refractory SE as soon as benzodiazepines are unsuccessful in
terminating seizures.
Table 8.1 Overview of Therapeutic Options for SE
Drug Dosing Mechanism Metabolism/ Side Effects Comments
Elimination
Emergent management
Lorazepam 0.1 mg/kg IV, up to 4 mg GABAA agonist, Hepatic/renal Respiratory depression, Preferred first-line
per dose (in 5-to 10-min increases frequency Low lipid solubility, some hypotension treatment for in-hospital SE
intervals) of Cl- channel stays in intravascular (less than propofol or Significant tachyphylaxis
opening space barbiturates)
IV contains propylene
glycol
Midazolam 0.2 mg/kg IM (max 10 mg) GABAA agonist, Hepatic/renal Respiratory depression, In prehospital setting, has
increases frequency some hypotension been shown to be at least
of Cl- channel (less than propofol or as effective as IV lorazepam
opening barbiturates) in cessation of SE
Significant tachyphylaxis
Diazepam 0.15 mg/kg IV (max 10 mg GABAA agonist, Hepatic/renal Respiratory depression, Can accumulate in the
per dose), repeat in 5 min increases frequency Highly lipid-soluble, some hypotension body with repeated doses
0.2 mg/kg PR (for age of Cl- channel quickly absorbed and (less than propofol or Significant tachyphylaxis
>12 years) opening redistributed out of barbiturates)
IV contains propylene
intravascular space glycol
(within 15–30 mins)
Phenytoin/ 20 mg/kg IV (can load an Blocks voltage-gated Hepatic/renal and Hypotension, bradycardia, Fosphenytoin is water-
Fosphenytoin additional 10 mg/kg IV if Na+ channels biliary arrhythmias with infusion soluble prodrug, less likely
seizures persist) Cytochrome P450 Therapeutic levels can to cause local reactions,
Check level 2 hrs after inducer, multiple drug cause or exacerbate can be infused three times
loading: interactions pancytopenia, drug fevers, faster than phenytoin (150
and rash (including SJS) mg PE/min vs. 50 mg/min)
▸ If in desired range, start
maintenance dose (typically Toxic levels can cause 90% protein bound,
100 mg q8 hrs) sedation, nystagmus, should check free level; if
diplopia, dysarthria, ataxia, unavailable, check serum
▸ If supratherapeutic, albumin and calculate
hold dose until in tremor, and coma
corrected level
therapeutic range
Normal therapeutic range
▸ If subtherapeutic, give 1–2 free, 10–20 corrected;
another smaller loading in SE, typically aim for 2–
dose and start maintenance 2.5 free, 20–25 corrected
Valproic acid 20-40 mg/kg IV (can load Blocks voltage-gated Hepatic/renal Hyperammonemia, Noninferior to phenytoin
with another 20 mg/kg IV if Na+ channels and T- Cytochrome P450 thrombocytopenia, in clinical studies, has
seizures persist) type Ca2+ channels inhibitor; multiple drug sedation, pancreatitis, advantage of multiple
Check level 2 hr after May increase interactions tremor, elevated mechanisms
loading, but do not delay turnover of GABA transaminases Typical therapeutic range
maintenance dose (typically (and therefore Significant teratogenicity, 50–100 µg/ml (total), but
starting at 15 mg/kg divided increase GABA avoid in pregnant patients can be as high as 150 in SE
over two doses); lab levels) unless absolutely necessary
processing time longer than
PHT levels
Levetiracetam Load with 1 g–3 g IV bolus, Binds to synaptic Enzymatic hydrolysis/ May cause mood Also a preferred agent for
then start maintenance dose vesicle glycoprotein renal disturbances seizure prophylaxis
(typically 500–1500 mg SV2A, inhibits No major drug Levels not monitored
q12 hrs) presynaptic Ca2+ interactions
channels
(continued)
Table 8.1 Continued
Elimination
Adjust dose for renal function:
▸ CrCl <30: 250–500 mg q12 hr
▸ HD dosing: 500–1000 mg
daily with extra 250–500 mg
after HD
Refractory management
Midazolam (0.2 mg/kg IV boluses see Emergent see Emergent see Emergent Management Preferred first infusion
(max 2 mg/kg) until Management Management for RSE
seizure cessation; then
maintenance infusion of
(0.05–2.9 mg/kg/hr)
Propofol 1–2 mg/kg boluses until Acts on GABAA Hepatic Significant hypotension and Less tachyphylaxis than
seizure cessation, then receptor (different glucuronidation/ respiratory depression BZDs
maintenance infusion site than BZDs hepatic PRIS: cardiac arrhythmias, PRIS associated with
30–100 µg/kg/min or barbiturates), Rapidly redistributed heart failure, hyperkalemia, prolonged use and high
potentiating its into peripheral tissues, lactic acidosis, ↑ creatinine doses, especially in
activity leading to a short kinase and transaminases, pediatric and septic patients
duration of clinical rhabdomyolysis, renal Monitor BMP, LFTs, and
effect failure triglycerides
Pentobarbital 5–15 mg/kg boluses until GABAA agonist, Hepatic/renal Hypotension, prolonged Was mainstay of treatment
seizure cessation, then increase duration of 15–50 hr half-life sedation and respiratory for refractory SE, now
maintenance infusion 1–10 Cl- channel opening depression, cardiotoxicity, typically reserved for
mg/kg/hr Also block excitatory paralytic ileus; at SE refractory to other
AMPA and kainite high doses, potential infusions
receptors immunosuppression IV contains propylene
glycol
Ketamine Retrospective studies NMDA receptor Hepatic/renal May cause hypertension as Clinical data limited;
suggest 1.5 mg/kg bolus antagonist opposed to hypotension retrospective studies
dose, followed by minimum Previously thought to suggest ketamine infusion
maintenance infusion of 0.9 potentially increase ICP, but is safe and potentially
mg/kg/hr this is now controversial efficacious, especially when
initiated earlier
May have synergistic effect
when combined with BZDs
Phenobarbital 15–20 mg/kg IV loading GABAA agonist, Hepatic/renal Hypotension, prolonged Can be considered as
dose increase duration of 60–120 hr half-life sedation, and respiratory a bridge while weaning
Maintenance dose 50–100 Cl- channel opening depression pentobarbital
mg two to three times daily Also block excitatory Monitor levels; typical
AMPA and kainate range 30–50 µg/ml in SE
receptors IV contains propylene
glycol
Lacosamide Load with 200–400 mg IV Enhances slow Demethylation/renal May cause arrhythmias and Limited evidence in
bolus, then start 100–200 inactivation of No major drug bradycardia; caution with treatment of SE
mg q12 hrs voltage-gated Na+ interactions prolonged QTc Levels not monitored
channels
Adjust dose for renal
function:
▸ CrCl <30: 50–150
mg q12 hr
▸ HD dosing:
Supplement with 50%
of dose after HD
(continued)
Table 8.1 Continued
Elimination
Topiramate Load with 400–1600 mg PO Blocks voltage-gated Hepatic/renal Sedation, nephrolithiasis, No IV form
over 24 hrs (two to four Na+ channels and At high doses can be acute angle-closure Has the advantage of
divided doses), then start high voltage-activated a cytochrome P450 glaucoma multiple mechanisms; also
100–200 mg BID Ca2+ channels inducer Malignant hyperthermia has carbonic anhydrase
Enhances GABAergic and severe metabolic activity
activity; inhibits AMPA acidosis are rare, but high Levels not monitored
and kainate receptors doses increase the risk
Clobazam 10–40 mg PO per day over GABAA agonist Hepatic/renal Sedation (though less than No IV form
two divided doses As a 1,5- May interact with other other BZDs) Can be used as a bridge
benzodiazepine, cytochrome P450 while weaning other BZDs
structurally different substrates 80%–90% protein bound
from and provides
for more selectivity Levels not monitored
than traditional
1,4-benzodiazepines
Super-refractory considerations (with only limited evidence from case series)
Other AEDs (e.g. felbamate, clonazepam, carbamazepine, oxcarbazepine, etc.)
Ketogenic diet (consult nutritionist for tube feeding recommendations to induce ketosis)
IV steroids and/or IVIG (if there is concern for autoimmune encephalitis
Hypothermia
Inhaled anesthetics
Electroconvulsive therapy
Surgical resection (if epileptogenic focus identified)
Allopregnanolone (currently being investigated in clinical trials)
SE = status epilepticus; IV = intravenous; PO = by mouth; IM = intramuscular; PHT = phenytoin; HD = hemodialysis; RSE = refractory status epilepticus; BZD = benzodiazepine;
PRIS = propofol infusion syndrome; BMP = basic metabolic panel; LFT = liver function tests; ICP = intracranial pressure; AED = antiepileptic drug; IVIG = intravenous immunoglobulin therapy.
Special Considerations
• In hepatic failure, strong consideration should be given to using levetiracetam
and lacosamide as AEDs of choice.
• In nonoliguric renal failure, renally cleared AEDs may still be used as long as
they are appropriately dose-adjusted; however, they should be avoided in
patients who are oliguric or anuric and not receiving dialysis.
• Consider a benzodiazepine challenge in situations where it is unclear
whether EEG (and clinical) findings represent ictal or interictal activity. This
involves administration of a relatively small dose of a short or intermediate
acting benzodiazepine (e.g., 1–2 mg midazolam or 0.5–1 mg lorazepam and
assessing for both electrographic and clinical improvement). If after 10 to
Chapter 8
15 minutes there is no improvement, another dose is given (at the same
or slightly larger dose) and assessment is repeated, for a total of up to
three doses.
• Myoclonic SE, while historically thought to portend a poor prognosis
after anoxic brain injury, may still have the potential for a good outcome
and should therefore be treated aggressively when consistent with goals
of care, especially since it is still unclear whether treatment intensity may
have a benefit on patient outcomes. The AEDs of choice are valproic acid
79
and levetiracetam, along with continuous infusions of GABAA agonists like
midazolam.
• Alcohol withdrawal seizures are primarily treated with GABAA agonists, es-
pecially phenobarbital or benzodiazepines; other AEDs are generally not
recommended unless RSE ensues.
Outcomes after Status Epilepticus

SE remains a frequently lethal disorder despite medical intervention, with
in-hospital mortality rates ranging from 9.4% to 21% and increasing to 23%
to 61% for RSE. Separate analyses of NCSE, meanwhile, have described
mortality rates ranging from 18% to 52%. Much of this variability is due to
characteristics of the patient population studied, especially whether or not
myoclonic SE is included, which various studies have suggested carries a poor
prognosis. However, many of these studies are inevitably fraught with bias as a
result of self-fulfilling prophecy, as myoclonic SE is often not treated as aggres-
sively as other forms of SE. Even so, cases of good outcomes after myoclonic
SE have been described.
For survivors of RSE, up to three-quarters of patients will have a poor
functional outcome (modified Rankin Scale 4–5) on discharge; Kilbride et al.
also followed survivors for at least six months and noted a long-term poor
functional outcome in two-thirds of these patients. The most consistent
predictors of discharge outcome are etiology (with outcomes worse for SE
secondary to new brain injury or lesion, especially from anoxia), duration
of coma, and mechanical ventilation. Other factors that may lead to worse
outcomes include older age, medical comorbidities, and high initial APACHE
scores, though these have not been consistently reproduced across studies.
Note again, though, that some of these same studies also showed that some
patients may survive with a good outcome despite negative prognostic
indicators.
Meanwhile, regardless of functional outcome, many patients who survive
RSE will continue to have seizures, though exactly how many is unclear given a
lack of longitudinal data. However, one study by Hesdorffer et al. showed that
the risk of recurrent SE over a 10 year follow-up period was 32%.
SECTION 1
EEG Monitoring in the Intensive Care Unit

The availability of continuous EEG monitoring in some centers has in-
creased the awareness of seizures and SE, particularly when they are
nonconvulsive; up to a third of patients with brain injury in the intensive
care unit (ICU) can have nonconvulsive seizures depending on etiology.
This awareness is also spreading to other patient populations: retrospec-
tive studies by Oddo et al. and Kurtz et al. showed that between 10% and
16% of patients undergoing EEG monitoring for altered mental status in
80
medical and surgical ICUs, respectively, were found to have electrographic

seizures, while a prospective study by Gilmore et al. showed that 11% of
all patients admitted to a medical ICU with severe sepsis were found to
have electrographic seizures on continuous EEG. Therefore, there should
be a low threshold to consider usage of continuous EEG monitoring in
the ICU, especially when actively treating RSE and as infusions are weaned
(until the patient awakens and/or remains seizure-free for several days), in
coma status-post cardiac arrest, with otherwise unexplained coma in any
other etiology, and with otherwise unexplained fluctuating mental status in
patients with critical illness or brain injury.
Once started, however, the dilemma becomes when to discontinue EEG
monitoring. In one study by Claassen et al. 20% of patients with recorded
seizures did not have their first one until after the first 24 hours of monitoring,
while 13% did not have one until after 48 hours of monitoring. Meanwhile,
a more recent study by Westover et al. showed that a lack of epileptiform
discharges in the first two hours of monitoring was predictive of a <5% risk
of seizures by 72 hours; whereas if epileptiform discharges were present, 16
hours of seizure-free monitoring was required to have a similarly low risk
of seizures at 72 hours. Despite this, there remains no consensus on dura-
tion of monitoring. However, it is reasonable to discontinue continuous EEG
monitoring after 12 to 24 hours of recording if patients have not had any
epileptiform discharges whatsoever and are not comatose. If there are ep-
ileptiform discharges but they are relatively infrequent, at least 24 hours of
recording is still recommended; otherwise, more aggressive electrographic
patterns warrant monitoring for at least 48 to 72 hours, especially if patients

are comatose.
Another dilemma is posed by electrographic patterns that are ambiguous.
While EEG patterns for NCSE can vary, the typical appearance includes
rhythmic spike-and-wave discharges that may be continuous or intermittent
with evolution. Sometimes, high-amplitude rhythmic delta with intermixed
spike or sharp waves may also represent NCSE. However, borderline
patterns, including periodic epileptiform discharges, may be more difficult
to interpret, as these may also represent seizures if there is a progression
to faster frequencies (>2 Hz), higher amplitudes, signs of evolution with in-
creasing rhythmicity or spread, or positive clinical symptoms. Otherwise,
their presence indicates a high risk of developing seizures. Meanwhile, depth
EEG recordings have revealed that some patients with only rhythmic slowing
Chapter 8
on surface EEG may have concurrent deep focal seizures, which may even
be more prevalent than seizures involving larger areas of the brain (though
they typically go unrecognized). Their significance is unclear, but they appear
to be associated with worse outcomes than if there were no seizures or only
surface seizures.
If the EEG remains ambiguous, other means for determining whether findings
represent ictal or interictal activity have been occasionally employed. This includes
a benzodiazepine challenge as previously mentioned or studies of metabolic ac-
81
tivity, including either noninvasive studies with perfusion or SPECT imaging or in-
vasive procedures with multimodality monitoring (examining for changes in brain
tissue oxygenation and microdialysis measures that indicate a hypermetabolic
state). Continuous EEG itself is a major component of multimodality monitoring,
and quantitative EEG analysis with compressed spectral array can be used to de-
tect a number of cerebral disturbances aside from seizures.
Further Reading
Alldredge BK, Gelb AM, Isaacs SM, et al. A comparison of lorazepam, diazepam,
and placebo for the treatment of out-of-hospital status epilepticus. N Engl J Med.
2001;345(9):631–637.
Brophy GM, Bell R, Claassen J, et al. Guidelines for the evaluation and management of
status epilepticus. Neurocrit Care. 2012;17(1):3–23.
Claassen J, Mayer SA, Kowalski RG, Emerson RG, Hirsch LJ. Detection of electrographic
seizures with continuous EEG monitoring in critically ill patients. Neurology.
2004;62:1743–1748.
Claassen J, Riviello JJ, Silbergleit R. Emergency neurologic life support: status epilepticus.
Neurocrit Care. 2017;27(Suppl. 1):152–158. doi:10.1007/s12028-015-0172-3
DeLorenzo RJ, Waterhouse EJ, Towne AR, et al. Persistent nonconvulsive status epilep-
ticus after the control of convulsive status epilepticus. Epilepsia. 1998;39(8):833–840.
Gaspard N, Foreman B, Judd LM, et al. Intravenous ketamine for the treatment of
refractory status epilepticus: a retrospective multicenter study. Epilepsia. 2013
Aug;54(8):1498–1503.
Gilmore EJ, Gaspard N, Choi HA, et al. Acute brain failure in severe sepsis: a pro-
spective study in the medical intensive care unit utilizing continuous EEG monitoring.
Intensive Care Med. 2015 Apr;41(4):686–694.
Hesdorffer DC, Logroscino G, Cascino GD, Hauser WA. Recurrence of afebrile status
epilepticus in a population-based study in Rochester, Minnesota. Neurology. 2007 Jul
3;69(1):73–78.
Hocker SE, Britton JW, Mandrekar JN, Wijdicks EF, Rabinstein AA. Predictors of out-
come in refractory status epilepticus. JAMA Neurol. 2013 Jan;70(1):72–77.
Kilbride RD, Reynolds AS, Szaflarski JP, Hirsch LJ. Clinical outcomes following prolonged
refractory status epilepticus (PRSE). Neurocrit Care. 2013 Jun;18(3):374–485.
SECTION 1
Kurtz P, Gaspard N, Wahl AS, et al. Continuous electroencephalography in a surgical

intensive care unit. Intensive Care Med. 2014 Feb:40(2):228–234.
Legriel S, Azoulay E, Resche-Rigon M, et al. Functional outcome after convulsive status
epilepticus. Crit Care Med. 2010;38:2295–2303.
Lowenstein DH, Alldredge BK. Status epilepticus at an urban public hospital in the
1980s. Neurology. 1993;43:483–488.
Novy J, Logroscino G, Rossetti AO. Refractory status epilepticus: a prospective obser-
vational study. Epilepsia. 2010;51(2):251–256.
Oddo M, Carrera E, Claassen J, Mayer SA, Hirsch LJ. Continuous electroencephalog-
raphy in the medical intensive care unit. Crit Care Med. 2009 Jun;37(6):2051–2056.
Silbergleit R, Durkalski V, Lowenstein D, et al. Intramuscular versus intravenous therapy
82
for prehospital status epilepticus. N Engl J Med. 2012;366(7):591–600.

Sutter R, Marsch S, Fuhr P, Rüegg S. Mortality and recovery from refractory status
epileticus in the ICU: a 7-year observational study. Epilepsia. 2013;54(3):502–511.
Sutter R, Kaplan PW. Can anesthetic treatment worsen outcome in status epilepticus?
Epilepsy Behav. 2015 Aug;49:294–297.
Treiman DM, Meyers PD, Walton NY, et al. A comparison of four treatments for gener-
alized convulsive status epilepticus. N Engl J Med. 1998;339(12):792–798.
Westover MB, Shafi MM, Bianchi MT, et al. The probability of seizures during EEG mon-
itoring in critically ill adults. Clin Neurophysiol. 2015 Mar;126(3):463–471.
Chapter 9
The Management of Traumatic

Brain Injury
Jeremy G. Stone, David M. Panczykowski,
and David O. Okonkwo
General Principles
Epidemiology
In the United States, traumatic brain injury (TBI) results in 2.5 million emer-
gency department (ED) visits per year with 280,000 requiring hospitaliza-
tion and is responsible for 53,000 deaths annually. The leading causes of TBI
83
resulting in hospitalization include falls, motor-vehicle collisions, and assaults.
Age-adjusted rates of TBI-related ED visits are increasing (457.5 in 2007 to
715.7 in 2010) whereas TBI-related hospitalizations have remained the same
(91.7 per 100,000 people) and TBI-related deaths have decreased (18.2 to
17.1 per 100,000 people) over the same time period. Taken together, the
prevalence of TBI confers an immense burden on the health care system with
regard to direct medical expenditures, rehabilitation of long-term TBI-related
disability, and indirect cost related to loss of productivity.
Definitions
The most widely utilized grading system for TBI severity is the Glasgow Coma
Scale (GCS), which stratifies TBI into mild (e.g., concussion; GCS 13–15),
moderate (GCS 9–12), and severe (GCS 3–8). The GCS assesses a patient’s
level of consciousness based on scores for eye opening, verbal response, and
motor response (detailed in assessment section later).
The primary injury, which occurs at the time of initial trauma, may be
caused by blunt, blast, and/or penetrating mechanisms and is defined as death
or dysfunction of central neurons secondary to the initial traumatic impact.
Secondary injury includes progressive neuronal injury that develops subse-
quent to the initial trauma caused by inflammatory, ischemic, and/or cyto-
toxic insults that may be secondary to or modulated by both intracranial and
systemic processes. In the critical care setting, nothing can be done to mitigate
primary injury, which by definition has already occurred. However, prophy-
laxis, early recognition, and rapid treatment of subsequent insults may modu-
late secondary injury and improve outcome after TBI.
Key physiologic factors implicated in secondary brain injury include hypo-
tension, hypoxia, intracranial hypertension, and inadequate cerebral perfusion

pressure, all of which ultimately result in cerebral ischemia. Evidence-based
protocols in severe TBI management are directed toward avoidance and treat-
ment of cerebral ischemia and have reduced mortality rates in severe TBI
from 50% to less than 25% over the past 30 years.
Pathophysiology of Cerebral Ischemia
When there is insufficient cerebral blood flow (CBF) to meet the metabolic
demands of the brain, cerebral ischemia results. Unique among our organs,
SECTION 1
the brain lacks the capacity for significant oxygen storage, has minimal capacity
to store glycogen, and is nearly completely dependent on blood flow to pro-
vide substrates of metabolism. As such, normal central nervous system phys-
iology has autoregulatory mechanisms which couple the cerebral metabolic
rate of oxygen to CBF to ensure that the high metabolic demands of the brain
parenchyma are met. Cerebral autoregulation allows CBF to be kept constant
over a wide range of blood pressures. Through dynamic vasoconstriction and
dilation, cerebral arterioles can maintain a consistent level of blood and ox-
ygen delivery when systolic blood pressure ranges between 50 and 150 mm
Hg (Figure 9.1). Outside of this autoregulatory threshold, however, arteri-
olar vasoconstriction or dilation fail to maintain constant CBF and cerebral
84
ischemia may ensue.
75
Cerebral Blood Flow (ml/100 g/min)
Passive Maximum Zone of Normal Maximum

Collapse Dilatation Autoregulation Constriction
50
25
0
0 25 50 75 100 125 150
Systolic Blood Pressure (mmHg)
Figure 9.1 Cerebral blood flow is maintained at relatively constant rate over a wide range
of cerebral perfusion pressures via dynamic arteriolar dilation and vasoconstriction.
Source: White H, Venkatesh B. Cerebral perfusion pressure in neurotrauma: a review. Anesth Analg.
2008;107:979–988. Printed with permission of Wolters Kluwer Health, Inc.
Cerebral autoregulation is often impaired in TBI, thus several physiologic
Traumatic Brain Injury

parameters involved in autoregulation are implicated in increased risk for
cerebral ischemia (Figure 9.2). Cerebral perfusion pressure (CPP) is defined
as mean arterial pressure (MAP) minus intracranial pressure (ICP) and is
a surrogate marker for CBF. Low CPP is associated with higher mortality
and morbidity in TBI, particularly when systemic hypotension occurs con-
comitantly. Likewise, hypoxia (arterial oxygen saturations <60%), systemic
hypotension (systolic blood pressures <100–110 mm Hg), elevated ICP
(ICP >20 mm Hg), low brain tissue oxygen saturation (PbtO2 <15–20 mm
Hg), and hyperthermia (core temperature >38.5 °C, or brain tissue tem-
Chapter 9
perature >37.5 °C) are linked with higher morbidity and mortality in severe
brain injury.
ICP is a key factor influencing CPP. Dictated by the fundamental principles
of the Monro-Kellie doctrine, ICP is determined by the dynamic interplay
between blood, cerebrospinal fluid (CSF), and brain parenchyma, the three
components which occupy the fixed volume within the skull. If any one of
these three components enlarges in volume or there is a new mass lesion
introduced within the skull by trauma, there must be a compensatory cor-
responding decrease in volume of the other components in order to main-
tain the same pressure. The system has some capacity to compensate for
new intracranial mass lesions by decreasing venous blood volume and/or CSF
85
volume; however, once compensatory measures are overcome, there is an in-
evitable rise in ICP. Elevated ICP may ultimately lead to decreased CPP and ce-
rebral ischemia or brain herniation, thereby contributing to secondary injury.
(a) (b)
CBF (ml/g/min)
50 150 90 110
MABP (mm Hg)
Figure 9.2 Cerebral perfusion pressure autoregulation in normal patients (A) versus
patients with traumatic brain injury (B).
Source: Rangel-Castilla L, Gasco J, Nauta HJ, Okonkwo DO, Robertson CS. Cerebral pressure
autoregulation in traumatic brain injury. Neurosurg Focus. 2008;25:1–8.
Clinical Assessment and Classification

General Principles
As in all traumatic injury, primary assessment and management should proceed
based on Advanced Trauma Life-Support recommendations. Of patients with
GCS ≤8, 60% have one or more other organ systems injured. As such, a focus
on airway, breathing, and circulatory systems precedes neurologic assessment.
Neurologic Assessment and Classification
SECTION 1
The GCS developed by Teasdale and Jennett in 1974 is the most widely used,
reproducible, and rapid initial clinical assessment tool in TBI. Three components
including eye opening, verbal response, and motor response are assessed and
summed to quickly determine a patient’s overall level of consciousness (Table
9.1). Mechanical ventilation via endotracheal tube or tracheostomy precludes
a patient’s ability to produce a verbal response, thus the verbal score of 1 is
designated with a following “T” to highlight the mechanically ventilated status
of the patient. Limitations of the GCS include scenarios with concomitant use
Table 9.1 The Glasgow Coma Scale

86
Best Response Score

Eye Opening (E)
• Spontaneous 4
• To speech 3
• To pain 2
• Not open 1
Verbal Response (V)
• Conversant and oriented 5
• Confused 4
• Nonsense 3
• Sounds (moans) 2
• Silent 1
• Intubated 1T
Motor Response (M)
• Follows commands 6
• Localizes to pain 5
• Withdraws from pain 4
• Arm flexion to pain 3
• Arm extension to pain 2
• No response 1
GCS score = E + V + M (range 3 or 3T to 15)
of sedating alcohol, drugs, and/or administered medications; presence of facial

trauma limiting eye opening or verbal responses; and/or assessment shortly
after rapid sequence intubation in which neuromuscular blockade is in effect.
Coupled with the GCS, pupillary and motor examinations are key
components of an initial rapid neurologic assessment. A unilateral dilated,
nonreactive pupil suggests ipsilateral uncal herniation with subsequent ipsi-
lateral oculomotor nerve dysfunction. Alternatively, bilaterally dilated and
unresponsive pupils may by observed in severe irreversible brainstem injury,
hypotension, hypoxia, or bilateral oculomotor nerve dysfunction. Patients
with an admission GCS of 3 and bilaterally dilated unresponsive pupils have
Chapter 9
near 100% mortality. The presence of focal weakness on motor examination
may suggest a lateralized mass lesion. Most often, uncal herniation of the me-
dial temporal lobe causes disruption of the corticospinal tracts on the side of
the mass lesion and presents with contralateral weakness. However, in 10%
of cases, a Kernohan notch phenomenon or “false localizing sign” may be ob-
served in which a lateralized mass lesion will cause weakness on the same side
as the lesion due to the contralateral corticospinal tracts being compressed
against the contralateral edge of the tentorium cerebelli.
Monitoring in TBI
87
CPP is used as a surrogate for the adequacy of CBF, therefore ICP and MAP
should be monitored in severe TBI. Continuous systemic blood pressure
monitoring via indwelling arterial line, ICP monitoring via invasive intracranial
monitor, and continuous pulse oximetry are recommended in the critical care
setting. ICP monitors may take the form of intraparenchymal monitors or
external ventricular drains (EVD), which are placed within the ventricle. With
regard to intraparenchymal monitors, devices on the market employ various
technologies including fiberoptic transduction, microtip strain gauge transduc-
tion, or pneumatic transduction to measure ICP. Notably, EVDs have the dual
ability to both directly transduce ICP and treat elevated ICP via diversion of
CSF. However, when an EVD is open to drain, ICPs are unable to be trans-
duced. Thus, per protocol at our institution, all severe TBI patients undergo
tandem placement of EVDs and intraparenchymal ICP monitors, permitting
ICP to be continuously transduced even while CSF is actively being diverted
via EVD. This practice is not universal, and local protocols should be estab-
lished by the managing teams.
While these physiologic parameters serve as surrogate markers for ce-
rebral oxygenation, brain tissue oxygenation (PbtO2) may also be directly
assessed via invasive fiberoptic catheter placement into the brain parenchyma.
Low PbtO2 levels (<15–20 mm Hg) and longer duration of decreased PbtO2
(>30 minutes) are associated with higher mortality rates in severe TBI. Many
protocols set a PbtO2 level of ≥20 mm Hg as the targeted goal for patient
management.
Management of TBI
Care Setting
Patients with severe TBI should be cared for in a tertiary care center with a
high volume of TBI patients. All severe TBI patients should be managed in the
intensive care unit.
CPP: Goal >60 mm Hg
Current management guidelines for severe TBI advocate for a CPP threshold
SECTION 1
of 60 to 70 mm Hg. Driving CPP above 70 mm Hg with the use of vasopressors

and/or intravascular volume expansion has been associated with significantly
higher rates of acute respiratory distress syndrome. Thus, aggressive man-
agement of CPP greater than 70 mm Hg is not recommended. Overall, hy-
potension should be avoided, with a goal systolic blood pressure of 100 to
110 mm Hg and greater. When inadequate CPP is encountered clinically
without elevated ICP, differential diagnosis should include systemic causes of
hypotension including bleeding, tension pneumothorax, and spinal or cardiac
shock. Intravascular resuscitation with isotonic fluids should be an initial step
in therapy. In cases of refractory low CPPs, vasopressors may be required.
ICP: Goal <20 mm Hg
88
Intracranial hypertension has been associated with worse patient outcomes

in severe TBI. In the neurocritical care setting, aggressive management of
ICP is a management cornerstone. Intracranial hypertension treatment is
based on a multimodal approach involving patient positioning, pharmacologic
interventions, and potentially surgical intervention.
First-line treatments, when intracranial hypertension is encountered, in-
clude the following:
• Patient positioning: head in neutral position with head of bed raised to
at least 30 degrees
• Cervical spine clearance and removal of rigid cervical collars should be
expedited
• Normocapnea (PaCO2 between 35 and 40 mm Hg)
• Appropriate analgesia and pharmacologic sedation
• Hyperosmolar therapy
Second-line treatments must be employed in up to 25% of severe TBI
patients who will demonstrate intracranial hypertension refractory to first-line
therapies:
• Neuromuscular blockade
• Hyperventilation (PaCO2 between 30 and 35 mm Hg)
• Hypothermia (Temperature <36° C)
• Barbiturate-induced burst suppression
• Decompressive craniectomy
First-Line ICP Treatments

Rational for prioritizing these first-line treatments includes that the treatments
are relatively low risk with a low adverse event profile, they can be rapidly
administered, and they are the least invasive. Elevating the head of bed and
maintaining a neutral head position serves to reduce ICP via maximization
of venous blood outflow from the head, thereby decreasing intracerebral
blood volume while concomitantly modulating CBF and CPP. Regarding cer-
vical spine clearance, expeditious removal of external cervical collars allows
decreased kinking of neck veins which, similar to bed positioning, can facili-
Chapter 9
tate venous outflow from the head. Based on mounting evidence highlighting
the sensitivity and specificity of helical computed tomography (CT) scanning
of the cervical spine for determination of unstable cervical spine injuries, our
institutional policy is to clear cervical spine precautions following negative
helical CT scans in patients with no neurological deficit that can be ascribed
to a cervical spinal cord injury.
Additionally, analgesia and pharmacologic sedation are first-line therapies
with the goal of minimizing pain and agitation, which physiologically are shown
to induce elevations in blood pressure, ICP, body temperature, and resistance
to controlled mechanical ventilation, all of which can contribute to secondary
brain injury. Sedation in TBI patients can render accurate neurologic assessment
problematic, however, a variety of pharmacologic agents are good candidates
89
for use in the TBI population. Propofol and dexmedetomidine are have rela-
tively short half-lives and thus are able to be turned off for accurate neurologic
assessments and are shown to reduce ICP and have neuroprotective proper-
ties, respectively. Boluses of short-acting opioids, such as fentanyl, are also
highly effective for pain and agitation control without significantly impacting
the ability to complete a neurologic exam.
Last among first-line therapies, hyperosmolar therapy, including use of hy-
pertonic saline or mannitol, may be employed for effective medical control
of ICP. Hypertonic saline in the form of 3% continuous infusions and/or 30
cc boluses of 23.4% saline may be introduced via central venous catheter
to reduce ICP via both an osmotic and rheologic effect. Mannitol is admin-
istered via bolus dosing of 0.25 g/kg to 1 g/kg body weight; it reduces ICP
and also exhibits dual rheologic and osmotic effects. Rheologic properties in-
clude the ability to expand plasma volume, reduce hematocrit, and increase
red blood cell deformability, all of which ultimately decrease blood viscosity,
increase CBF, and subsequently decrease ICP via arteriolar vasoconstriction.
Osmotically, mannitol creates a gradient between intravascular and interstitial
space in brain parenchyma, thereby encouraging water diffusion from the pa-
renchyma into the cerebral vasculature.
Increasing evidence suggests hypertonic saline is overall more effective in
a greater percentage of patients for ICP reduction than mannitol. Rebound
intracranial hypertension is a troubling adverse phenomenon which can be
observed with mannitol use and is thought to be secondary to uptake of man-
nitol into the brain, causing a paradoxical reverse osmotic gradient. The blood
brain barrier, however, is thought to be less permeable to hypertonic saline
solutions, and thus rebound ICP spikes are less commonly seen with hyper-
tonic saline use. Importantly, mannitol can precipitate acute tubular necrosis,
leading to acute renal failure, particularly in cases with elevated serum osmo-
larity greater than 320 mOsm. Mannitol encourages profound diuresis, which
can precipitate electrolyte abnormalities and intravascular volume depletion
as well. Hypertonic saline use is also not without risk. Vigilant attention to
serum sodium levels is critical when administering hypertonic saline, as a rapid
rise in serum sodium can precipitate central pontine myelinolysis, especially in
patients with underlying chronic hyponatremia. Other risks of hypertonic sa-
SECTION 1
line use include pulmonary edema, seizures, coagulopathy, phlebitis, and other
electrolyte abnormalities. Goal sodium levels in this setting are generally 140
to 160 mEq/L.
Second-Line ICP Treatments
When first-line therapies are exhausted and intracranial hypertension remains
refractory, second-line therapies may be necessary, including neuromuscular
blockade, hypothermia, barbiturate-induced coma, hyperventilation, and sur-
gical decompression. Neuromuscular blockade via continuous infusion of
agents with mechanism of action at the neuromuscular junction reduces ICP
via decreasing intrathoracic pressure, facilitating venous blood outflow from
the head and completely removing the patient’s ability to resist against con-
90
trolled mechanical ventilation. Agents with a short half-life (e.g., cisatracurium)

are preferred to reduce interference with neurologic examination as much as
possible, and dosing can be titrated to a patient’s twitch response to a train of
four peripheral nerve stimulator. Complications of neuromuscular blockade
include higher rates of infection (e.g., pneumonia, sepsis) and longer intensive
care unit stays.
Induction of hypothermia to a temperature below 36° C may lower brain
metabolism and reduce ICP and can be trialed in refractory intracranial hyper-
tension. This treatment should not be used prophylactically, as multiple studies
have shown there is no benefit on long-term outcomes. Hyperventilation has
also been shown to reduce ICP by producing vascular constriction and low-
ering CBF. This intervention should only be used transiently as it may lead to
cerebral ischemia and infarction, thus resulting in additional secondary injury.
Therapy with barbiturate-induced coma requires continuous EEG moni-
toring. The most common agent used is pentobarbital, and treatment should
be initiated with a test bolus dose of 10 mg/kg. If there is a reduction in the
ICP in response to the test dose, then an infusion of pentobarbital should be
started at 1 mg/kg. The infusion dose is titrated up until electrographic burst
suppression of cerebral activity is achieved on EEG. Additional boluses of 5
mg/kg should be given every 30 to 60 minutes up to a total bolus dose of 20
mg/kg to achieve loading and facilitate transition to a burst suppression pat-
tern. The mechanism by which pharmacologic burst suppression promotes
ICP reduction is incompletely understood but thought to be secondary to
decreased cerebral metabolism, cerebral vasoconstriction, and prevention of

neurotoxic excitatory cascades. There is no benefit to prophylactic treatment
with burst suppression in severe TBI; however, patients with intracranial hy-
pertension refractory to all other interventions have decreased mortality if
ICPs respond to barbiturate therapy. Complications of barbiturate-induced
burst suppression include systemic hypotension and increased infection rates.
Pharmacologic burst suppression unfortunately obviates the ability to com-
plete neurologic exams for long periods of time, often 24 to 48 hours after
barbiturate infusions are stopped.
Chapter 9
Surgical Intervention
In cases where medical therapy fails and in other specific circumstances in-
volving focal mass lesions with mass effect and midline shift, surgical interven-
tion may be indicated. In patients with severe TBI, 25% will have lesions in
which surgical intervention is generally indicated. Additionally, patients who
show decreases in GCS score by ≥2 from the field to the ED are more likely
to require neurosurgery. Neurosurgical intervention for management of ICP
involves decompression via removal of a portion of the cranial vault. Taking
into account the Monro-Kellie doctrine, surgical decompression serves to
treat elevated ICP by removing the volumetric constraint of the nonexpansile
rigid skull (Figure 9.3). Most commonly, lateralized mass lesions such as sub-
91
dural hematoma, epidural hematoma, or contusion are treated with ipsilateral
decompressive hemicraniectomy. Much less commonly, bifrontal lesions such
as bifrontal contusions or generalized multifocal brain edema are treated with
bifrontal craniectomy. Rarely, a posterior fossa traumatic mass lesion will re-
quire a posterior approach for suboccipital craniectomy. Surgical intervention
can be lifesaving and highly effective for ICP management in medically refrac-
tory intracranial hypertension.
Brain Tissue Oxygen: PbtO2 >20 mm Hg
Current guidelines state that the use of advanced monitoring of brain oxygen-
ation in conjunction with ICP and CPP monitoring should provide additional
information regarding the metabolic demands of the brain. Evidence to sup-
port the routine use of PbtO2 monitoring remains controversial. Based on
clinical evidence of worse outcomes in patients with low PbtO2, if monitoring
is undertaken, the usual goal is to maintain levels >20 mm Hg. This can be
achieved in a number of ways, such as adjusting ventilator settings, increasing
CPP, red blood cell transfusions, or temperature control. There are potential
risks with any of these interventions, thus careful consideration must be used
based on the individual clinical scenario.
Hypoxia: Goal Arterial Oxygen Saturation >90%
A multimodal approach is necessary to prevent secondary brain injury re-
lated to cerebral ischemia. Establishing an early protected airway via endotra-
cheal intubation improves outcomes in severe TBI. Arterial oxygen saturations
1 2 3 ICP
Venous
50
CSF
Mass/
Venous
CSF
edema
Mass/
edema 30
SECTION 1
Arterial Arterial Arterial

blood blood blood
10
Brain Brain Brain
Normal Compensated Uncompensated
Figure 9.3 The rigid skull with fixed intracranial volume (represented by the rectangle)
contains brain, cerebrospinal fluid (CSF), and blood. Intracranial pressure (ICP) is
influenced by the dynamic interplay among the individual components (1). The addition of
another component within the intracranial space, such as posttraumatic cerebral edema
92
or mass lesion, will result in concomitant decrease in volume of another component in s

compensated state where ICP does not rise (2). However, an uncompensated state with
elevations in ICP will eventually ensue when compensatory measures are exhausted (3).
Source: Exo J, Smith C, Bell MJ. Emergency treatment options for pediatric traumatic brain injury. Pediatric
Health. 2009;3:533–541.
should be treated to maintain above 90%. Given oxygen delivery to the brain
is highly dependent on oxygen content of blood, anemia may also enhance
secondary brain injury. There is no current consensus on appropriate trans-
fusion thresholds in severe TBI; however, low hemoglobin is associated with
increased morbidity and mortality. Transfusion of packed red blood cells may
improve outcomes for the severely injured via multiple mechanisms including
improving cerebral oxygenation and increasing blood pressure which thereby
modulates CPP. However, transfusion is also associated with increased throm-
boembolic events, transfusion reactions, and increased rates of acute respira-
tory distress syndrome.
Hyperpyrexia: Goal Normothermia,
Core Body Temperature <38.5°C
An elevated core body temperature >38.5°C is shown to increase cerebral
metabolism and increases mortality in severe head injury. Cornerstones in
effective fever management include antipyretics, cooling blankets, and intra-
vascular cooling catheters. Our institutional protocol for severe TBI patients
involves immediate placement of a femoral or subclavian intravascular cooling
catheter which remains in place for at least 36 hours for close temperature

control in the acute postinjury period. Infectious workups should be pursued
aggressively in appropriate cases.
Glycemic Control: Goal Serum Glucose <180 mg/dL
Hyperglycemia is common after TBI secondary to activation of the sympa-
thetic nervous system and adrenergic catecholamine release. Both early and
persistent hyperglycemia is associated with poor functional outcomes in se-
vere TBI. Frequent glucose checks and tight glycemic control via sliding scale
insulin administration are paramount.
Chapter 9
Fluid Balance and Electrolytes
Electrolyte abnormalities are common in the TBI population, with up to
60% demonstrating an electrolyte abnormality in the acute posttrauma pe-
riod. Abnormal serum sodium concentrations are the most common alter-
ation, and posttraumatic hyponatremia is associated with worse outcomes
and increased hospital days. Underlying pathophysiology involved with post-
traumatic hyponatremia most often involves cerebral salt wasting or syndrome
of inappropriate antidiuretic hormone (SIADH). These two entities are distin-
guished based on volume status with SIADH resulting in eu-or hypervolemia,
whereas cerebral salt wasting results in hypovolemia. Distinguishing between
93
the two is important, as management differs. In SIADH, fluid restriction is in-
dicated, while sodium supplementation is the mainstay of therapy for cerebral
salt wasting.
Hypernatremia is not uncommon after TBI and is most often secondary to
central diabetes insipidus related to pituitary stalk dysfunction. Incidence of
posttraumatic central diabetes insipidus is directly related to severity of brain
injury. Management may involve a combination of free-water boluses or ad-
ministration of hypotonic fluids, and rarely necessitates use of desmopressin
(DDAVP). Following initial fluid resuscitation, the intravenous maintenance
fluid of choice in TBI is normal saline (0.9% NaCl). Fluids with dextrose should
be avoided given concern for inducing or worsening hyperglycemia.
Nutrition
Severe TBI and other frequently associated polytraumas can cause a hyper-
metabolic state. Enteral feeding (unless there is some absolute contraindica-
tion) should commence as soon as possible with full caloric replacement by
postinjury day 5. Patients with malnutrition during the first two weeks after
injury have significantly increased mortality when compared to patients who
have caloric needs met by postinjury day 7. Additionally, early initiation of en-
teral feeding is associated with decreased infection rates.
In choosing between orogastric and nasogastric enteral feeding tubes, one
must note the presence or absence of anterior skull base fractures. Placement
of nasogastric tubes in the setting of severe skull base fractures can adversely
result in intracranial penetration of the feeding tube. Other considerations
in TBI patients include the increased prevalence of delayed gastric emptying
and altered lower esophageal function which predispose to aspiration. The

need for long-term feeding may indicate placement of percutaneous endo-
scopic jejunal or gastrostomy tube.
Prophylaxis
Seizure Prophylaxis
Seizures can profoundly enhance secondary brain injury via increased ICP and
cerebral metabolic rate; decreased CBF, blood pressure, and oxygen delivery;
and promotion of neurotoxic excitatory neurotransmitter release. Seizures
SECTION 1
are more likely to be observed in patients with depressed skull fractures, cor-
tical contusions, penetrating injury mechanisms, and subdural/epidural hema-
tomas and in those with initial GCS <10. Early posttraumatic seizures, which
occur within seven days of injury, are observed in between 4% and 25% of
TBI patients. Late posttraumatic seizures, which occur after the seven-day
window, are seen in 9% to 42% of patients. Seizure prophylaxis with antiepi-
leptic (AED) agents does not impact incidence of late posttraumatic seizures;
however, it is shown to reduce early posttraumatic seizures. Thus, seven days
of seizure prophylaxis with an AED remains a standard recommendation.
Phenytoin is the most widely studied drug in the TBI population, thus it is
frequently used; however, newer AEDs, such as levetiracetam, have reduced
94
side effect profiles, particularly in the elderly, and may be another option.
Dosing regimens that achieve therapeutic levels are recommended. Ongoing
research seeks to elucidate if there is clinical superiority in seizure prevention
and outcomes between the various AEDs that are available.
Gastrointestinal Prophylaxis
Sympathetic nervous system activation with resulting catecholamine surges
can promote gastric ulcer formation after severe TBI. Routine use of hista-
mine type 2 antagonists or proton pump inhibitors for ulcer prevention is
common practice among mechanically ventilated patients with severe TBI.
Additionally, early enteral feeding and appropriate fluid resuscitation aid in
prevention of this complication.
Venous Thromboembolism Prophylaxis
Without the use of mechanical and pharmacologic prophylaxis, the incidence
of deep venous thrombosis (DVT) in the severe TBI population may be as
high as 20%. DVTs may lead to potentially life-threatening pulmonary embo-
lism (PE), which is seen in 0.38% of TBI patients during the acute recovery
phase. Given the immediate danger of PEs and the risk of progression of
intracranial bleeding in TBI patients who must be fully anticoagulated to treat
a DVT/PE, prevention of venous thromboembolism is critical. All in-hospital
patients are fitted with pneumatic compression devices for mechanical DVT/
PE prophylaxis. Pharmacologic prophylaxis does increase the risk a trauma-
related intracranial bleed will progress; thus, our institutional policy is to com-
mence pharmacologic prophylaxis with enoxaparin 30 mg every 12 hours on
posttrauma day 1, as long as CT scanning demonstrates stability of trauma-

related intracranial lesions.
Management to Avoid: Steroids
Steroid administration should be avoided in severe TBI as it is shown to in-
crease mortality and is associated with other complications including hypergly-
cemia, gastric ulcer formation, and gastrointestinal bleeding.
Conclusions
Chapter 9
The prevalence and consequences of TBI confer an immense burden on patients,
families, and our health care system. Evidence-based managements strategies
seek to intervene on pathophysiology which ultimately leads to cerebral is-
chemia and secondary brain injury. Modern care for TBI with multidisciplinary
treatment teams employing multimodal physical, pharmacologic, and surgical
therapies resulted in a drop in mortality from 50% to 25% in patients with severe
head injury. While much progress has been made, there remains vast room for
improvement in outcomes for TBI victims. Ongoing research is assessing novel
neuroprotective drugs and the role of advanced neuromonitoring management
of these patients, but much work remains to be done.
95
Further Reading
Andrews PJ, Sleeman DH, Statham PF, et al. Predicting recovery in patients suffering
from traumatic brain injury by using admission variables and physiological data: a
comparison between decision tree analysis and logistic regression. J Neurosurg.
2002;97:326–336.
Arabi B, Hsdorffer DC, Ahn ES, Aresco C, Scalea TM, Eisenberg HM. Outcome fol-
lowing decompressive craniectomy for malignant swelling due to severe head injury.
J Neurosurg. 2006;104:469–479.
Bouderka MA, Fakhir B, Bouaggad A, Hmamouchi B, Hamoudi D, Harti A. Early tra-
cheostomy versus prolonged endotracheal intubation in severe head injury. J Trauma.
2004;57:251–254.
Carney N, Totten AM, OʼReilly C, et al. Guidelines for the management of severe trau-
matic brain injury, fourth edition. Neurosurgery. 2017. 80(1):6–15
Deogaonkar A, Gupta R, DeGeorgia M, et al. Bispectral index monitoring correlates
with sedation scales in brain-injured patients. Crit Care Med. 2004;32:2403–2406.
Hartings JA, Vidgeon S, Strong AJ, et al. Surgical management of traumatic brain injury: a
comparative-effectiveness study of 2 centers. J Neurosurg. 2014:120(2):434–446
McIntyre LA, Fergusson DA, Hutchinson JS, et al. Effect of a liberal versus restrictive
transfusion strategy on mortality in patients with moderate to severe head injury.
Neurocrit Care. 2006;5:4–9.
Moro N, Karayama Y, Igarashi T, Mori T, Kawamata T, Kojima J. Hyponatremia in
patients with traumatic brain injury: incidence, mechanism, and response to
sodium supplementation or retention therapy with hydrocortisone. Surg Neurol.
2007;68:387–393.
Ogden AT, Mayer SA, Connolly ES Jr. Hyperosmolar agents in neurosurgical prac-
tice: the evolving role of hypertonic saline. Neurosurgery. 2005;57:207–215.
Roberts I, Sydenham E. Barbiturates for acute traumatic brain injury. Cochrane Database
Sys Rev. 1999;3:CD000033. doi:10.1002/14651858.CD000033
Robertson CS, Hannay HJ, Yamal JM et al. Effect of erythropoietin and transfusion
threshold on neurological recovery after traumatic brain injury: a randomized clinical
trial. JAMA. 2014;312(1):36–47.
Struchen MA, Hannay HJ, Contant CF, Roberson CS. The relation between acute phys-
SECTION 1
iological variables and outcome on the Glasgow Outcome Scale and Disability Rating
Scale following severe traumatic brain injury. J Neurotrauma. 2001;18:115–125.
Taylor SJ, Fettes SB, Jewkes C, Nelson RJ. Prospective, randomized, controlled trial to
determine the effect of early enhanced enteral nutrition on clinical outcome in me-
chanically ventilated patients suffering head injury. Crit Care Med. 1999;27:2525–2531.
Timofeev I, Dahyot-Fizelier C, Keong N, et al. Ventriculostomy for control of raised ICP
in acute traumatic brain injury. Acta Neurochir. 2008;102(Suppl.):99–104.
Tomycz ND, Chew BG, Chnag YF, et al. MRI is unneccessary to clear the cervical spine
in obtunded/comatose trauma patients: the four-year experience of a level I trauma
center. J Trauma. 2008;64:1258–1263.
Winchel RJ, Hoyt DB. Endotracheal intubation in the field improves survival in patients
with severe head injury. Arch Surg. 1997;132:592–597.
96
Chapter 10
The Management of Traumatic

Spinal Cord Injury
David M. Panczykowski, Jeremy G. Stone,
and David O. Okonkwo
Epidemiology
In the United States, the annual incidence of traumatic spinal cord injury (SCI)
ranges from 25 to 58 per million. The average age at time of injury is 32, with
a 4:1 male to female ratio. Traumatic brain injury (TBI) occurs in 25% to 50%
of patients with SCI; conversely, 5% to 10% of TBI patients sustain SCI. The
highest mortality is within the first three months of injury (~20%); however
97
the average life expectancy of a complete paraplegic SCI patient is only 16%
shorter than that of an uninjured peer and only 8% shorter for patients with
incomplete SCI.
Pathophysiology
Spinal cord injury, as with acute TBI, is a dynamic process with the full extent
of injury not initially apparent. Primary SCI is defined as neuronal death or
dysfunction as a consequence of initial impact, transient or persistent com-
pression, distraction, and laceration/ transection. Secondary SCI involves
progressive ischemic, inflammatory, and cytotoxic processes initiated or po-
tentiated by systemic and/or local insults. The prophylaxis and/or correction
of systemic insults such as hypotension, decreased oxygen content, and hy-
perthermia are the central goals of management to prevent secondary SCI.
Diagnosis
Clinical Assessment
The initial assessment of a patient suspected of SCI should be in accordance
with Advanced Trauma Life-Support recommendations, beginning with an
evaluation of airway, breathing, and circulatory function. Concomitant injuries
to other organ systems occur in 20% to 60% of patients suffering SCI.
The guiding principle for spinal assessment following trauma is to diagnose
neural element compression and spinal instability in the most accurate and
efficient manner possible. The American Spinal Injury Association (ASIA)
Classification System is the most widely used, rapid, and reproducible method
of neurologic classification of SCI (i.e., injury level and degree of impairment;
Figure 10.1). Strength is graded (0–5) in 10 bilateral motor segments (C5–T1
and L2–S1); a level is considered intact if the motor grade ≥3. The sensory
level is defined by testing 28 bilateral dermatome points, assessed for pin-
prick and light touch. Neurologic level of injury is defined as the most caudal
segment of the spinal cord with normal bilateral motor (≥3/5) and sensory
SECTION 1
(light touch and pinprick) function. The degree of impairment is categorized

as complete (e.g., no voluntary anal contraction, no S4–5 sensation, and no
deep anal pressure sensed) or incomplete (e.g., presence of sacral sparing).
Complete SCI involves the total loss of any motor, sensory, or sphincter
control below lesion and is often associated with neurogenic shock. Several
incomplete SCI syndromes have been described. Central cord syndrome
involves disproportionate weakness of the upper extremities (especially distal
98
Figure 10.1 International Standards for Neurological Classification of Spinal Cord Injury.
© 2011 American Spinal Injury Association.
motor groups) compared to the lower extremities, with signs of myelopathy
Traumatic Spinal Cord Injury

(usually urinary retention) and variable sensory disturbances (hyperpathia,
hypesthesia, etc.). Fifty percent recover the ability to ambulate; most recover
bladder control. The anterior cord syndrome involves dissociated sensory
loss (no pain or temperature sensation, preserved proprioception, vibratory
sense, and deep pressure) with either paraplegia or quadriplegia (if above C7).
Prognosis for recovery is poor. The posterior cord syndrome, which is rela-
tively rare, includes pain and paresthesias (burning sensations) of the neck,
upper arms, and torso with mild paresis of upper extremities. Hemisection of
the spinal cord (secondary to penetrating injury) results in the Brown-Sequard
syndrome with ipsilateral motor paralysis and loss of proprioception and vi-
bratory sense with contralateral loss of pain and temperature sensation. The
Chapter 10
conus medullaris syndrome consists of bilateral sacral sensory deficit (“saddle
anesthesia”), pronounced autonomic dysfunction (especially urinary reten-
tion), and symmetric paraparesis.
The neurologic exam may be confounded by spinal shock in the first 24 to
72 hours following traumatic injury. Spinal shock presents as transient loss
of all neurologic function below the level of injury demonstrated by flaccid
paralysis and areflexia, the resolution of which is often heralded by return of
anal-cutaneous and/or bulbocavernosus reflexes. This syndrome should not
to be confused with neurogenic shock, which is characterized by bradycardia,
99
hypotension, hypothermia, and priapism caused by loss of sympathetic tone.
Radiographic Assessment
Myriad specialty recommendations and practice guidelines have delineated
that all blunt trauma patients suffering clinical symptoms, neurologic deficits,
altered mental status, distracting injuries, or significant traumatic mechanisms
should undergo multidetector row computed tomographic (CT) scans with
axial collimation as the primary screening modality for acute cervical, thoracic,
and lumbar spinal injury. Magnetic resonance imaging (MRI) is the primary mo-
dality for characterizing traumatic, soft-tissue lesions of the spinal cord, inter-
vertebral disks, and spinal ligaments and is indicated for the evaluation of gross
neurologic deficits, CT findings suggestive of neurologic impingement, and/or
neurologic examination findings despite the absence of radiographic abnor-
malities. At this time plain and dynamic x-ray radiography for acute evaluation
of suspected of spinal injury should be discouraged as these modalities lack
the necessary sensitivity for screening.
Imaging modalities should be evaluated for characteristics suggestive of
spinal column stability. Lateral spine CT reconstructions should demonstrate
smooth, uninterrupted planes along the anterior vertebral body, posterior ver-
tebral body, and spinolaminar lines. Spinal canal diameter should exceed 13 mm
at every vertebral level. Instability is suggested by subluxation of >3.5 mm or
>20% listhesis, angulations exceeding 11o (cervical) or 20o (thoracolumbar),
and/or loss of >50% anterior compared with posterior vertebral body height.
Cervical Spine Clearance
Cervical collars must be removed as soon as it is safe to do so. They have

been associated with skin breakdown, increased intensive care unit (ICU)
stays, and elevated intracranial pressure. Cervical collars may be removed in
patients who are awake, alert, and without neurologic deficit or distracting
injury who have no neck pain/tenderness and full range of motion of the
cervical spine without the need for imaging. Meta-analysis of the current liter-
ature demonstrates that multislice helical CT alone is sufficient to detect un-
stable cervical spine injuries in trauma patients unable to be clinically cleared,
thus permitting the removal of the cervical collar from obtunded or intubated
SECTION 1
trauma patients if a modern CT is negative for acute injury.
Treatment
Patients with acute traumatic SCI should be managed at a Level 1 trauma center
to optimize outcomes. Patients suffering SCI, regardless of severity, frequently
experience cardiovascular instability and pulmonary insufficiency necessitating
vigilant monitoring. Management in an ICU has been shown to improve neuro-
logic outcome and reduce cardiopulmonary-related morbidity and mortality.
100
Respiratory Management
Ventilatory dysfunction correlates with spinal level and completeness of injury.
Cervical injuries (C2 through C6) may result in 80% to 95% reductions in vital
capacity, with absent or impaired cough. Low cervical and high thoracic cord
injuries (i.e., C7 through T4) may also lead to reduced vital capacity (30%–50%
of normal) and ineffective cough. Patients who may be able to initially com-
pensate for reduced ventilation can rapidly fatigue and progress to respiratory
arrest. Endotracheal intubation is frequently indicated in the setting of airway
compromise, respiratory failure (PaO2 <60 mm Hg or PaCO2 >60 mm Hg),
and/or associated severe TBI (GCS ≤8), with up to 30% of patients with
cervical SCI requiring intubation in the first 24 hours. Care should be taken
to avoid spinal movement during intubation. Options include awake fiber-
optic guidance or direct laryngoscopy with manual in-line spine stabilization.
Ventilation typically worsens between postinjury days two and five, followed
by gradual improvement; the mean duration of mechanical ventilation is
22 days in cervical SCI and 12 days in thoracic SCI. Management of respiratory
problems related to neuromuscular weakness after SCI should mirror res
piratory management of other neuromuscular conditions (see Chapter 13).
Establishing an ICU protocol to guide management of respiratory issues (e.g.,
gradual weaning, secretion clearance, cough assist, inspiratory force and vital
capacity measurements) after SCI is strongly recommended.
Pulmonary complications are the leading causes of death and morbidity in
the SCI population; atelectasis, pneumonia, aspiration pneumonitis, pulmonary
edema, and pulmonary embolism occur in >60% of patients with cervical and

upper thoracic SCI. Additionally, SCI patients frequently suffer from comorbid
traumatic injuries, further predisposing them to acute lung injury and acute
respiratory distress syndrome.
Management strategies advocate lung protective ventilation utilizing low
tidal volumes (6–8 mL/kg predicted body weight) with steps to enhance alve-
olar recruitment (positive end-expiratory pressure). Early conversion to tra-
cheostomy should be considered in patients anticipated to require more than
two weeks ventilatory support. Early tracheostomy has been associated with
better subjective tolerance, improved ventilation, reduced airway resistance,
shorter ventilator weaning, and shorter ICU stays.
Chapter 10
Cardiovascular Management
Management should be tailored to the etiology of hemodynamic disturbance
while optimizing spinal cord perfusion and avoiding injury to other organ
systems. Potential causes of hypotension include neurogenic shock, occult
hemorrhage, tension pneumothorax, myocardial injury or tamponade, and
sepsis. Hypotension immediately following acute SCI is most commonly due
to hemorrhage; however, neurogenic shock is an important etiology that
must be diligently sought and treated. Neurogenic shock (classically char-
acterized by hypotension, hypothermia, and bradycardia) occurs in up to
101
90% of patients suffering complete cervical SCI with lesions above the T6
neurological level (compared to 50% of those with incomplete SCI). Spinal
cord injury with interruption of the anterior interomedial tract results in
sympathetic denervation, thus leading to arteriolar dilation and hypotension
(SBP ≤80 mm Hg) with relative hypovolemia (venous pooling), as well as
unopposed parasympathetic drive (bradycardia and decreased contractility).
Hemodynamic instability secondary to neurogenic shock may be corrected
by judicious volume resuscitation, avoiding pulmonary edema, followed by
norepinephrine infusion (increases systemic vascular resistance and has ino-
tropic properties) as needed.
Optimal blood pressure management in patients with SCI is largely inferred
from data on cerebral autoregulation and perfusion pressure goals for treat-
ment of severe TBI. Class II and III evidence suggests hemodynamic augmen-
tation in SCI with a goal of maintaining mean arterial pressure >80–85 mm Hg
for at least seven days is safe and may be associated with improved neurologic
outcome in SCI. Hypotension should be aggressively avoided.
Venous Thromboembolism
Patients suffering SCI have the highest risk of venous thromboembolic disease
of any hospitalized patient population, with the risk being greatest during the
first three months. Untreated, 40% to 100% will develop deep venous throm-
bosis (DVT), while even with adequate prophylaxis (unfractionated heparin and
pneumatic compression stockings vs. low molecular weight heparin alone) 12 to
16% will develop major venous thromboemboli. Although the necessity of DVT/
PE prophylaxis has been established, the optimal treatment strategy remains

elusive. Systematic reviews and evidence-based consensus conferences have
recommended low molecular weight heparin alone, adjusted dose unfractionated
heparin, or unfractionated heparin combined with a nonpharmacologic device;
treatment should begin no later than 24 to 72 hours postinjury.
Other Critical Care Concerns
Infectious complications most frequently involve the respiratory or urinary
tract and are a leading cause of death and morbidity following SCI. Fever or
SECTION 1
leukocytosis not explained by an infectious source should prompt investiga-

tion of acute abdominal processes (e.g., pancreatitis, cholecystitis, bowel ob-
struction/ischemia/perforation) that may be occult with SCI.
Gastrointestinal stress ulceration is a known complication in trauma,
with SCI conferring independent risk. Either H2-blockers or proton pump
inhibitors are indicated and should be started at admission and continued for
at least four weeks.
Autonomic dysreflexia is characterized by symptoms of paroxysmal hyper-
tension, cutaneous flushing, blurred vision, and nausea in response to a stim-
ulus (hollow viscera distention or surgical procedures) below the level of the
cord lesion. If untreated, this may result in encephalopathy, seizures, stroke,
102
myocardial infarction, arrythmias, and death. Management priorities are re-

moval of stimulus and correction of hypertension.
Surgical Management
Surgical intervention plays a role in the management of spine trauma regard-
less of whether SCI exists. Surgery is performed with two goals in-mind: (a)
to decompress neural elements in those with neurologic deficit and (b) to
re-establish spinal alignment and stability in order to prevent further cord in-
jury and facilitate early mobilization. Early surgical intervention is not associ-
ated with increased complication rates and may improve neurologic outcome.
In the cervical spine, cord decompression may be performed through either
closed or open means. Closed reduction of cervical spinal fracture-dislocation
injuries with craniocervical traction is utilized to restore of anatomic alignment
of the cervical spine as a bridge to definitive surgical fixation.
Neuroprotective Strategies and Future Research
Research in SCI over the past several decades has elucidated many of the
mechanisms at play in secondary injury. Pharmacologic therapies investigated
to date in large multicenter prospective randomized controlled trials have in-
cluded methylprednisolone and the related compound tirilizad mesylate, GM-1
ganglioside, thyroid releasing hormone, gacyclidine, naloxone, and nimodipine.
None of these therapies has yet been shown in a randomized control trial to
definitively improve neurologic outcome.
Prospective, multicenter trials of methylprednisolone for neuroprotection

in SCI have demonstrated modest motor score benefits in secondary anal-
yses. However, subsequent studies and analyses have found significantly
higher rates of severe pneumonia, severe sepsis, and death. The American
Association of Neurological Surgeons/Congress of Neurological Surgeons
systematic review of this literature in 2002 concluded that “treatment with
methylprednisolone for either 24 or 48 hours is recommended as an option
in the treatment of patients with acute spinal cord injuries that should be un-
dertaken only with the knowledge that the evidence suggesting harmful side
effects is more consistent than any suggestion of clinical benefit.”
There is much ongoing clinical research in both pharmacologic as well
nonpharmacologic interventions for SCI. These include trials on the optimal
Chapter 10
timing of surgical decompression, therapeutic hypothermia, cerebrospinal
fluid drainage, cellular transplantation (e.g., Schwann cells, stem cells, etc.), as
well as targeting specific cellular inhibitors of regeneration.
Further Reading
Berlly M, Shem K. Respiratory management during the first five days after spinal cord
injury. J Spinal Cord Med. 2007;30(4):309–318.
103
Bracken MB, Shepard MJ, Collins WF Jr, et al. Methylprednisolone or naloxone treat-
ment after acute spinal cord injury: 1-year follow-up data. Results of the second
National Acute Spinal Cord Injury Study. J Neurosurg. 1992;76(1):23–31.
Bracken MB, Shepard MJ, Holford TR, et al. Methylprednisolone or tirilazad mesylate
administration after acute spinal cord injury: 1- year follow up. Results of the
third National Acute Spinal Cord Injury randomized controlled trial. J Neurosurg.
1998;89(5):699–706.
Casha S, Christie S. A systematic review of intensive cardiopulmonary management
after spinal cord injury. J Neurotrauma. 2011;28(8):1479–1495. PMCID: 3143388.
Chappell ET. Pharmacological therapy after acute cervical spinal cord injury. Neurosurgery.
2002;50(3 Suppl.):S63–72.
Chiodo AE, Scelza WM, Kirshblum SC, Wuermser LA, Ho CH, Priebe MM. Spinal cord
injury medicine. 5. Long-term medical issues and health maintenance. Arch Phys Med
Rehabil. 2007;88(3 Suppl. 1):S76–83.
Consortium for Spinal Cord Medicine. Respiratory management following spinal cord
injury: a clinical practice guideline for health-care professionals. J Spinal Cord Med.
2005;28(3):259–293.
Dhall SS, Hadley MN, Aarabi B, et al. Deep venous thrombosis and thromboembolism
in patients with cervical spinal cord injuries. Neurosurgery. 2002;50(3 Suppl.):S73–80.
Frankel HL, Coll JR, Charlifue SW, et al. Long-term survival in spinal cord injury: a fifty
year investigation. Spinal Cord. 1998;36(4):266–274.
Ganuza JR, Garcia Forcada A, Gambarrutta C, et al. Effect of technique and timing of
tracheostomy in patients with acute traumatic spinal cord injury undergoing mechan-
ical ventilation. J Spinal Cord Med. 2011;34(1):76–84.
Hadley MN, Walters BC, Grabb PA, et al. Management of acute spinal cord inju-
ries in an intensive care unit or other monitored setting. Neurosurgery. 2002;50(3

Suppl.):S51–57.
Hawryluk GW, Rowland J, Kwon BK, Fehlings MG. Protection and repair of the injured
spinal cord: a review of completed, ongoing, and planned clinical trials for acute spinal
cord injury. Neurosurg Focus. 2008;25(5):E14.
Hoffman JR, Mower WR, Wolfson AB, Todd KH, Zucker MI. Validity of a set of clinical
criteria to rule out injury to the cervical spine in patients with blunt trauma. National
Emergency X-Radiography Utilization Study Group. N Engl J Med. 2000;343(2):94–99.
Kirshblum SC, Priebe MM, Ho CH, Scelza WM, Chiodo AE, Wuermser LA. Spinal cord
injury medicine. 3. Rehabilitation phase after acute spinal cord injury. Arch Phys Med
SECTION 1
Rehabil. 2007;88(3 Suppl. 1):S62–70.

Krassioukov A, Warburton DE, Teasell R, Eng JJ. A systematic review of the man-
agement of autonomic dysreflexia after spinal cord injury. Arch Phys Med Rehabil.
2009;90(4):682–695.
Marino RJ, Barros T, Biering-Sorensen F, et al. International standards for neurological
classification of spinal cord injury. J Spinal Cord Med. 2003;26(Suppl. 1):S50–S56.
Panczykowski DM, Tomycz ND, Okonkwo DO. Comparative effectiveness of using
computed tomography alone to exclude cervical spine injuries in obtunded or
intubated patients: meta-analysis of 14,327 patients with blunt trauma. J Neurosurg.
2011;115(3):541–549.
Paralyzed Veterans of America, Consortium for Spinal Cord Medicine. Early Acute
104
Management in Adults with Spinal Cord Injury: A Clinical Practice Guideline for Health-
Care Providers. Washington, DC: Consortium for Spinal Cord Medicine; 2008.
Spinal Cord Injury Thromboprophylaxis Investigators. Prevention of venous thrombo-
embolism in the acute treatment phase after spinal cord injury: a randomized, mul-
ticenter trial comparing low-dose heparin plus intermittent pneumatic compression
with enoxaparin. J Trauma. 2003;54(6):1116–1124; discussion 25–26.
Rogers FB, Cipolle MD, Velmahos G, Rozycki G, Luchette FA. Practice management
guidelines for the prevention of venous thromboembolism in trauma patients: the
EAST practice management guidelines work group. J Trauma. 2002;53(1):142–164.
Wilson JR, Fehlings MG. Emerging approaches to the surgical management of acute
traumatic spinal cord injury. Neurotherapeutics. 2011;8(2):187–194.
Wuermser LA, Ho CH, Chiodo AE, Priebe MM, Kirshblum SC, Scelza WM. Spinal cord
injury medicine. 2. Acute care management of traumatic and nontraumatic injury.
Arch Phys Med Rehabil. 2007;88(3 Suppl. 1):S55–61.
Chapter 11
Meningitis and Encephalitis

Ruchira Jha
Meningitis
Meningitis can have multiple etiologies classified as either infectious (bacte-
rial, viral, fungal, tick-borne) or noninfectious (neoplastic, drug-related). This
chapter outlines key features in the epidemiology, diagnosis, and management
of common causes of meningitides. Key points are summarized in Table 11.1.
Bacterial Meningitis
Epidemiology
Acute bacterial meningitis (ABM) typically presents within hours and requires
early recognition and aggressive management including antibiotics, steroids if
105
indicated, isolation, and prophylaxis of contacts when necessary. Mortality
in adults remains up to 20% depending on the organism and immune con-
stitution of the patient. Fortunately, only 10% of diagnosed meningitis cases
have a bacterial origin. Five pathogens are responsible for 80% of ABM
cases: Streptococcus pneumoniae, Neisseria meningitidis, Haemophilus influenzae,
Group B streptococcus, and Listeria monocytogenes. The H. influenzae type B,
pneumococcal, and meningococcal vaccines have dramatically reduced the
global incidence of meningitis. Several studies have indicated a potential ge-
netic predisposition to ABM (particularly related to complement deficiencies)
however these details are beyond the scope of this chapter. In immunocom-
promised intensive care unit patients, aerobic gram-negative bacilli like the
Enterobacteriaceae family are also important pathogens. Other risk factors
such as sinusitis, endocarditis, penetrating trauma, neurosurgical procedures,
or nosocomial infections further broaden the potential spectrum of causative
microbes (Table 11.1).
Diagnosis and Presentation
The clinical presentation of ABM classically includes two of the four findings
of headache, fever (77%), nuchal rigidity (83%), or altered mental status (69%)
in about 95% of patients. Seizures occur in approximately 20% of patients.
Lethargy is also common, most often in pneumococcal meningitis. The need
for neuroimaging prior to lumbar puncture (LP) is controversial, as hernia-
tion related to ABM is rare. Practice guidelines recommend computed to-
mography (CT) prior to LP in patients with an immunocompromised state,
Table 11.1 Meningitis Evaluation and Treatment
A: Bacterial Meningitis
• Typical CSF Profile: OP ↑↑ >20 cmH2O, WBC ↑↑ 250–100,000/mm3 (polymorphonuclear predominance), Protein ↑↑ 100–500 mg/dL, Glucose ↓↓ <40 mg/dL,
CSF-blood glucose ratio <0.4
• Empiric Treatment: Vancomycin + Ceftriaxone; Dexamethasone 10 mg Q6H × 4 days with first dose prior to antibiotics unless there is a contraindication
• Special Populations: Add Ampicillin to above regimen if pregnant, immunocompromised, or >50 years
Organism Risk Factors Treatment
Streptococcus pneumoniae Asplenia, alcoholism, complement • Options based on sensitivities: Penicillin G, Ampicillin, Ceftriaxone, or
deficiency Vancomycin + Ceftriaxone
Neisseria meningitidis Multiperson dwelling (military, college) • Options based on sensitivities Penicillin G, Ampicillin, or Ceftriaxone
Haemophilus influenza Newborns–adults • B-lactamase negative: Ampicillin
• B-lactamase positive: Ceftriaxone
Listeria monocytogenes Immunodeficient (alcoholism, HIV, • Penicillin G or Ampicillin
steroids, newborns)
Staphylococcus aureus Nosocomial, penetrating trauma, • Methicillin-sensitive: Nafcillin or Oxacillin
recent neurosurgical procedure • Methicillin-resistant: Vancomycin
Gram negative bacilli (Escherichia Nosocomial (33% of cases), sinusitis • Pseudomonal species: Ceftazidime or Cefepime
coli, Klebsiella, Pseudomonas species, • Non-pseudomonal: Third-generation cephalosporin
Enterobacteriacae)
B: Viral Meningitis
• Typical CSF Profile: OP normal, WBC ↑ 100–1,000/mm3 (lymphocytic predominance), Protein ↑ 50–250 mg/dL, Glucose normal
Organism Diagnosis Treatment
Enterovirus CSF PCR • Supportive, IVIG if patient has hypogammaglobulinemia
HSV-2 CSF PCR • If severe and recurrent Acyclovir
• If frequent episodes, prophylaxis with Valacyclovir
VZV CSF IgM and IgG • Supportive, IV Acyclovir if immunocompromised
HIV Serum diagnosis • Supportive, initiate antiretrovirals
WNV CSF IgM • Supportive
C: Fungal Meningitis
• Typical CSF Profile: OP ↑↑ >20 cmH2O, WBC ↑↑ 100–500/mm3, Protein ↑ 100–500 mg/dL, Glucose ↓ <30–45 mg/dL CSF-blood glucose ratio <0.4
Cryptococcus CSF cryptococcal Ag + CSF India ink • Induction: Liposomal Amphotericin B + flucytosine
+ CSF Cx • Consolidation: Fluconazole
• Maintenance: Fluconazole for life
• Serial LP until opening pressure normal
• Antiretroviral therapy
Coccidioides Complement fixation antibody titer • Intrathecal amphotericin
• Fluconazole for life
Histoplasma Complement fixation antibody • Liposomal Amphotericin B then Itracnazole
Aspergillus CSF PCR, CSF antigen galactomannan, • Voriconazole
antibody
D: Other
Mycobacteria Acid fast bacillus (se 25%), culture (se • Initial regimen (2 months): Isoniazid, Rifampin, Ethambutol, or
50%–80%), PCR (se 50%–100%, sp Streptomycin, Pyrazinamide
100%) • Subsequent 10 months: Isoniazid
Lyme (Borrelia burgdoferi) Serologic (IgM and IgG)—CSF PCR and • Ceftriaxone or Doxycycline
culture low se and sp
CSF = cerebrospinal fluid; OP = opening pressure; WBC = white blood count; PCR = polymerase chain reaction; IVIG = intravenous immunoglobulin; IV = intravenous; LP = lumbar
puncture; se = sensitivity; sp = specificity.
history of central nervous system disease, seizure, focal neurological deficits,
or depression in mental status; otherwise patients are considered safe for LP

prior to CT. The cerebrospinal fluid (CSF) profile in ABM is outlined in Table
11.1. Independent CSF predictors of ABM with >99% certainty include glu-
cose <34 mg/dL, CSF:serum glucose ratio <0.23, protein >220 mg/dL, total
pleiocytosis >2000 cells/uL, neutrophils >1180 cells/uL. Importantly, these
findings are not significantly changed by antibiotic pretreatment compared to
the gram stain yield which is reduced by 16%. Given the reduction in diagnostic
yield of CSF gram stain and culture after antimicrobial treatment, molecular
methods are becoming increasingly important to diagnose ABM and other
SECTION 1
CNS infections including PCR (that can detect organisms for several days
after antibiotics), multiplex PCR, 16S PCR, MALDI-TOF, and whole genome
sequencing. The details of these technologies are beyond the scope of this
chapter since they are not currently standard of care, but are important to be
aware of since they may soon revolutionalize clinical microbiology in this field.
Management
Prompt diagnosis and management of ABM is essential to decreasing the
morbidity and mortality of ABM. Treatment should not be delayed to obtain
imaging or LP. Empiric treatment should be initiated immediately upon suspi-
cion for ABM based on the patient’s risk for specific organisms (Table 11.1).
The classic regimen includes vancomycin and a third-generation cephalosporin
108
with good central nervous system (CNS) penetration such as ceftriaxone (see
Table 11.1 for doses). In patients who are above 50 years old, pregnant, or im-
munocompromised, ampicillin should be added. Those with penetrating head
trauma or neurosurgery are at risk for pseudomonal infections, and third-
generation cephalosporins should be replaced by cefepime or ceftazidime.
Unless there are contraindications to steroids, patients with suspected ABM
should receive 10 mg dexamethasone every six hours for four days, with the
first dose administered prior to antibiotics. Once culture data are available,
antibiotics should be narrowed appropriately (Table 11.1).
Viral Meningitis
Epidemiology
Most cases of meningitis are viral. Enteroviruses are the most common of-
fender, causing approximately 60% of all cases. Transmission is usually fecal-
oral. Herpesviruses (such as herpes simplex virus [HSV-2]), Varicella Zoster
virus (VZV), and arboviruses (like West Nile virus [WNV]) are also neuro-
tropic and can cause meningitis, though HSV-1 typically causes an aggressive
encephalitis. HSV-2 can cause meningitis at the time of initial genital infection.
Arboviruses such as WNV typically cause disease in the mid-to late summer
months. Most cases of WNV are asymptomatic or cause a mild febrile ill-
ness; only 1% develop neuroinvasive disease such as meningitis, encephalitis,
or a flaccid paralysis. Acute HIV may cause meningitis in 5% to 10% of newly
infected patients.

Viral meningitis commonly presents with fever, severe headache, photo-
phobia, and nausea. Altered level of consciousness is rare. There may be evi-
dence of a viral exanthema in enteroviruses and occasionally also with WNV.
The CSF profile shows a normal opening pressure, mild leukocytosis (100–
1000/mm3), lymphocytic predominance (although neutrophils may prevail
early in the course in approximately 40% of patients), mildly elevated protein
(50–250 mg/dL), and a normal glucose. CSF evaluations for specific viral eti-
ologies are summarized in Table 11.1. HIV-related acute meningitis can be di-
agnosed by CSF polymerase chain reaction (PCR) since patients may not have
seroconverted at the time of meningitis presentation.
Chapter 11
Management
Management of viral meningitis is predominantly supportive. Enteroviral in-
fection is usually mild and self-limited. HSV-1 typically causes an encephalitis
and is treated emergently with intravenous (IV) acyclovir (see section on
encephalitis). HSV-2 commonly causes acute meningitis and is the most
common cause of recurrent viral meningitis. Like most viral meningitides,
the course is generally self-limited; however, severe recurrent episodes can
be treated with IV acyclovir. If episodes are particularly frequent, then inter-
mittent prophylaxis can be considered with oral valacyclovir. VZV-related
109
meningitis has become increasingly recognized; however, CSF VZV may be
positive in the absence of clinical symptoms. Immunocompromised patients
are treated with IV acyclovir (Table 11.1).
Fungal Meningitis
Most fungal meningitides are subacute to chronic and present in immuno-
compromised hosts. The following section focuses on four relatively common
fungal infections: Cryptococcus neoformans, Coccidioides immitis, Histoplasma
capsulatum, and Aspergillus fumigatus.
Epidemiology
Cryptococcus neoformans has four serotypes, with type A (C. neoformans grubbi)
being the most common. Cryptococcal meningitis has a prevalence of approx-
imately 10% in the United States and is as high as 30% in sub-Saharan Africa
and is most common in patients with cell-mediated immunocompromise.
Coccidioides immitis is endemic to southwestern United States and typ-
ically results in a self-limited respiratory infection (valley fever). Patients at
risk for disseminated infection including cavitary pneumonia, osteomyelitis,
and meningitis include pregnant women, those on long-term immunosuppres-
sive therapy, and those with HIV. Histoplasma capsulatum is endemic to the
Ohio and Mississippi river valleys. It has also been related to exposure to bird
excrement and bat guano. Patients often have an acute pulmonary infection.
Immunocompromised patients have disseminated disease including meningitis.
Aspergillus fumigatus is an increasing cause of infection related to immuno-
suppression. Patients susceptible to invasive aspergillosis can be neutropenic,
have defects in phagocyte function, or have decreased cell-mediated im-
munity. CNS infection is typically from direct extension (e.g., sinusitis) or

hematogeneous spread.
In general, the CSF profile of fungal meningitides includes elevated opening
pressure, moderate pleiocytosis (100–500/mm3) with a lymphocytic predom-
inance, elevated protein (100–500 mg/dL), and decreased glucose (20–40
mg/dL with a CSF to blood glucose ratio of <0.4–0.6). Specific presentations
and diagnostic tests for various organisms are discussed next.
Cryptococcus neoformans meningitis typically occurs in HIV patients with
SECTION 1
CD4+ counts <50/uL. Symptoms in immunocompromised patients may be

vague, indolent, and nonspecific: 75% have headache and fever developing
over two to four weeks, 50% have altered mental status, nausea, and vomiting.
Immunocompetent hosts present with typical signs of meningitis, and approxi-
mately 20% develop signs of elevated intracranial pressure. Focal neurological
deficits are due to mass effect from cryptococcomas. Radiographic imaging
(CT or magnetic resonance imaging [MRI]) may show cerebral edema, hydro-
cephalus, leptomeningeal enhancement, or cryptococcomas. CSF findings are
typical for fungal meningitides. India ink stain is 75% sensitive. Cryptococcal
capsular polysaccharide antigen assay is sensitive and specific and provides a
quantitative titer. CSF fungal culture can also be useful with reported sensitiv-
110
ities from 75% to 95%.

Coccidioides immitis meningitis has a predilection for the basilar meninges and
presents with headache (75%), altered mental status (39%–73%), focal neu-
rologic deficits including facial palsies and diplopia, and neck stiffness (20%).
Patients with HIV also typically have a reticulonodular infiltrative pneumonia.
Patients may develop hydrocephalus (20%– 50%), vasculitis, and infarction
including dural venous thromboses. In addition to parenchymal lesions or
parameningeal abscesses, MRI may also demonstrate meningeal enhancement
or spinal arachnoiditis. The definitive diagnosis of C. immitis meningitis involves
CSF culture or identification using complement-fixation antibody assay. Of
note, antibody tests may be negative early in the course of infection. PCR is not
usually available. CSF eosinophilia is noted in approximately 70% of patients.
Serologic tests or identification of the organisms from other body fluids such
as sputum or broncheoalveolar lavage (BAL) fluid can also aid in the diagnosis.
Histoplasma capsulatum meningitis occurs in less than 20% of patients
infected with this organism and most often affects those with cell-mediated
immunocompromise. Patients present with headache, altered mental status,
and occasionally with focal neurologic deficits, seizures, stroke, and mass
lesions. History and identification of risk factors are crucial to the diagnosis.
CSF findings are typical for fungal meningitis but also have mononuclear
pleiocytosis. CSF culture is not sensitive, and complement fixation antibody
assays can be used. Serologic or bone marrow identification of the organism
are helpful adjunctive tests.
Aspergillus fumigatus infection of the CNS can have multiple

presentations: solitary aspergillomas, cavernous sinus thrombosis, multiple
abscesses, acute or chronic meningitides (typically basilar), or aggressive vas-
culitis that can result in infarction, aneurysm formation, or hemorrhage. These
findings are often confirmed on MRI, which can also show dural enhance-
ment adjacent to the infected paranasal sinuses indicative of direct extension.
Diagnosis is often supported by culture from extra-CNS sites of infection such
as a BAL or blood culture. PCR can detect aspergillus in blood and BAL; how-
ever, it is limited in CSF unless it is used serially. Serial CSF PCR may also be
useful in assessing treatment response.
Management
Chapter 11
Cryptococcus neoformans: The treatment is divided into three stages (Table
11.1): (a) induction with liposomal amphotericin B and flucytosine for 14 days,
(b) consolidation with fluconazole for 10 weeks, and (c) maintenance with
lifelong lower dose fluconazole. During the induction stage, it is also impor-
tant to perform serial daily LPs if the opening pressure is above 25 cm H2O.
This practice should be continued until the opening pressure is normal, and
some patients may require a temporary lumbar drain or ventriculostomy. In
HIV patients, antiretroviral therapy should be initiated. There is no role for
steroids. Mortality has improved with treatment but remains 10% to 25% even
in developed countries.
111
For Coccidioides immitis, induction therapy with oral fluconazole is required
followed by lifetime maintenance therapy. IV combined with intrathecal am-
photericin B is reserved for those who fail oral azole therapy, pregnant patients,
those with complicated infections, and those who are intolerant of azole
treatment. Most patients (66%–80%) respond to azole treatment, but due to
the high relapse rate, lifelong prophylaxis with fluconazole is recommended.
Mortality without treatment is extremely high.
Histoplasma capsulatum meningitis has a high mortality (up to 40%) despite
treatment, and even those who initially respond to therapy relapse after dis-
continuation. Treatment involves IV liposomal amphotericin B with eventual
transition to itraconazole or fluconazole maintenance therapy for the subse-
quent 9 to 12 months.
Aspergillus fumigatus infection of the CNS has a high mortality despite
treatment. Data suggest that treatment with voriconazole is more effective
than amphotericin B. Salvage therapy in patients with invasive disease may
include combining voriconazole with caspofungin. In patients who have single
CNS aspergillomas that do not involve eloquent regions of the brain, surgical
debulking combined with systemic antifungals has been shown to be curative.
Other Infectious Meningitides
While there are numerous additional infectious etiologies for meningitis,
this section focuses on two common organisms: Borrelia burgdoferi and
Mycobacterium tuberculosis.
Epidemiology
Borrelia burgdoferi affects the CNS in 10% to 15% of patients with Lyme disease,
two-thirds of whom have lymphocytic meningitis. Other CNS manifestations
include cranial neuropathies or radiculitis. Meningitis typically occurs within the
first few months of initial infection.
Mycobacterium tuberculosis causes meningitis when a caseating CNS
granuloma expels its contents into the subarachnoid space. Tubercular
meningitis is aggressive and endemic to countries such as India, China, and
South Africa with an increasing incidence since the advent of HIV infection.
Immunocompromised patients are more likely to have extrapulmonary infec-
SECTION 1
tion such as meningitis.

Borrelia burgdoferi, or Lyme meningitis, has a similar clinical presentation to
viral meningitis. CSF findings include a mild pleiocytosis (lymphocytic predomi-
nance), normal glucose, and mildly elevated protein (Table 11.1). CSF may also
be positive for oligoclonal bands. CSF PCR and culture have low sensitivities
and specificities, so diagnosis is primarily dependent on the history (tick ex-
posure, rash of erythema migrans) and demonstration of serologic response
to the organism combined with clinical symptoms of meningitis. Treatment is
initiated if patients are either IgM positive (if symptoms began <30 days prior
112
to presentation) or IgG positive (if symptoms have persisted for >30 days).
For Mycobacterium tuberculosis, immunocompetent patients typically pre-
sent with headache (96%), meningismus (79%), cranial nerve deficits, and signs
of raised intracranial pressure (decreased level of consciousness, vision loss,
vomiting). Acute deep territory stroke from vascular thrombosis can also be
a common presentation. M. tuberculosis has a propensity for the basilar me-
ninges and can produce a thick exudate that blocks CSF flow, causing hydro-
cephalus. CSF abnormalities (Table 11.1) show a mononuclear pleiocytosis,
low glucose, and high protein. The acid-fast bacillus smear is positive in only
5% to 30% of cases. CSF culture takes weeks to process and is positive in 45%
to 90% of infected patients. CSF PCR has a varying sensitivity (30%–50%);
however, it has 98% specificity. Frequently, the diagnosis may be made with a
combination of evidence of systemic tuberculosis, clinical CNS disease, and
response to empiric treatment.
Management
Borrelia burgdoferi is sensitive to many common antibiotics with good CNS
penetration. Treatment for patients with parenchymal CNS disease or menin-
gitis involves IV ceftriaxone or penicillin G or oral doxycycline for two to four
weeks. These treatments are typically curative.
Mycobacterium tuberculosis has a high mortality without prompt treatment.
First line therapy includes two months of rifampicin, isoniazid, pyrazinamide,
and ethambutol or streptomycin followed by 9 to 12 months of isoniazid +/–
rifampicin. Multidrug-resistant tuberculosis may require prolonged intensive
treatment and potentially additional medications depending on the sensitivities.

There are data suggesting that corticosteroids reduce morbidity and mortality.
Ventriculoperitoneal shunting may be required for managing hydrocephalus.
Noninfectious Meningitides
While not the focus of this chapter, it is important to note that meningitis
can have noninfectious etiologies including medications, neoplasms, and
systemic inflammatory diseases. Common drug offenders include nonste-
roidal anti-inflammatory drugs, antibiotics (particularly penicillins, trimetho-
prim/sulfamethoxazole), and immunosuppressants, and symptoms are like
viral meningitis. Leptomeningeal carcinomatosis (carcinomatous meningitis) is
commonly associated with hematologic malignancies; however, it can occur in
Chapter 11
patients with solid tumors.
Encephalitis
Encephalitis refers to inflammation of the brain parenchyma, and it can have
multiple etiologies including various infectious (viral, bacterial, fungal) or
noninfectious causes (postinfectious and autoimmune or paraneoplastic).
The classic presentation includes headache, fever, and altered mental status.
113
Patients may also have focal neurologic signs or seizures.
Viral Encephalitis
There are approximately 20,000 cases of viral encephalitis in the United States
annually. This section focuses on the three of the most common causes: HSV-
1, VZV, and WNV. Table 11.1 summarizes key findings in other common viral
encephalitides.
Epidemiology
HSV-1 is the most common cause of sporadic viral encephalitis in the United
States affecting 1/250,000 to 1/500,000 patients per year and representing
10% to 20% of all viral encephalitis cases. It has no predilection for a season,
age, gender, or immune status. Although the pathogenesis is not clearly de-
fined, it is thought that HSV encephalitis occurs due to spread from the trigem-
inal ganglia to the meninges over the temporal lobes or following reactivation
of latent virus within the brain.
VZV is another member of the herpesviridae family that has been increasingly
recognized as a cause of CNS syndromes including meningitis, encephalitis,
and myelitis. CNS involvement is likely due to neuronal reactivation of latent
virus in the trigeminal or other ganglia. Unlike HSV1, it commonly affects im-
munocompromised and elderly patients.
WNV is a single-stranded RNA virus of the flaviviridae family that is now
the leading cause of epidemic viral encephalitis in the United States. It causes
neuroinvasive disease in the form of encephalitis, meningitis, or acute flaccid
paralysis. Most adults are asymptomatic, but 20% will develop a flu-like illness
(malaise, fatigue, fever, nonspecific rash, headache) lasting for seven days, and
only 1% develop neuroinvasive disease. Risk of encephalitis increases signifi-

cantly with age or an immunocompromised state.
HSV-1 encephalitis can mimic ABM with rapid onset, severe alteration in
mental status (confusion to coma), and high fever. This virus has a predilec-
tion for temporal and orbitofrontal cortex thereby predisposing patients to
seizures, behavioral changes/ personality alterations, and aphasias. MRI is
often useful and may show T2 hyperintensity in the affected cortex (most
commonly the temporal lobe) that will enhance with gadolinium. There may
SECTION 1
also be associated diffusion restriction. It is typically unilateral but can be bi-

lateral. Advanced infection results in hemorrhagic destruction of the involved
cortex. In patients who cannot undergo MRI, CT with electroencephalography
(EEG) can be useful. Characteristic EEG findings demonstrate lateralized peri
odic discharges. CSF findings can vary. Pleiocytosis can range from 10/mm3 to
2000/mm3 with a lymphocytic predominance; however, a neutrophil predom-
inance may occur early in the course. Xanthochromia may be present due to
the hemorrhagic changes. Protein is mildly to moderately elevated, and glu-
cose is generally normal but may be slightly low. The diagnostic test of choice
is CSF PCR, which has a high sensitivity (96%) and specificity (99%). In some
patients, PCR may be negative at presentation and should be repeated in four
114
to seven days if there is a high clinical suspicion.

VZV encephalitis is frequently associated with a large and small vessel cere-
bral vasculitis. Patients may develop multifocal ischemic injury, hemorrhage,
aneurysm formation, and subarachnoid hemorrhage. Symptoms are more
severe in immunosuppressed patients. Often encephalitis follows a zoster
outbreak. Unlike many viral encephalitides, CSF in VZV vasculitis may have
a mononuclear pleiocytosis with low glucose. Oligoclonal bands may be pre-
sent. Diagnosis is by CSF IgG and IgM since PCR has a lower sensitivity and
does not exclude VZV.
WNV encephalitis typically presents with the triad of fever, headache, and
altered mental status (usually stupor or coma). The virus has a predilection for
the basal ganglia, thalamus, and substantia nigra; thus patients may also have
tremor, myoclonus, parkinsonian symptoms, or cerebellar ataxia. Patients may
also develop weakness (including respiratory failure), due to invasion of the
anterior horn cells, and hyponatremia. MRI may show T2 hyperintensity in the
basal ganglia or thalamus. The diagnosis is made by demonstrating CSF IgM
antibodies specific to the virus.
Management
HSV-1:
Mortality from HSV encephalitis has decreased significantly since the intro-
duction of IV acyclovir. Standard treatment is with IV acyclovir (Table 11.2)
for three weeks. Treatment should not be delayed if there is any clinical sus-
picion of HSV encephalitis. Despite adequate treatment, approximately 30%
Table 11.2 Encephalitis Evaluation and Management
A: Viral Encephalitis
Herpesviridae
HSV-1 CSF PCR Acyclovir
CMV CSF PCR Ganciclovir; Foscarnet
EBV CSF PCR Supportive
HHV6 CSF PCR Treatment if immunocompromised: Ganciclovir
Foscarnet
VZV CSF IgG and IgM Acyclovir
Flaviviridae
WNV CSF IgM Supportive
St. Louis encephalitis CSF IgM acute IgG Supportive
Paramyxoviridae
Measles Serology Supportive (preventative with vaccine)
Mumps CSF IgM, culture, acute IgG Supportive (preventative with vaccine)
Picornaviridae
Enterovirus CSF PCR Supportive (Pleconaril, not available in United States)
Retrovirus
HIV CSF PCR Supportive
B: Nonviral Infectious Encephalitis
Listeria monocytogenes CSF Culture Ampicillin
Mycoplasma pneumoniae Serum IgM, PCR low yield Tetracycline, Doxycycline, Erythromycin
Naegleria fowleri CSF trophozoites wet mount None
Rickettsia rickettsia (Rocky Mountain Spotted Fever) Acute serology Tetracyclines (Doxycyline) or Chloramphenicol
CFS = cerebrospinal fluid; PCR = polymerase chain reaction.
of patients have permanent neurologic deficit including seizures, impaired
memory or cognition, speech difficulties, or other focal deficits. The prognosis

is worse in patients who have a decreased level of consciousness, advanced
age, evidence of extensive disease on initial MRI or CT, or delayed initiation
of treatment.
VZV: Encephalitis is treated with IV acyclovir for a minimum of 14 days.
Oral prednisone (1 mg/kg) for five days can be considered for the treatment
of vasculitis.
WNV: Treatment is supportive. The prognosis is variable. However, re-
covery is extremely prolonged, and many patients do not return to premorbid
SECTION 1
function with complaints of persistent headache, fatigue, and myalgia being

common.
Other Infectious Etiologies of Encephalitis
Epidemiology
Nonviral infectious etiologies of encephalitis are less common and include
multiple organisms such as Bartonella henselae, Listeria monocytogenes,
Mycoplasma pneumonia, Coxiella burnetti, Naegleria fowleri, Ricketsia ricketsii,
and Toxoplasma gondii. Specific features are outlined in Table 11.2. Of these
organisms, Listeria monocytogenes, Mycoplasma pneumoniae, and Ricketsia
ricketsii are the most common. Listeria tends to affect neonates, adults over
116
the age of 50, and patients with impaired cell mediated immunity. Naegleria
fowleri is associated with fresh water swimming. Rocky mountain spotted
fever (RMSF) is caused by Rickettsia , which is endemic in the central and
southeastern regions of the United States, and this tick-borne encephalitis
tends to occur between April and September. Mycoplasma does not have
a geographic distribution, seasonal incidence, or predilection for specific
populations.
Diagnosis and prevention vary depending on the specific agent involved.
Listeria typically causes a rhombencephalitis but can also result in a cor-
tical encephalitis that resembles HSV-1. The diagnosis is made by CSF
culture. Mycoplasma can be associated with a postinfectious encephalitis
such as acute disseminated encephalomyelitis but can also cause an acute
infectious encephalitis or encephalomyelitis. The diagnosis is made by
serum IgM and IgG since CSF PCR is typically low yield. RMSF causes a
fever, headache, altered mental status, seizures, and a classic centripetal
maculopapular rash (beginning on the wrists and ankles) and is diagnosed
using acute serology.
Management
As summarized in Table 11.2, treatment for Mycoplasma is tetracycline, doxy-
cycline, or erythromycin. Listeria is treated with IV ampicillin, and Rickettsia
is treated with either tetracyclines or chloramphenicol. There is no effective
treatment for Naegleria, and meningoencephalitis rapidly progresses to coma
and death. Miltefosine is a potential investigational drug but is not clinically

available.
Immune-Mediated Encephalitis
There is increasing recognition of immune-mediated encephalitis as impor-
tant etiologies of CNS inflammation. Antibody-mediated encephalitides can
be paraneoplastic, idiopathic autoimmune, or postinfectious autoimmune.
Diagnosis relies on serum and CSF testing, imaging, and EEG findings, and
treatment varies based on the underlying etiology as discussed in a different
chapter in this handbook.
Chapter 11
Intracranial Abscesses and Subdural
Empyemas
Intracranial Abscess
Intracranial abscesses are focal infections of the brain parenchyma. The
manifesting symptoms depend on the location of the abscess. In addition
to focal neurologic deficits, they can result in elevated intracranial pressure
or seizures which may also require appropriate diagnosis and treatment.
Predisposing risk factors include trauma, hematogenous spread (influenced by
117
the patient’s age and immune status, endocarditis), meningitis, or contiguous
source of infection (sinusitis, otitis). Abscesses can also occur in the epidural
space. Abscess diagnosis is initially suggested by MRI with gadolinium. Most
commonly isolated organisms tend to be anaerobic, Staphylococcus aureus,
Enterobacteriaceae, and Streptococci. Treatment of encapsulated abscesses
requires a combination of surgical excision or drainage and prolonged antibi-
otic therapy typically for six to eight weeks tailored to the isolated pathogen.
Subdural Empyema
Subdural empyemas (infections between the dura mater and arachnoid)
are less common—they share risk factors with abscesses and in particular
may occur as complications of local infections, meningitis, or neurosurgical
procedures/external instrumentation. Diagnosis and management is similar to
that of intracranial abscesses.
Further Reading
Blair JE. Coccidioidal meningitis: update on epidemiology, clinical features, diagnosis, and
management. Curr Infect Dis Rep. 2009;11(4):289–295.
Cho TA, Mckendall RR. Clinical approach to the syndromes of viral encephalitis, mye-
litis, and meningitis. Handb Clin Neurol. 2014;123:89–121.
de Gans J, van de Beek D. Dexamethasone in adults with bacterial meningitis. N Engl J
Med. 2002;347(20):1549–1556.
Dubos F, Martinot A, Gendrel D, Bréart G, Chalumeau M. Clinical decision rules for
evaluating meningitis in children. Curr Opin Neurol. 2009;22(3):288–293.

Greenlee JE. Encephalitis and postinfectious encephalitis. 2012;18(6 Infectious Disease):
1271–1289.
Halperin JJ. Lyme disease: a multisystem infection that affects the nervous system.
Continuum. 2012;18(6 Infect Dis):1338–1350.
Hasbun R, Abrahams J, Jekel J. Computed Tomography of the Head before Lumbar
Puncture in Adults with Suspected Meningitis. N Engl J Med. 2001;345:1727–1733.
McGill F, Heyderman RS, Panagiotou S, Tukel AR, Solomon T. Acute bacterial meningitis
in adults. Lancet. 2016;388:3036–3047.
SECTION 1
Prasad K, Singh MB. Corticosteroids for managing tuberculous meningitis. Cochrane

Database Syst Rev. 2008;(1):CD002244.
Sejvar JJ, Haddad MB, Tierney BC, et al. Neurologic manifestations and outcome of
West Nile virus infection. JAMA. 2003;290(4):511–515.
Spanos A, Harrell FE, Durack DT. Differential diagnosis of acute meningitis: an analysis
of the predictive value of initial observations. JAMA. 1989;262(19):2700–2707.
Tunkel AR, Glaser CA, Bloch KC, et al. The management of encephalitis: clinical
practice guidelines by the Infectious Diseases Society of America. Clin Infect Dis.
2008;47(3):303–327.
Tunkel AR, Hartman BJ, Kaplan SL, et al. Practice guidelines for the management of bac-
terial meningitis. Clin infect Dis. 2004;39:1267–1284
118
Wheat J, Sarosi G, McKinsey D, et al. Practice guidelines for the management of

patients with histoplasmosis. Infectious Diseases Society of America. Clin Infect Dis.
2000;30(4):688–695.
Van de Beek D, de Gans J, Spanjaard L. Clinical Features and Prognostic Factors in Adults
with Bacterial Meningitis N Engl J Med. 2004;35:1849–1859.
Chapter 12
Inflammation and
Demyelination
Daniel B. Rubin and Henrikas Vaitkevicius
Introduction
This chapter focuses on autoimmune and noninfectious inflammatory
disorders of the central nervous system (CNS) encountered in the neurolog
ical intensive care unit. This category of diseases encompasses both the ful
minant presentations of acute and previously undiagnosed syndromes as
well as the severe exacerbations of chronic conditions. Traditionally, these
disorders are reviewed as a list of diseases with corresponding diagnostic
119
guidelines and treatment options. However, this approach is less relevant in
fulminant disorders when a comprehensive diagnostic workup cannot fea
sibly be completed prior to the need for intervention. It is also not infrequent
that neurointensivists encounter patients with disease processes that, despite
having the clinical, laboratory, and radiographic markers of autoimmunity,
elude formal diagnosis.
In recent years, however, research into the basic immunological processes
underlying many of these syndromes has revealed several common patho
physiologic mechanisms. As such, many of these disorders can be classi
fied by their common underlying mode of immune dysfunction, and clinical
syndromes that elude diagnosis can at the very least be characterized immu
nologically in an effort to guide a rational approach to empiric therapy. As our
understanding of the basic pathologic mechanisms underlying autoimmunity
continues to grow, it is very likely that this approach will become of increasing
importance in the treatment of acute neuroinflammatory disorders. In the
present, this approach will hopefully be of help in guiding treatment decisions
in these otherwise seemingly “data-free” zones of critical care neurology. For
this reason, we organize our discussion of these disorders on the pathophysi
ology of the underlying immune dysfunction, workup strategies, and available
interventions. While there is an attempt to cluster these groups of disorders
based on immunological processes involved, we do recognize that frequently
multiple immunological pathways are implicated and multiple classes of
interventions may be effective. Treatment strategies focus on the spectrum
of potential risk/benefit ratios for currently available immunologic therapies.
Diagnostic Strategy
The differential diagnosis of new onset CNS dysfunction in the neurological
intensive care unit is almost always first described generically as infection,
neoplasia, toxic/metabolic, or primary vascular. Autoimmune disorders are
suspected when the disease course is subacute (<3 months), there is evidence
of inflammation, and toxic/metabolic, infectious, and primary neoplastic pro
cesses are less likely. Occasionally, pathognomonic clinical syndromes can be
identified, and these are extremely helpful; examples are listed in Table 12.1.
SECTION 1
More frequently though, the clinical findings are nonspecific and additional ev
idence is sought through diagnostic testing.
All patients in the neurological intensive care unit should undergo a routine
workup for systemic medical conditions that may be associated with or un
derlying the presenting neurologic syndrome. These medical conditions may
also affect the treatment decisions. Table 12.2 lists a set of basic screening labs
that can be helpful to rule out alternative diagnoses or support the diagnosis
of autoimmune dysfunction.
One of the first diagnostic studies obtained in most cases is some form of
neuroimaging. Although head computed tomography (CT) is often the most
rapidly attainable neuroimaging study, the diagnostic yield from this study is
120
generally poor. CT angiogram and postcontrast scans are more useful, but
magnetic resonance imaging (MRI) of the neuraxis with and without gado
linium has become the cornerstone of advanced neurological workup in the
intensive care unit. Patterns of enhancement, T2 abnormalities, diffusion re
striction, and even perfusion sequences may be pathognomonic for specific in
fectious, toxic, or inflammatory conditions. More advanced imaging, including
spectroscopy and fluorodeoxyglucose-positron emission tomography (FDG-
PET; body and brain) scans have also become standard diagnostic tools during
Table 12.1 Classic Autoimmune Syndromes of the Central

Nervous System
Classic Syndromes Possible Diagnosis
Limbic encephalopathy Paraneoplastic, HSV, HHV6
Faciobrachial dystonic seizures Lgl1 antibodies
Cerebellar degeneration Opsoclonus/myoclonus Paraneoplastic, parainfectious
Optic neuritis Plus NMO, MS
Miller-Fisher syndrome Anti-GQ1b
Stiff-person syndrome Anti-GAD, Anti-GABA-a,
Anti-amphiphysin
Neuromyotonia Anti-CAPR2
HSV = herpes simplex virus; NMO = neuromyelitis optica; MS = multiple sclerosis.
Inflammation and Demyelination
Table 12.2 Suggested Clinical and Laboratory
Tests of Suspected Autoimmune CNS Disease
Alternative Pathologies Inflammatory Markers
CBC w/diff ESR
Electrolytes CRP
BUN/Cr ANA
Glucose RF
UA and culture ANCA
LFTs Antiphospholipid screen
Ammonia Anti-dsDNA
B12 Anti-Ro/La
RPR Compliment levels
Chapter 12
TSH Cryoglobulins
Cortisol Paraneoplastic panels
CXR Anti-TPO AB
Medication review
Toxicology screen
Constipation screen
Blood cultures
121
Viral serologies
Serum flow cytometry
β2 microglobulin
CNS = central nervous system.
the workup up of these neurologically ill patients. Table 12.3 describes imaging
modalities and possible diagnostic conclusions that these studies may lead to.
The most important test in the workup of inflammatory disorders of the
nervous system remains the lumbar puncture, and the diagnostic value of ce
rebrospinal fluid (CSF) analysis in CNS dysfunction cannot be emphasized
enough. Table 12.4 lists initial CSF studies that should be sent in all patients
when autoimmune CNS disease is on the differential diagnosis. CSF inflam
mation is most frequently defined by the presence of one or more of the
following: pleocytosis (>5 cells), elevated protein (>50 mg/dL), the presence
of oligoclonal bands (>1), and an elevated IgG index (>0.66). It is important
to recognize that many pathologic antibodies may be present in CSF only and
without significant evidence of CSF inflammation.
Electroencephalography (EEG) is rarely specific but often very informative
diagnostically in understanding encephalopathy or even localizing pathologic
regions within the brain. Occasionally, certain findings may be suggestive of
specific disease processes (see Table 12.3 for details).
Table 12.3 Suggested Imaging and Diagnostic Studies for the

Evaluation of Autoimmune Central Nervous System Disease
Diagnostics Finding Potential Diagnosis
CT/CTA Head Vascular beading Vasculitis
Venous engorgement AVM, fistula, VST
Atrophy Neurodegenerative process
CT Chest/Abd/ Mass Malignancy screen
Pelvis
MRI Leptomeningeal enhancement Meningitis
SECTION 1
Mesotemporal T2/DWI changes Limbic encephalitis

Focal cortical DWI changes Focal seizures
Cortical DWI ribboning CJD, anoxia
Dawson’s fingers Multiple sclerosis
Basal ganglia T2 changes Arbovirus, toxydromes
Microhemorrhages CAA, TTP, DIC
MRI spectroscopy Lactate peak Metabolic abnormalities
FDG PET Focal uptake Malignancy
Focal cerebellar uptake Acute cerebellar degeneration
Mesotemporal uptake Limbic encephalitis
122
Spinal cord uptake Transverse myelitis

EEG Extreme delta brush NMDA encephalitis
PSWC CJD
Periodic temporal discharges HSV
Diffuse slowing with triphasics Metabolic encephalopathy
Transvaginal US Ovarian mass NMDA encephalitis
CT = computed tomography; CTA = computed tomography angiography; AVM = arteriovenous
malformation; VST = venous sinus thrombosis; MRI = magnetic resonance imaging; FDG-
PET = fluorodeoxyglucose-positron emission tomography; EEG =electroencephalography;
PSWC = periodic sharp wave complexes; CJD = Creutzfeldt-Jakob disease; CAA = cerebral amyloid
angiopathy; TTP = thrombotic thrombocytopenia purpura; DIC = disseminated intravascular
coagulopathy; HSV = herpes simplex virus; US = ultrasound.
Often the workup of neurologic disease yields only nonspecific markers

of inflammation without providing a specific diagnosis, in which case biopsy
may be required to further narrow the pathologic process and guide treat
ment. When considering CNS biopsy, the potential diagnostic benefit must be
weighed against the very significant risk of potentially permanent neurologic
injury. In general, targets for biopsy should be in regions of active disease in
volvement on neuroimaging. When the area of active involvement is inacces
sible, the potential diagnostic yield of biopsy drops considerably, and the utility
of such an intervention should be further considered. When feasible, brain
and/or meningeal biopsy yields invaluable information regarding the nature of
the inflammatory response, the underlying cellular/immune process, and the
Table 12.4 CSF Studies Consistent with
Inflammation of the Central Nervous System
CSF Study Result
Glucose Normal
Protein elevated
IgG index >0.66
Oligoclonal bands >1
Paraneoplastic panel positive
New generation sequencing negative
WBCs 5–100
Flow cytometry normal
Cytology normal
Chapter 12
HSV1/HSV2 PCR negative
VZV PCR and Ab negative
β2 microglobulin normal
IgH gene rearrangement absent
CSF = cerebrospinal fluid; WBC = white blood count;
PCR = polymerase chain reaction.
123
microstructural distribution of the inflammation, all of which can have signifi
cant impact on the choice of therapy.
Treatment Options
In the intensive care unit (ICU) setting, the clinical examination and diagnostic
workup must be focused on identifying specific pathophysiologic processes
that allow for early targeted treatment, rather than solely aimed at securing
a specific diagnosis. The balance between the risks of treatments, diagnostic
confidence, and the risk of disease progression will ultimately define individual
patient care. Table 12.5 lists categories of autoimmune pathophysiology and
associated disorders as well as potential acute interventions. The risk of
these interventions is real and may significantly contribute to secondary se
quelae such as opportunistic infections. This is true for other widely accepted
indications for these medications including immune suppression in organ
transplantation.
The treatment of immunologic diseases of the brain is not limited to
immunomodulation, and it is important to recognize that immunolog
ical attacks on central nervous tissue may manifest as seizures, vascular
events, cerebral edema, pain, or psychiatric symptoms. These clinically rel
evant symptoms may have to be treated directly in the acute setting and
may be even exacerbated by disease-modifying therapies. Finally, many of
Table 12.5 Autoimmune Central Nervous System Diseases and Treatments Classified by Mechanism
of Underlying Immune Dysfunction
Predominant T-Cell Mediated B-Cell Mediated Granulomatous Autoinflammatory NOS
Pathophysiology Disorders Disorders
Disorders Multiple sclerosis SLE Sarcoidosis Behçet’s disease Susac disease
ADEM NMO (anti-AQP4) GCA Churg-Strauss
Syndrome
PACNS (PCNSV) Transverse myelitis NOS Wegener’s Microscopic
granulomatosis Polyangiitis
Aβ-related angiitis Miller-Fisher syndrome
IgG4-RD Bickerstaff encephalitis
Sjögren’s syndrome Antiphospholipid syndrome
Intracellular antigen SREAT
antibodies: Hu, Ma2, Stiff-Person syndrome
GAD, CV2, Ri, Yo,
Amphiphysin
CLIPPERS Synaptic antibodies:
NMDA, AMPA, GABA,
mGluR5, D2 receptor
Channel targeting
antibodies:
LGl1, CASPR2, DPPX,
MOG
Treatments Glucocorticoids Glucocorticoids Glucocorticoids Glucocorticoids Glucorticoids
Cyclophosphamide Plasma exchange TNF-alpha inhibitors TNF-alpha inhibitors Cyclophosphamide
(GCA does not
respond)
Anti-CD20 targeting IVIG Anti-CD20 targeting Anti-CD20 targeting
therapies therapies therapies
Natalizumab Anti-CD20 targeting Cyclophosphamide
therapies
Anti-C5 (eculizumab)
Anti-IL-6R (tocilizumab)
ADEM = acute disseminated encephalomyelitis; PACNS = primary angiitis of the central nervous system; SLE = systemic lupus erythematosus; NMO = neuromyelitis optica;
SREAT = steroid responsive encephalopathy associated with autoimmune thyroiditis; GCA = giant cell arteritis; CLIPPERS = chronic lymphocytic inflammation with pontine
perivascular enhancement responsive to steroids.
Compiled with assistance of Ivana Vodopivec, MD, PhD.
these autoimmune disorders have been associated with neoplasia or benign
tumors. These lesions are frequently small but need to be identified and
treated rapidly.
Treatment Risks
Treatment with immunomodulatory agents incurs a significant risk for infec
tion and other systemic side effects, and patients should be properly screened
and protected to minimize potential complications. Most complications re
SECTION 1
lated to chronic immune suppression are due to opportunistic infections; de

spite these, treatment is generally necessary as the autoimmune disease may
lead to loss of life or severe neurologic disability. Vaccinations play a significant
role in prevention against opportunistic infections in patients who are chron
ically immune suppressed. However, routine vaccinations against Influenza,
Streptoccocus pneumoniae, and latent Zoster are frequently avoided during the
period of acute illness as they may conceivably worsen immunologic disease.
Table 12.6 lists standard laboratory tests that should be obtained prior to
the initiation of immune suppression in order to better understand the risks
of these interventions. Frequently, these labs are positive but the immuno
logic treatment is still given; in these cases, additional antimicrobials, an infec
126
tious disease consultation, or additional discussions with the family may be

warranted. For example, JC virus serologies are often positive prior to initia
tion of therapy, in which case the JC virus (JCV) antibody index may be useful
to assess the relative risk of progressive multifocal leukoencephalopathy
(PML) in the individual patient ( JCV antibody index >1.5 indicates an in
creased risk for PML).
Glucocorticoids and many immunomodulators increase the risk for
Pneumocystis jiroveci pneumonia; this is more relevant in chronically
Table 12.6 Suggested Pretreatment

Screening Studies for Immunomodulatory
Therapy
Infection Screens Other Labs
Hepatitis B sAB BUN/Cr
Hepatitis B cAB LFTs
Hepatitis C AB hCG
HIV AB/CD4 IgM, IgG, IgA
HIV PCR Vitamin D
Strongyloides TMPT
T. cruzi
JC virus AB index
PPD/IGRA
Table 12.7 Doses and Appropriate Prophylaxis for Commonly
Used Immunomodulatory Drugs
Agents Dosing Major risks Prophylaxis
Glucocorticoids Methylprednisolone Psychosis, Proton pump
1000 mg IV QD myopathy, inhibitor
× 3–5 days hyperglycemia, Vitamin D/calcium
Dexamethasone 4– osteonecrosis,
TMP/SMX, Mepron,
10 mg IV Q6H × hypertension
Dapsone
3–5 days
Cyclophospha Partners MS Hemorrhagic IVF
mide protocol: 800 cystitis, Mesna
mg/m2 IV Q4 lymphoma,
antiemetics
weeks × 6 sterility,
hepatotoxicity TMP/SMX
EULAR protocol: 15
Chapter 12
mg/kg IV Q2
weeks × 3
SLE NIH protocol:
0.5–1 g/m2 Q4
weeks × 6
EURO lupus
protocol: 500 mg
IV Q2 weeks × 6
127
Anti-CD20 Rituximab 1000 mg Anaphylaxis, Acetaminophen
Q2 weeks × 2 hypogamma Diphenhydramine
Rituximab 375 mg/ globulinemia,
Methylprednisolone
m2 Qweek × 4 PML, fever
Anti-TNFalpha Infliximab IV 3–10 Anaphylaxis, Treat latent Tb
mg/kg Q2 weeks hepatotoxicity, TMP/SMX
Adalimumab SC 40 optic neuritis, CNS
acetaminophen
mg Q2 weeks demyelination
IVIG 2 g/kg over 3–5 days Anaphylaxis, Acetaminophen
hypercoagulability, Diphenhydramine
aseptic meningitis,
anemia
PLEX 1–1.5 plasma Hypotension, IVF
volumes × 5 over coagulopathy
10 days
Anti-C5 Eculizumab 400–1200 Hypertension, Acetaminophen
mg IV Q2 weeks anemia Diphenhydramine
Anti-IL-6R Tocilizumab 8 mg/kg Anaphylaxis, GI Acetaminophen
IV Q4 weeks perforation Diphenhydramine
Natalizumab 300 mg IV Q4 weeks PML, anaphylaxis Acetaminophen
Diphenhydramine
SLE = systemic lupus erythematosus; NIH = National Institutes of Health; GI = gastrointestinal;
PML = progressive multifocal leukoencephalopathy; IV = intravenous.
immune- suppressed patients, but antibiotic prophylaxis is frequently
considered. Table 12.7 lists available acute immunomodulatory regimens

and associated prophylaxis. While there is limited data guiding the use of
these medications, there is often little to lose in critically ill patients with
rapidly progressive immunologic disorders, and the risk of withholding
potentially efficacious treatment often outweighs those associated with
these agents.
Finally, immunologic interventions may affect the yield of future diagnostic
studies; for this reason it is reasonable to collect extra serum and necessary
tissue biopsies prior to the initiation of therapy. If a paraneoplastic disorder
SECTION 1
is being treated, the treatment should be focused against the primary ma
lignancy in addition to immunologic modulation. This requires a multidisci
plinary approach involving oncologists, surgeons, radiation oncologists, and
neurologists.
Conclusion
Immunologic disorders in the neurocritical care unit often cause significant
morbidity and mortality and are associated with prolonged and expensive ICU
stays. Fortunately, if treated rapidly these are potentially reversible disorders.
128
Expanding our understanding of the basic pathophysiology of these diseases

and focusing on currently available immunomodulatory tools will allow us to
benefit more patients in the future.
Further Reading
Agnihotri S, Singhal T, Stern B, Cho T. Neurosarcoidosis. Semin Neurol.
2014;34(4):386–394.
Antiphospholipid Antibodies in Stroke Study Group. Clinical and laboratory findings
in patients with antiphospholipid antibodies and cerebral ischemia. Stroke.
1990;21(9):1268–1273.
Bhattacharyya S, Berkowitz AL. Primary angiitis of the central nervous system: avoiding
misdiagnosis and missed diagnosis of a rare disease. Pract Neurol. 2016;16(3):195–
200. doi:10.1136/practneurol-2015-001332
Bhattacharyya S, Helfgott SM. Neurologic complications of systemic lupus erythema
tosus, Sjögren syndrome, and rheumatoid arthritis. Semin Neurol. 2014;34(4):425–436.
Biotti D, Deschamps R, Shotar E, et al. CLIPPERS: chronic lymphocytic inflamma
tion with pontine perivascular enhancement responsive to steroids. Pract Neurol.
2011;11(6):349–351.
Castillo P, Woodruff B, Caselli R, et al. Steroid-responsive encephalopathy associated
with autoimmune thyroiditis. Arch Neurol. 2006;63(2):197–202.
Dalmau J, Rosenfeld MR. Paraneoplastic syndromes of the CNS. Lancet Neurol.
2008;7(4):327–340.
Dalmau J, Tüzün E, Wu HY, et al. Paraneoplastic anti-N-methyl-D-aspartate receptor

encephalitis associated with ovarian teratoma. Ann Neurol. 2007;61(1):25–36.
Eng JA, Frosch MP, Choi K, Rebeck GW, Greenberg SM. Clinical manifestations of ce
rebral amyloid angiopathy-related inflammation. Ann Neurol. 2004;55(2):250–256.
Graus F, Titulaer MJ, Balu R, et al. A clinical approach to diagnosis of autoim
mune encephalitis. Lancet Neurol. 2016;(15):S1474– S4422. doi:10.1016/
S1474-4422(15)00401-9
Gwathmey K, Balogun RA, Burns T. J Neurologic indications for therapeutic plasma ex
change: 2013 update. J Clin Apher. 2014;29(4):211–219.
Hajj-Ali RA, Singhal AB, Benseler S, Molloy E, Calabrese LH. Primary angiitis of the
CNS. Lancet Neurol. 2011;10(6):561–572.
Hardy TA, Chataway J. Tumefactive demyelination: an approach to diagnosis and man
agement. J Neurol Neurosurg Psychiatry. 2013;84(9):1047–1053.
Hauser SL, Waubant E, Arnold DL, et al. B-cell depletion with rituximab in relapsing-
Chapter 12
remitting multiple sclerosis. N Engl J Med. 2008;358(7):676–688.
Lafitte C, Amoura Z, Cacoub P, et al. Neurological complications of primary Sjögren’s
syndrome. J Neurol. 2001;248(7):577–584.
Lancaster E, Dalmau J. Neuronal autoantigens—pathogenesis, associated disorders and
antibody testing. Nat Rev Neurol. 2012;8(7):380–390.
Linker RA, Gold R. Use of intravenous immunoglobulin and plasma exchange in neurol
ogical disease. Curr Opin Neurol. 2008;21(3):358–365.
129
Lu LX, Della- Torre E, Stone JH, Clark SW. IgG4- related hypertrophic
pachymeningitis: clinical features, diagnostic criteria, and treatment. JAMA Neurol.
2014;71(6):785–793.
McDaneld LM, Fields JD, Bourdette DN, Bhardwaj A. Immunomodulatory therapies in
neurologic critical care. Neurocrit Care. 2010;12(1):132–143.
McKeon A, Pittock SJ, Lennon VA. CSF complements serum for evaluating paraneoplastic
antibodies and NMO-IgG. Neurology. 2011;76(12):1108–1110.
Olmez I, Moses H, Sriram S, Kirshner H, Lagrange AH, Pawate S. Diagnostic and ther
apeutic aspects of Hashimoto’s encephalopathy. J Neurol Sci. 2013;331(1–2):67–71.
Plavina T, Subramanyam M, Bloomgren G, et al. Anti-JC virus antibody levels in serum
or plasma further define risk of natalizumab- associated progressive multifocal
leukoencephalopathy. Ann Neurol. 2014;76(6):802–812.
Saip S, Akman- Demir G, Siva A. Neuro- Behçet syndrome. Handb Clin Neurol.
2014;121:1703–1723.
Scolding NJ, Joseph F, Kirby PA, et al. Abeta-related angiitis: primary angiitis of the cen
tral nervous system associated with cerebral amyloid angiopathy. Brain. 2005;128(Pt.
3):500–515.
Shahrizaila N, Yuki N. Bickerstaff brainstem encephalitis and Fisher syndrome: anti-
GQ1b antibody syndrome. J Neurol Neurosurg Psychiatry. 2013;84(5):576–583.
Shams’ili S, Grefkens J, de Leeuw B, et al. Paraneoplastic cerebellar degeneration as
sociated with antineuronal antibodies: analysis of 50 patients. Brain. 2003;126(Pt.
6),1409–1418.
Tenembaum S, Chitnis T, Ness J, Hahn JS, International Pediatric MS Study Group. Acute
disseminated encephalomyelitis. Neurology. 2007;68(Suppl. 2):S23–S36.
Vodopivec I, Miloslavsky E, Kotton C, Cho T. a neurologist’s guide to safe use of immu
nomodulatory therapies. Semin Neurol. 2014;34(4): 467–478.

von Geldern G, McPharlin T, Becker K. Immune mediated diseases and immune modu
lation in the neurocritical care unit. Neurotherapeutics. 2011;9(1):99–123.
Yu Z, Kryzer TJ, Griesmann GE, Kim K, Benarroch EE, Lennon VA. CRMP-5 neuronal
autoantibody: marker of lung cancer and thymoma- related autoimmunity. Ann
Neurol. 2001;49(2):146–154.
130 SECTION 1
Chapter 13
Neuromuscular Conditions
Deepa Malaiyandi and Saša A. Živković
Introduction
The clinical manifestation of neuromuscular disorders in the intensive care
unit (ICU) setting includes focal or diffuse weakness, dysphagia, sensory
loss, and, when severe, respiratory failure. Physical signs of impending respi
ratory failure include tachypnea, slow and deliberate speech, use of acces
sory respiratory muscles at rest, and ineffective cough, which in combination
with dysphagia may lead to aspiration and sudden worsening of respiratory
function. Various other comorbidities can precipitate worsening of respiratory
function, including infections and toxic/metabolic disturbances. Pulmonary
function testing and clinical alertness are essential to avoid unexpected res
131
piratory failure as delayed intubation is directly related to worse outcomes
and increased morbidity and mortality. In addition to severe weakness and
respiratory failure, progressive cardiomyopathy associated with muscular dys
trophies can precipitate cardiac failure and even necessitate cardiac transplan
tation. Guillain-Barré syndrome (GBS) is one of few neuromuscular disorders
that can be associated with significant dysautonomia. Clinical manifestations
of dysautonomia include orthostatic hypotension, cardiac arrhythmia, gas
trointestinal dysmotility, and anhidrosis. Prolonged ICU stay in patients with
severe medical illnesses may also lead to ICU-acquired weakness or nerve
entrapments.
Based on pathophysiology and site of neuromuscular dysfunction we can
divide neuromuscular disorders into (a) disorders of muscle; (b) disorders of
neuromuscular junction; (c) disorders of peripheral nerve, and (d) disorders
of motor neurons.
Myopathies
Respiratory muscle weakness associated with myopathies typically manifests
late in the course of disease as a result of severe diffuse muscle weakness.
However, respiratory failure may also be the initial presentation in ambula
tory patients with mild or minimal limb weakness. Weakness of respiratory
muscles can directly result in respiratory failure, and contributing factors may
include kyphoscoliosis, weakness of pharyngeal and laryngeal muscles, and un
derlying pulmonary disorders. Inflammatory myopathies may be associated

with ventilatory dysfunction due to inflammation of respiratory muscles and
also due to interstitial lung disease (especially with elevated Jo-1 antibody
titers). Nevertheless, respiratory failure is rare in patients with inflamma
tory myopathies. Hereditary myopathies are typically associated with slowly
progressive weakness but can manifest with sudden respiratory failure fol
lowing unrecognized chronic respiratory insufficiency. Rhabdomyolysis in ICU
patients often presents in the absence of muscle weakness and pain, and it has
been estimated that 10% to 50% of patients with rhabdomyolysis may develop
SECTION 1
acute kidney injury. Risk factors for rhabdomyolysis in the ICU setting include
trauma, recent surgery, and vascular occlusions. An underlying myopathy can
also present with rhabdomyolysis following use of medications or strenuous
exertion, and it is recommended to wait at least six weeks from resolution for
muscle biopsy, as necrosis may mask the underlying primary disease of the
muscle. Treatment of rhabdomyolysis is mostly focused on preserving renal
function with fluid replacement and potentially dialysis.
Myasthenia Gravis
132
Myasthenia gravis (MG) has a prevalence of 14 to 20 per 100,000 in the

United States. The typical clinical course of MG is characterized by fatigable
weakness and episodic exacerbations. It often begins with ocular symptoms
such as ptosis or diplopia, but more than 80% of patients eventually develop
generalized weakness. However, if symptoms remain isolated to the eyes for
>3 years, the likelihood of generalization is <10%. Weakness in MG results
from impaired activation of post-synaptic acetylcholine receptors (AChR),
often associated with high titers of antibodies targeting AChR (blocking,
binding, and modulating) receptors, muscle specific kinase, or lipoprotein-
protein related protein 4. Yet, 10% to 15% of patients will have seronegative
MG with normal antibody titres. Repetitive nerve stimulation (RNS) during
nerve conduction study (NCS) is a common diagnostic test for MG and may
show the classic decremental responses (Figure 13.1). Single-fiber electromy
ography (EMG), which is the most sensitive test for MG, is often not feasible
in the ICU setting.
Myasthenic crisis is a medical emergency defined as an exacerbation se
vere enough to require intubation or delay postoperative extubation. It
manifests with bulbar or generalized weakness that progresses to respiratory
failure. It can be triggered by various medications (Table 13.1), intravenous
magnesium, infection, surgery, pregnancy, or labor, and the diagnosis can
be challenging when crisis is the first manifestation of disease. The standard
treatment of myasthenic crisis includes IVIG or plasma exchange (Table 13.2).
Subsequent maintenance therapy requires a disease-modifying agent such as
glucocorticoids, often in combination with steroid-sparing medications (e.g.,
Chapter 13
Figure 13.1 Decremental response on repetitive nerve stimulation in a patient with
myasthenia gravis.
133
Table 13.1 Medications Associated with Abnormal
Neuromuscular Junction Transmission
Absolute Relative Caution Typically Tolerated
Contraindication Contraindication (May (Rarely Exacerbate)
(Exacerbate) (Likely Exacerbate)
Exacerbate)
Anesthetic Agents
Neuromuscular Inhalation agents
blockers Local anestheticsa
Anti-infectious Agents
Ketolides Aminoglycosides Clindamycin, Antitretrovirals
Fluoroquinolones Vancomycin Metronidazole
Macrolides Nitrofurantoin
Quinine derivative Tetracyclines
Cardiovascular Agents
Procainamide, Calcium channel
Quinidine blockers
Beta-blockers
Other
Botulinum toxin Magnesium salts Iodinated contrasts Carbamazepine
D-Penicillamine Statins Haldol
Interferon-α Glucocorticoids Lithium
Ophthalmic drops
Phenobarbital
Phenytoin
a
Risk of exacerbation exists only when inadvertently administered intravascularly.
Table 13.2 Treatment of Myasthenic Crisis
Drug Indication Starting Dose Effect Time Caution Monitoring
IVIG Symptomatic/Acute 1–2 g/kg over 2–5 days 1 week Transfusion reaction, anaphylaxis, Vital signs, rash, fever
HTN, AKI, aseptic meningitis viral
transmission, VTE
PLEX Symptomatic/Acute 3–5 L QOD × 7–14 days 1 week Hypotension Vital signs
Rituximab Refractory 375 mg/m2 IV 1×/week × 4 4–12 weeks Leukopenia, opportunistic infections CBC
Cyclophosphamide Refractory 150–200 mg/daya IV 3–12 weeks Hemorrhagic cystitis,b Cr, CBC, UA
myelosuppression, malignancy
a
Dosing regimens vary and some studies have used doses as high as 500 mg/m2 or 50 mg/kg IV; b Co-administration of MESNA prevents hemorrhagic cystitis with IV cyclophosphamide.
IVIG = intravenous immunoglobulin therapy; HTN = hypertension; AKI = acute kidney injury; VTE = venous thromboembolism; PLEX = plasma exchange; IV = intravenous;
Cr = creatinine; CBC = complete blood count; UA = urine analysis.
azathioprine, mycophenolate), that can be started in the ICU as it takes
months to achieve a full response. The use of cholinesterase inhibitors is gen
erally avoided in intubated patients to reduce secretions. Also, the half-life of
these agents can vary greatly in patients on multiple medications and dosing
schedule should be considered when planning extubation.
Preferences for rapid sequence intubation strategies in MG vary greatly as
myasthenics are often sensitive to nondepolarizing neuromuscular blockers
(NMB), are resistant to depolarizing agents, and have varying responses to
reversal agents, which are not routinely recommended. The combination of
propofol and remifentanyl (or fentanyl) offers a non-paralytic option with
which optimal conditions can be achieved within 2.5 minutes. Epinephrine
Chapter 13
should be readily available to mitigate potential bradycardia and hypotension.
Other intravenous anesthetics such as barbiturates, ketamine, etomidate,
and midazolam have all been used without complication in myasthenic
patients. Local or regional anesthesia is preferred when possible. If general
anesthesia is used, the lowest possible dose or no NMB should be used as
inhaled anesthetics afford a dose-dependent neuromuscular relaxation akin
to NMB in MG.
Botulism
135
Botulism is a toxin-mediated, life-threatening neuroparalytic syndrome. The
toxin is produced by Clostridium botulinum, a gram-positive, spore-forming,
obligate anaerobic bacillus. The clinical syndrome results from the toxin
binding to and inhibiting various proteins involved in the presynaptic release
of acetylcholine. Infant botulism is the most common variant and results from
ingestion of spores following an incubation period of hours to seven days.
Wound botulism results from inoculation of a wound, then vascular dissemi
nation and in vivo toxin production. It is unique in presentation due to the ab
sence of gastrointestinal (GI) symptoms and an incubation period of 10 days.
Heroin and cocaine abusers are at increased risk. Food-borne illness from in
gestion of pre-formed toxin manifests within 36 hours. Iatrogenic intoxication
is truly rare with only five confirmed cases following Botox® use. Illness from
inhalation of toxin presents in <6 hours and carries important implications
for bioterrorism. Illness often presents with GI symptoms, followed by de
scending paralysis yielding bilateral cranial nerve palsies and symmetric neu
rological deficits. Definitive diagnosis is made by isolation of toxin in stool or
serum bioassay in mice. Electrodiagnostic testing typically shows fibrillations
on EMG and incremental responses with high-frequency RNS.
Once concern for botulism arises, the state health department should be
contacted to assist in determining appropriateness of antitoxin, facilitate its
delivery from the Centers for Disease Control and Prevention, and provide
a designated testing center. The human-derived botulism immune globulin is
reserved for infants. All others receive equine serum heptavalent antitoxin and
need close monitoring for severe reactions including serum sickness and ana
phylaxis. Respiratory failure is the primary cause of death, and early intubation
improves mortality. Antibiotics are only indicated in wound botulism as bacte
rial lysis in other cases releases more toxin. If wound or serum is positive for
bacteria or toxin, penicillin G or metronidazole is used. Aminoglycosides are
avoided given their ability to impair neuromuscular junction (NMJ) transmis
sion. Cathartics are also used in the absence of ileus. Botulism is not conta
gious, and standard precautions suffice. Intoxication with toxin A, misdiagnosis,
and presentation with isolated shortness of breath are poor prognostic factors.
SECTION 1
Organophosphate Toxicity
Organophosphate agents are cholinesterase inhibitors used as insecticides that can
also affect NMJ transmission. Acute poisoning with organophosphates may result
in severe weakness and respiratory failure with alteration of consciousness and
seizures. A delayed intermediate syndrome that occurs 24 to 96 hours after ini
tial exposure may require prolonged ventilator support. EMG/NCS show repet
itive responses after a single stimulus and decremental or decrement-increment
response with high- frequency RNS. Organophosphorus chemicals are most
commonly used as insecticides but have been also used as chemical weapons,
136
commonly known, as “nerve agents.” Treatment is based on antimuscarinic

agents, oxime therapy, and benzodiazepines (for control of seizures).
Guillain-Barré Syndrome
GBS is a rare, potentially fatal neuromuscular disorder of peripheral nerves
with an annual incidence of 1 to 2 per 100,000. GBS is often triggered by var
ious infections and manifests with rapidly progressive symmetric weakness.
Diagnosis is based on clinical features, elevated protein content in cerebro
spinal fluid, and typical findings on EMG/NCS. Rapidly progressive weakness
may involve respiratory muscles, and respiratory failure requiring mechanical
ventilation has been reported in 20% to 30% of patients. Delayed recognition
of respiratory insufficiency and late intubation are associated with prolonged
hospital stay and increased morbidity and mortality. Additionally, prolonged
mechanical ventilation (>2 months) is also associated with worse outcomes
and delayed recovery. Mortality associated with GBS remains at 4% to 5%,
and the leading cause of death is ventilator-associated pulmonary infections,
especially in elderly patients with pre-existing pulmonary conditions. Mortality
is substantial in ventilated GBS patients reaching up to 20%, but 79% of
survivors will eventually regain independent ambulation. Dysautonomia is often
overshadowed by other more obvious clinical manifestation of GBS, but ev
idence of autonomic dysfunction has been reported in up to two-thirds of
patients with GBS. Severe arrhythmias have been reported in up to 11% of
patients, and these patients more often develop systolic hypertension. Acute
idiopathic isolated dysautonomia, specifically without weakness or sensory loss,
that responds to intravenous immunoglobulin therapy (IVIG) has also been rec
ognized as a rare variant of GBS. Overall, severity of ICU complications in GBS
patients correlates with long-term mechanical ventilation and dysautonomia
and increases with prolonged ICU stay beyond three weeks. GBS is treated
with IVIG (2 g/kg over two to five days) or plasma exchange (usually five
exchanges), in addition to supportive measures for respiratory failure and
dysautonomia if needed. Treatment of dysautonomia in GBS is complex given
the various clinical manifestations and severity. These patients may exhibit an
increased sensitivity to treatment with antihypertensive medications, and car
diac arrhythmias may require pacing. Additionally, treatment with plasma ex
Chapter 13
change is usually avoided in the presence of cardiovascular dysautonomia.
Motor Neuron Disease

Amyotrophic lateral sclerosis (ALS) is the most common motor neuron dis
ease in adults with a lifetime risk of 1 in 800. ALS is typically a chronic pro
gressive disease which eventually affects respiratory muscles with an average
survival of 26 months. Typically, respiratory failure presents late in the course
with diffuse generalized weakness. However, ALS with respiratory onset has
137
been reported in 2% to 3% of patients, and one in seven of these patients
may require emergency intubation. Similarly, as with other neuromuscular
disorders, ALS patients may develop aspiration pneumonia, especially in
the setting of dysphagia, and this has been reported in 13% of ALS patients.
Treatment of ALS in the ICU focuses on supportive care.
ICU-Acquired Weakness
ICU-acquired weakness (ICUAW) is an umbrella term applied to those ICU
patients with symmetric limb weakness and/or difficulty weaning from me
chanical ventilation. It includes critical illness polyneuropathy, critical illness my
opathy (CIM), critical illness neuromyopathy, and cases where definitive testing
is not feasible because of electrical interference, the patient’s inability to co
operate, or peripheral edema, all of which may limit electrodiagnostic testing.
EMG/NCS may show evidence of CIM, but muscle biopsy is required for the
definitive diagnosis of CIM (Figure 13.2). Nevertheless, biopsy is rarely pursued
after clinical diagnosis is made. Independent risk factors for ICUAW include
female gender, sepsis/systemic inflammatory response syndrome, multiorgan
and renal failure, immobility, hyperglycemia, and use of glucocorticoids or NMB.
The exact mechanisms remain unknown, but hypotheses implicate microvas
cular and mitochondrial failure as well as sodium channelopathy. At this time
there is no specific treatment for ICUAW. Intensive insulin therapy has been
shown to decrease incidence, accelerate weaning off ventilator support, and
improve 180-day mortality, but it also increases hypoglycemia, thus it remains
138 SECTION 1
Figure 13.2 Critical illness myopathy with loss of myosin thick filaments shown on an
electron photomicrograph.
unclear whether the benefits outweigh the risks. However, ICU practices
have shifted and now emphasize daily coordination of awakening and sponta
neous breathing trials, delirium assessments, and early exercise to prevent and
manage ICUAW.
Neuromuscular Respiratory Failure

In patients with NMD, respiratory failure can develop from progression of
a chronic condition, such as ALS, an exacerbation of a fluctuating disorder,
such as MG, or a sudden onset of acute illness, such as in GBS. Three main
mechanisms of respiratory failure in NMD include (a) loss of upper airway
muscle tone with dysphagia leading to increased risk of aspiration and
varying degrees of airway obstruction; (b) concurrent cardiopulmonary dis
ease such as cardiomyopathy and congestive heart failure, and, when these
limit mobility, risk of pulmonary embolism increases; and (c) weakness of
respiratory muscles including the diaphragm, intercostal, and accessory mus
cles. The diaphragm is most important when supine, and its weakness may
manifest as orthopnea. It is the only respiratory muscle active during rapid
eye movement sleep. Hypoventilation provokes tachypnea, further straining
Box 13.1 The Cardinal Signs of Impending Neuromuscular
Respiratory Failure
Tachypnea, orthopnea, paradoxical breathing
Dyspnea at rest or with minimal exertion
Accessory muscle contraction on inspiration (visible or palpable)
Sinus tachycardia
Diaphoresis (on forehead in particular)
Anxiety or generalized fatigue
Staccato speech, dysphonia, hypophonia
Dysphagia (pooling of secretions or coughing with swallowing), Weak cough
Frequent nocturnal awakenings or nonrefreshing sleep
Chapter 13
Weakness of neck flexors and extensors
Single breath count <15 (a count of 10 approximates a forced vital capacity [FVC] of 1 L)
*FVC < 30%–35% of predicted (~ < 20 mL/kg)
*Negative inspiratory force (NIF) or maximal inspiratory pressure (MIP) > –30 cmH2O
*Maximal expiratory pressure (MEP) < 40 cm H2O
*Decreased in FVC, NIF, or MEP by > 30%
Arterial partial pressure of oxygen (PaO2) < 70
**Arterial partial pressure of carbon dioxide (PaCO2) > 45–50
* All of these values can be falsely low in patients with significant bulbar weakness as they may be
139
unable to form a sufficient seal around their mouths to produce reliable values.
** Applies to patients without comorbid baseline hypercapnia.
already weakened muscles, and respiratory failure ensues (Box 13.1).

The first step in the management of neuromuscular respiratory failure
is to secure the airway. Systemic effects of NMD on cardiorespiratory
and autonomic reserve increase the risk of emergent intubation. In some
instances, noninvasive bilevel positive pressure ventilation (BiPAP) can be
trialed on a case-by-case basis in patients with little or no bulbar involve
ment who have not yet developed hypercapnia and have a history of
rapid recovery. If BiPAP is tried, respiratory decline or lack of improve
ment must be acted upon quickly. Failure to wean BiPAP after nocturnal
use or following use for two to three hours while awake are indications
for intubation.
Further Reading
Abel M, Eisenkraft JB. Anesthetic implications of myasthenia gravis. Mt Sinai J Med.
2002;69(1–2):31–37.
Adalja AA, Toner E, Inglesby TV. Clinical management of potential bioterrorism-related
conditions. N Engl J Med. 2015;372(10):954–962.
Fletcher DD, Lawn ND, Wolter TD, Wijdicks EF. Long-term outcome in patients
with Guillain- Barre syndrome requiring mechanical ventilation. Neurology.

2000;54(12):2311–2315.
Ghasemi M, Norouzi R, Salari M, Asadi B. Iatrogenic botulism after the therapeutic use
of botulinum toxin-A: a case report and review of the literature. Clin Neuropharmacol.
2012;35(5):254–257.
Gilhus NE, Verschuuren JJ. Myasthenia gravis: subgroup classification and therapeutic
strategies. Lancet Neurol. 2015;14(10):1023–1036.
Henderson RD, Lawn ND, Fletcher DD, McClelland RL, Wijdicks EF. The mor
bidity of Guillain-Barre syndrome admitted to the intensive care unit. Neurology.
2003;60(1):17–21.
SECTION 1
Hermans G, De Jonghe B, Bruyninckx F, Van den Berghe G. Interventions for preventing

critical illness polyneuropathy and critical illness myopathy. Cochrane Database Syst
Rev. 2014;1:CD006832.
Hermans MC, Pinto YM, Merkies IS, de Die-Smulders CE, Crijns HJ, Faber CG.
Hereditary muscular dystrophies and the heart. Neuromuscul Disord. 2010;20(8):
479–492.
Hughes RA, Cornblath DR. Guillain- Barre syndrome. Lancet. 2005;366(9497):
1653–1666.
Hulse EJ, Davies JO, Simpson AJ, Sciuto AM, Eddleston M. Respiratory complications
of organophosphorus nerve agent and insecticide poisoning. Implications for respira
tory and critical care. Am J Respir Crit Care Med. 2014;190(12):1342–1354.
140
Kiernan MC, Vucic S, Cheah BC, et al. Amyotrophic lateral sclerosis. Lancet.
2011;377(9769):942–955.
Kress JP, Hall JB. ICU-acquired weakness and recovery from critical illness. N Engl J Med.
2014;370(17):1626–1635.
Lacomis D. Neuromuscular disorders in critically ill patients: review and update. J Clin
Neuromuscul Dis. 2011;12(4):197–218.
Lawn ND, Wijdicks EF. Fatal Guillain-Barre syndrome. Neurology. 1999;52(3):635–638.
Nance JR, Mammen AL. Diagnostic evaluation of rhabdomyolysis. Muscle Nerve.
2015;51(6):793–810.
Nitahara K, Sugi Y, Higa K, Shono S, Hamada T. Neuromuscular effects of sevoflurane
in myasthenia gravis patients. Br J Anaesth. 2007;98(3):337–341.
Oosterhuis HJ. Observations of the natural history of myasthenia gravis and the effect
of thymectomy. Ann N Y Acad Sci. 1981;377:678–690.
Pfeffer G, Povitz M, Gibson GJ, Chinnery PF. Diagnosis of muscle diseases presenting
with early respiratory failure. J Neurol. 2015;262(5):1101–1114.
Phillips LH, 2nd. The epidemiology of myasthenia gravis. Ann N Y Acad Sci.
2003;998:407–412.
Shoesmith CL, Findlater K, Rowe A, Strong MJ. Prognosis of amyotrophic lateral scle
rosis with respiratory onset. J Neurol Neurosurg Psychiatry. 2007;78(6):629–631.
Sorenson EJ, Crum B, Stevens JC. Incidence of aspiration pneumonia in ALS in Olmsted
County, MN. Amyotroph Lateral Scler. 2007;8(2):87–89.
Wakerley BR, Uncini A, Yuki N, GBSC Group. Guillain- Barre and Miller Fisher
syndromes—new diagnostic classification. Nat Rev Neurol. 2014;10(9):537–544.
Winer JB, Hughes RA. Identification of patients at risk of arrhythmia in the Guillain-
Barre syndrome. Q J Med. 1988;68(257):735–739.
Chapter 14
Movement Disorder
Emergencies and Movement
Disorders in the ICU
Mihai C. Sandulescu and Edward A. Burton
Introduction
Movement disorders are characterized by slowness and paucity of voluntary
movement (hypokinetic disorders) or involuntary movements (hyperkinetic
disorders). Conventionally, motor weakness (paresis) and motor incoordina-
tion (ataxia) are not included, since their anatomy, diagnostic workup, and
141
clinical management differ from hypokinetic and hyperkinetic disorders. The
pathophysiology of many movement disorders involves diseases or drugs that
impair neurotransmission in the basal ganglia, although there are exceptions.
The most common hypokinetic disorders are caused by pathology affecting
the dopaminergic projection from the substantia nigra of the midbrain to
the putamen and caudate. In addition to degenerative diseases such as
Parkinson’s disease (PD), drugs that are antagonists at dopamine receptors
can cause a very similar clinical picture. In contrast, lesions in a variety of
anatomical locations can provoke involuntary movements, and a syndromic
approach based on recognizing the type of involuntary movement is usually
most helpful. In this chapter, we summarize diagnosis and management of
common movement disorders that require hospital admission or intensive
care unit (ICU) management. These are grouped by the clinical appearance of
the most prominent abnormalities (hypokinetic versus hyperkinetic), although
it should be recognized that different types of movement abnormality can be
seen concurrently.
Hypokinetic Movement Disorders

Hypokinetic disorders are characterized by an increase in muscle tone that
is equal in flexion and extension and throughout the range of movement (ri-
gidity), reduced spontaneous movement (akinesia), and slowness of move-
ment (bradykinesia). The most common hypokinetic disorders are PD and
drug-induced parkinsonism.
Neuroleptic Malignant Syndrome
Neuroleptic malignant syndrome (NMS) is caused by dopamine receptor an-

tagonist drugs and has a very high mortality (20%–30%). All dopamine re-
ceptor antagonists and drugs that disrupt other aspects of dopaminergic
function have been reported to induce this syndrome, regardless of the pres-
ence of a psychiatric or neurological condition. The neuroleptics showing the
weakest association with NMS are clozapine and quetiapine. Symptoms of
NMS typically start within two weeks from initiating treatment or following a
rapid dose escalation. Reported risk factors include abrupt dose increase, con-
current use of lithium or selective serotonin reuptake inhibitors (SSRIs), dehy-
SECTION 1
dration, high temperatures, exhaustion, and prior NMS episode. The classical
clinical picture includes a change in mental status, fever, rigidity, dysautonomia,
and an elevated serum creatinine kinase (CK) (Box 14.1). However, not all
patients show all of these signs, and the diagnosis should be considered in any
patient with a subset of these features following exposure to relevant drugs.
Patients are usually transferred to the ICU for management of complications
and airway protection. The offending drug(s) are discontinued immediately.
CK should be monitored and appropriate measures taken to prevent renal
injury from rhabdomyolysis. Dantrolene, bromocriptine, and benzodiazepines
are the standard treatments in the ICU setting, although there may be a role
for other dopaminergic agents. Neuroleptics can be restarted two weeks
142
after the acute phase of an episode of NMS, if clinically indicated in patients

with ongoing psychosis. Drugs should be reintroduced at low doses and
slowly titrated with close monitoring of response and adverse effects. It is
recommended that concurrent use of lithium is avoided in this situation.
Parkinsonism-Hyperpyrexia Syndrome
It is well recognized that patients with PD who stop taking dopaminergic
medication abruptly can develop a syndrome that is clinically indistinguish-
able from NMS. Parkinsonism-hyperpyrexia syndrome (HPS) has also been
Box 14.1 Diagnosis of Neuroleptic Malignant Syndrome

• Exposure to a dopamine antagonist or dopamine-agonist withdrawal
• Hyperthermia
• Rigidity
• Mental status alteration
• Elevated CK
• Sympathetic nervous system lability (hypertension, fluctuating blood pressure, diapho-
resis or urinary incontinence)
• Tachycardia and tachypnea
Patients may have only a subset of these features; there are currently no validated diagnostic criteria.
Gurrera RJ, Caroff SN, Cohen A, et al. An international consensus study of neuroleptic malignant
syndrome diagnostic criteria using the Delphi method. J Clin Psychiatry. 2011;72(9):1222–1228.
described in patients after decreasing the dose of dopaminergic medication,
Movement Disorder Emergencies

changing from immediate release to extended release levodopa, discontinuing
nondopaminergic medications including anticholinergics, or even when deep
brain stimulation is stopped. The reported time of onset of symptoms after
changing dopaminergic therapy ranges from 18 hours to seven days. Treatment
consists of prompt recognition, ICU admission for supportive treatment and
management of complications as in NMS, and reinstituting dopaminergic
therapy urgently. Options for treatment include non-ergot dopamine agonists,
such as ropinirole or pramipexole, levodopa preparations, or apomorphine
(see Table 14.1 for medications, including formulations that can be used in
patients who cannot swallow).
Managing PD Drugs in the ICU
Patients in the ICU are frequently unable to tolerate oral medications for a
Chapter 14
variety of reasons including unconsciousness, failure of bulbar or respiratory
function, loss of airway protection, and following some types of gastrointes-
tinal surgery. Due to the danger of parkinsonism-h yperpyrexia syndrome,
it is rarely advisable to discontinue dopaminergic medications abruptly in
these patients. Even though intravenous levodopa is considered generally
safe, it is not available in the United States. Immediate release levodopa/
carbidopa tablets can be crushed and dissolved and given by nasogastric
143
tube. In this case, care should be taken to avoid giving the tablets at the same
time as protein-rich enteral nutrition sources that compete with levodopa
for absorption. A nasogastric tube is often the most straightforward way
to administer PD medications to patients who cannot swallow tablets, and
many different formulations can be given this way. In situations where en-
teral absorption of levodopa is not possible, there are several other options
(Table 14.1).
In addition, it is important to remember that some drugs used commonly
in the ICU are harmful to PD patients. For example, several antiemetics (in-
cluding metoclopramide and prochlorperazine) are potent dopamine receptor
antagonists and are contraindicated in patients with PD. Alternatives, including
5HT3 receptor antagonist antiemetics such as ondansetron, can be used in
this setting if necessary. Other dopamine receptor blocking drugs that are
sometimes used in a hospital to manage agitation, such as haloperidol, are also
contraindicated in patients with PD, whereas benzodiazepines can be used ju-
diciously if necessary. For management of hallucinations and other psychotic
symptoms in PD, see later discussion.
Finally, it should be noted that monoamine oxidase B (MAO-B) inhibitors,
including rasagiline and selegiline, which are commonly used to treat symptoms
of PD, have the potential for significant drug interactions. Of particular impor-
tance is the risk of uncontrolled hypertension or serotonin syndrome (see
later discussion) with the analgesics meperidine and propoxyphene and the
antitussive dextromethorphan, all of which should be avoided. A large number
of drugs are known to interact with MAO-B inhibitors, and it is safest to check
Table 14.1 Dopaminergic Drugs: Formulations and Routes

of Administration
Drug Available Formulations Routes of Administration
Amantadine Capsules, tablets, oral PO; crushed/dissolved
suspension (Symmetrel®) tablets or oral suspension can
be given by NGT
Apomorphine Injectable (Apokyn®) subcutaneous
Carbidopa Tablets (Lodosyn®) PO; crushed/dissolved
tablets can be given by NGT
Entacapone Tablets (Comtan®) PO; crushed/dissolved
SECTION 1
tablets can be given by NGT

Levodopa Not available commercially PO; intravenous
Benserazide/Levodopa Available in Europe PO; (tablets cannot be
and Canada as tablets crushed)
(Madopar®, Prolopa®)
Carbidopa/Levodopa Tablets as IR (Sinemet®), PO; crushed/dissolved
ER (Sinemet CR®) or Sinemet® IR tablets can be
combination (RytaryTM); given by NGT; Parcopa®
orally disintegrating tablets can be given sublingually;
(Parcopa®); enteral DuopaTM can be given
suspension gel (DuodopaTM) enterally
Levodopa/Carbidopa/ Tablets (Stalevo®) PO; (tablets cannot be
144
Entacapone crushed)
Lisuride Available in Europe and PO; intravenous;
China as Dopergin® subcutaneous
Pramipexole IR and ER tablets (Mirapex®, PO; crushed/dissolved
Mirapex ER®) pramipexole IR tablets can be
given by NGT
Ropinirole IR and ER tablets (Requip®, PO; crushed/dissolved
Requip XL®) ropinirole IR tablets can be
given by NGT
Rotigotine Patch (Neupro®) Transdermal
Rasagiline Tablets (Azilect®) PO; crushed/dissolved
rasagiline tablets can be given
by NGT
Selegiline Capsules, tablets (Eldepryl®) PO; crushed/dissolved
Orally disintegrating tablet selegiline tablets can be given
(Zelapar®) by NGT; Zelapar® is given
sublingually; Emsam® is given
Patch (Emsam®)
transdermally
PO = by mouth; NGT = nasogastric tube; IR = immediate release; ER = extended release.
any individual drug before adding it to a regimen or to consider discontinuing

or switching an MAO-B inhibitor to another treatment.
Acute Parkinsonism
Parkinsonism can occasionally develop rapidly over hours to days, resulting in
admission to hospital and occasionally the ICU. This is a relatively infrequent
Table 14.2 Acute Parkinsonism
Mechanism Etiology
Drug- induced Amiodarone; carbamazepine; chemotherapeutic and cytotoxic
agents (including cyclosporin); dopamine receptor antagonists
(including all neuroleptics and many antiemetics); lamotrigine;
lithium; melatonin; phenytoin; valproic acid; tricyclic antidepressants
Genetic Rapid onset dystonia-parkinsonism
Acute decompensating Wilson’s disease
Infectious Coxsackie; Epstein Barr Virus; Human immunodeficiency virus;
influenza; Japanese B encephalitis; Murray Valley encephalitis;
poliovirus; post measles; western equine encephalitis; lymphocytic
choriomeningitis
Autoimmune Antiphospholipid syndrome; systemic lupus erythematous
Poisoning /Toxins Carbon monoxide; cadmium; cyanide; disulfiram; ethanol withdrawal;
Chapter 14
methanol; manganese; organophosphates; designer drugs
Metabolic Central pontine myelinosis; hepatic encephalopathy
Structural Ischemic stroke; obstructive hydrocephalus; subdural hematoma;
tumor
clinical situation, and dopamine receptor antagonist drugs are by far the most
145
common cause. In any patient that develops parkinsonism, especially acutely,
the most important clinical investigation is the medication and toxin history;
this should be elicited completely from patient, family, caregivers, and phar-
macy records scrutinized carefully for possible causes (drugs known to pro-
voke parkinsonism are listed in Table 14.2). There are several other causes
of acute parkinsonism that should be considered once drugs have been ex-
cluded (Table 14.2). Treatment should be targeted at the underlying etiologic
mechanism, but symptomatic management with dopaminergic medications
can also be considered.
Psychosis in PD
Up to half of PD patients develop psychosis at some point during the disease,
and it is one of the most common triggers for hospital admission and nursing
home placement. The clinical spectrum includes misidentification of visual
stimuli, visual hallucinations, delusions, and confusion or agitation.
As with many chronic neurological diseases, patients with PD are espe-
cially at risk for development of encephalopathy during intercurrent illness.
The acute development of psychotic symptoms in a PD patient should al-
ways prompt a thorough search for infection, metabolic abnormalities, or
superimposed neurological conditions such as seizure, stroke, or subdural
hematoma resulting from a fall, and a review of recent medication changes
that may have provoked symptoms. In the absence of an identifiable cause for
psychotic symptoms, consideration should be given to whether the symptoms
are caused by long-term medications or attributable to PD itself. Although
all medications used to treat PD can provoke hallucinations in susceptible
subjects, the worst offenders tend to be anticholinergics, followed by amanta-
dine, MAO-B inhibitors, dopamine agonists, COMT inhibitors, and levodopa.

These medications should be slowly withdrawn in the order listed; it is often
most effective to manage PD symptoms in patients with psychosis using levo-
dopa/carbidopa alone.
Delusions and hallucinations without insight, and which do not respond
to rationalization of the medication regimen as indicated previously, can be
suppressed in some patients using small doses of antipsychotic medications
under supervision. Most antipsychotic medications are contraindicated in this
setting because of the high risk of worsening motor symptoms and the possi-
SECTION 1
bility of NMS. However, clozapine or quetiapine are moderately effective and

relatively safe. Quetiapine is preferred since monitoring tests are unnecessary;
the starting dose in this setting should be no more than 12.5 mg daily, but the
medication can often be titrated slowly to 50 mg a day, given either at night or in
divided doses. Clozapine can also be effective at 6.25 to 50 mg daily (this dose
is much lower than the doses often used to treat schizophrenia) but requires
regular blood monitoring to detect agranulocytosis, a well-recognized adverse
effect of clozapine therapy. More recently, pimavanserin has been approved
by the Food and Drug Administration, specifically for treatment of psychosis
in PD. It is given as a single dose of 34 mg daily and in clinical trials produced
no worsening of motor function at doses that were efficacious for improving
146
hallucinations and delusions. Like quetiapine and clozapine, pimavanserin has

been associated with QT prolongation, and care should be taken when intro-
ducing these medications, particularly in patients with comorbidities or on the
presence of polypharmacy.
Hyperkinetic Movement Disorders –Chorea

Chorea is characterized by involuntary movements of short duration, irregular
frequency, and pseudo-random distribution. Movements are typically distal
and often also involve the face, oral, and lingual muscles. Especially when mild,
chorea is typically partially suppressible by the patient. Involuntary movements
can be incorporated into more purposeful movements and patients with
chorea often appear “fidgety” or “restless.”
Hemiballismus-Hemichorea
Hemiballismus-hemichorea is characterized by unilateral involuntary chorei-
form movements; there is a spectrum of abnormalities ranging from typical
chorea with predominately distal movements to ballismus, which describes
forceful, high-amplitude flailing movements caused by involvement of more
proximal muscle groups. Although uncommon, this is a dramatic clinical pres
entation that is not easily forgotten once seen; in severe cases, movements can
be continuous and violent such that injury is a recognized complication. It was
previously thought that hemiballismus-hemichorea was pathognomonic of
a lesion in the contralateral subthalamic nucleus (STN); however, STN lesions

were responsible for less than half of cases in recent series, and it is now
recognized that damage to other parts of the basal ganglia or contralateral
cortex can provoke the same clinical presentation. Cerebral infarction is the
most common recognized etiology of hemiballismus-hemichorea, followed by
non-ketotic hyperosmolar hyperglycemia. Multiple sclerosis, toxoplasmosis
and anticardiolipin syndrome can also cause a similar clinical presentation.
Hemiballismus- hemichorea patients usually require hospitalization for in-
vestigation and symptom control to prevent exhaustion or injury. Routine bio-
chemical studies and brain imaging often indicate the cause. In cases related to
hyperglycemia, symptoms frequently resolve within a few days of the patient’s
return to normoglycemia, although milder symptoms can persist beyond three
months. The movements may take much longer to resolve when caused by
other etiologies, although there is a good chance of remission by three months,
Chapter 14
and it may be possible to discontinue drugs without recurrence. There are no
randomized treatment trials for management of acute hemiballismus, and case
reports must be interpreted in light of the relatively high rate of spontaneous
resolution. Where movements are uncomfortable or violent, treatment is usu-
ally initiated with a neuroleptic agent such as haloperidol, with a starting dose of
0.5 to 1 mg twice a day, titrating upwards as needed. In emergency situations,
haloperidol may be given by intravenous injection at a dose of 1 to 4 mg every
147
four hours. If ongoing treatment is required beyond a few days, dopamine de-
pletion with tetrabenazine may be preferable for long-term suppression of
hemichorea, because of the lower risk for development of tardive dyskinesia.
Tetrabenazine is started at 12.5 mg three times a day and titrated slowly. If there
is no response at 50 mg a day, it is recommended that patients are genotyped
for CYP2D6 to determine if they are a poor metabolizer prior to further dose
increment. Doses up to 150 to 200 mg per day are sometimes required to
suppress movements; adverse effects such as hypotension or depression can
be limiting. In patients who do not respond to medication, stereotactic surgery
can be considered, although experience with this approach for hemiballismus-
hemichorea is currently limited.
Autoimmune Orobuccolingual Dyskinesia
Although uncommon, encephalitis associated with antibodies directed against
NMDA receptors can present as an emergency with a prominent movement
disorder. The typical history involves a prodromal illness with headaches and
flu-like symptoms, followed by agitation, paranoia, psychosis, or seizures.
Orobuccolingual dyskinesia is a characteristic feature of this condition. The
movements may include bruxism, grimacing, tongue protrusion and rolling,
nares flaring, palatal elevation, oculogyric crisis, smiling-like movements, and
forceful jaw opening and closing, severe enough to cause local injuries. In the
later stages of the disease, typical chorea, ballismus, and opisthotonic postures
have also been observed. Patients with advanced disease are frequently ad-
mitted to the ICU because of autonomic dysfunction, hypoventilation, or loss
of consciousness. The syndrome is readily recognized in its characteristic form
but should be considered in other patients with a subset of these symptoms.

The presence of serum NMDA receptor antibodies is diagnostic.
The importance of accurate diagnosis in this condition is twofold. In ap-
proximately 60% patients, this is a paraneoplastic autoimmune condition
(often associated with ovarian teratoma in young women) and the neurolog-
ical syndrome may improve with treatment of the malignancy. Consequently, a
thorough search for malignancy, including appropriate imaging studies, should
be carried out in patients with this condition. Second, even in cases not as-
sociated with malignancy, immunotherapy may be effective. Corticosteroids,
SECTION 1
intravenous immunoglobulin, and plasmapheresis are first-line treatments;

rituximab and cyclophosphamide have been employed in refractory cases
(discussed in more detail in Chapter 11).
There are two important caveats in the diagnosis and management of NMDA
receptor antibody encephalitis. First, since patients usually present initially with
neuropsychiatric disturbances, many are already taking neuroleptic medications
at time of admission. The orobuccolingual dyskinesia in these patients can be
misinterpreted as tardive dyskinesia, and the hyperthermia, elevated creatinine
kinase, and rigidity can be ascribed incorrectly to neuroleptic malignant syn-
drome. Second, patients frequently have both dyskinesia and seizures causing
involuntary movements; EEG monitoring is often necessary to distinguish these
148
possibilities to ensure that antiepileptic medications are deployed appropriately.
Hyperkinetic Movement Disorders—Myoclonus

Myoclonus describes rapid, brief, shock- like involuntary movements that
typically arise within the central nervous system. Myoclonus may be posi-
tive (caused by sudden muscle contraction) or negative (caused by sudden
disruption of ongoing contraction of postural muscles, sometimes also called
asterixis) and is usually not suppressible by the patient.
Serotonin Syndrome
Serotonin syndrome (SS) describes a constellation of symptoms and signs
caused by serotoninergic drugs. The diagnostic criteria include agitation,
tremor, hyperthermia, dysautonomia, and hypertonia/hyperreflexia/clonus
(Box 14.2). The characteristic clinical syndrome also includes prominent my-
oclonus, although this is not required for diagnosis. SS is an iatrogenic condi-
tion; there are multiple medications and combinations that can provoke this
syndrome (Table 14.3). The greatest risk for SS appears to be the combination
of SSRI antidepressants with MAO inhibitors. Diagnosis is clinical, but ancil-
lary tests are frequently employed to exclude other diagnostic possibilities,
for example EEG to exclude seizures in disoriented patients with myoclonus.
The diagnosis should be considered in any patient with a history of expo-
sure to relevant medications, presenting with confusion, autonomic findings,
hyperreflexia, or increased tone. The majority of cases show substantial
Box 14.2 Diagnosis of Serotonin Syndrome
• History of exposure to a relevant drug or other chemical agent
• Satisfies at least one of:
• Spontaneous clonus
• Inducible clonus + agitation or diaphoresis
• Tremor + hyperreflexia
• Hypertonia + temperature above 38oC + inducible clonus
Dunkley EJ, Isbister GK, Sibbritt D, Dawson AH, Whyte IM. The Hunter Serotonin Toxicity
Criteria: simple and accurate diagnostic decision rules for serotonin toxicity. QJM. 2003;96(9):635–642.
improvement within 24 hours of discontinuing the offending drug(s), which is

the most critical part of management. Supportive treatment is also necessary;
Chapter 14
in severe cases, sedation with benzodiazepines and even intubation might be
required. There are some data suggesting that serotonin antagonists may be
helpful. Cyproheptadine has been used as an initial 12 mg dose (oral or via na-
sogastric tube) followed by an additional 2 mg every hour if symptoms persist
and a maintenance dose of 8 mg four times a day, once the patient is stabilized.
Table 14.3 Drugs Implicated in Serotonin Syndrome
149
Mechanism Drugs
5HT receptor agonist Anticonvulsants: carbamazepine, valproic acid
Antidepressants: mirtazapine, trazodone
Opiates: fentanyl, meperidine
Hallucinogen: LSD
Mood stabilizer: lithium
Inhibit 5HT Anxiolytics: buspirone
metabolism Herbal supplements: St. John’s wort, Syrian rue
MAOIs: methylene blue, phenelzine, rasagilline, selegiline
Stimulate 5HT release Amphetamine
Antidepressants: mirtazapine
Weight loss drugs: phentermine
Certain opiates: meperidine, oxycodone
Over-the-counter cold remedies: dextromethorphan
Parkinson disease: Levodopa
Inhibit 5HT reuptake Amphetamine, cocaine, MDMA (Ecstasy)
Antiemetics: granisetron, ondansetron
Antihistamines: chlorpheniramine
Atypical Antidepressants: bupropion, trazodone
Opiates: levomethorphan, methadone, pethidine, tramadol
SNRIs: venlafaxine, duloxetine, venlafaxine
SSRIs: citalopram, escitalopram, fluoxetine, paroxetine, sertraline
TCAs: amitriptyline, desipramine, doxepin, nortriptyline
MAOIs = monoamine oxidase inhibitors; SNRI = serotonin and norepinephrine reuptake inhibitor;
SSRI = selective serotonin reuptake inhibitor; TCA = tricyclic antidepressant.
Chlorpromazine, 50 to 100 mg intramuscularly, can be considered if oral or
nasogastric administration of medication is not possible.

Post-hypoxic Myoclonus
Cerebral anoxia due to cardiac arrest can cause both acute myoclonic status
epilepticus and chronic post-hypoxic myoclonus.
Myoclonic status epilepticus manifests clinically as generalized myoclonus in
deeply comatose patients, usually within 12 to 24 hours after return of spon-
taneous circulation and is generally considered indicative of a poor prognosis
for survival (although some recent reports present more favorable outcomes
SECTION 1
with aggressive treatment). While phenytoin is seldom helpful, some benefits

have been reported with valproic acid, levetiracetam, and continuous infusions
of GABA-A agonists.
Note to editor-the A after GABA should be subscript and no dash between
GABA and the subscript A. Both the EEG findings (burst suppressions and
continuous generalized epileptiform discharges) and myoclonic movements
can be suppressed by the infusion of propofol and other anesthetics, although
there is no strong evidence that this improves the poor prognosis for recovery
(see Chapter 14).
Chronic post-hypoxic myoclonus (aka Lance Adams Syndrome) can occur
within a few days of return of spontaneous circulation but is typically a late com-
150
plication following recovery from anoxic brain injury. The clinical picture charac-
teristically includes action and stimulus-sensitive myoclonus in an awake patient.
Cognitive function may be preserved. The involuntary movements may resolve
over several years but if persistent, they are often difficult to control medically.
Myoclonus in Other Conditions
Prominent positive and negative myoclonus is commonly seen in hospital and
ICU patients with a variety of systemic conditions, often in the setting of enceph-
alopathy. The most common associations include renal failure, hepatic failure,
hypercapnia, and sepsis. Some medications also characteristically provoke myoc-
lonus, including SSRIs and gabapentin. EEG is frequently employed in this setting
to ensure that the combination of altered conscious level and myoclonus are not
caused by seizures. Usually, symptomatic treatment to suppress the movements
is unnecessary and therapy is directed at the underlying cause, such as addressing
uremia, or discontinuing offending medications, following which myoclonus usu-
ally resolves quickly. Oral clonazepam 0.5 mg can be considered if the movements
are uncomfortable or otherwise bothering the patient.
Hyperkinetic Movement Disorders—Dystonia

Dystonia refers to involuntary sustained muscle contractions that produce ab-
normal postures and twisting movements. Primary dystonia refers to diseases
in which dystonia is the sole or predominant manifestation and there is no
focal pathology or other extrinsic cause (which define secondary dystonia).
Many types of primary dystonia are genetically determined.
Dystonic Storm (Status Dystonicus)

Dystonic storm refers to a state of unremitting severe dystonia. This can
occur in patients with primary or secondary forms of dystonia, precipitated
by infection, medication changes, trauma, or abrupt failure of deep brain
stimulation (DBS) hardware. Patients are usually admitted to ICU for sup-
portive treatment including fluid and electrolyte balance. Treatment may also
include antimuscarinic medications, baclofen, or benzodiazepines, but these
are often ineffective and so sedation, airway protection, and even paralysis
are sometimes necessary to relieve the movements. Episodes may take days
to weeks to respond to treatment. Bilateral DBS of the globus pallidus in-
terna is now considered a treatment of choice for severe refractory forms
of dystonia.
Acute Dystonic Reactions Secondary to Drugs
Chapter 14
Drugs reactions are the most common cause of acute focal dystonia. If
maintenance of the airway or respiratory function is compromised, this can
be life-threatening. Acute dystonia usually appears within 24 hours following
drug exposure, and it often manifests with predominant orofacial dystonia.
Common presentations include tongue protrusion, oculogyric crisis, blepha-
rospasm, torticollis, trismus, and dysarthria. Neuroleptics and dopamine re-
ceptor antagonist antiemetics are the most commonly implicated drugs, but
151
SSRIs, tricyclic antidepressants, carbamazepine, phenytoin, or cocaine have
also been reported to provoke acute dystonia. Acute dystonic reactions are
usually self-limiting but when distressing they can frequently be reversed by
1 to 2 mg of benztropine given as an intravenous or intramuscular injection
and repeated 20 minutes later if needed. A course of oral benztropine ta-
pering over a week may be required to prevent the dystonia from returning.
Conclusions
Movement disorder emergencies can be associated with dramatic clinical
presentations that sometimes appear confusing or overwhelming. Diagnosis
is nearly always based on careful appraisal of the history and examination
findings, and is frequently relatively straightforward. Rapid clinical recognition
of these conditions in the emergency room and ICU often results in effective
treatment and a good prognosis for recovery, so that familiarity with diagnosis
and management of common movement disorders emergencies is both im-
portant and useful.
Further Reading
Artusi CA, Merola A, Espay AJ, et al. Parkinsonism-hyperpyrexia syndrome and deep
brain stimulation. J Neurol. 2015;262(12):2780–2782.
Boyer EW, Shannon M. The serotonin syndrome. New Engl J Med. 2005;352(11):
1112–1120.
Dalmau J, Lancaster E, Martinez-Hernandez E, Rosenfeld MR, Balice-Gordon R. Clinical
experience and laboratory investigations in patients with anti-NMDAR encephalitis.

Lancet Neurol. 2011;10(1):63–74.
Dunkley EJ, Isbister GK, Sibbritt D, Dawson AH, Whyte IM. The Hunter Serotonin
Toxicity Criteria: simple and accurate diagnostic decision rules for serotonin toxicity.
QJM. 2003;96(9):635–642.
English WA, Giffin NJ, Nolan JP. Myoclonus after cardiac arrest: pitfalls in diagnosis and
prognosis. Anaesthesia. 2009;64(8):908–911.
Gurrera RJ, Caroff SN, Cohen A, et al. An international consensus study of neuroleptic
malignant syndrome diagnostic criteria using the Delphi method. J Clin Psychiatry.
2011;72(9):1222–1228.
SECTION 1
Huddleston DES. Parkinsonism- hyperpyrexia syndrome in Parkinson’s disease.

In: Frucht SJ, ed. Movement Disorder Emergencies: Diagnosis and Treatment, 2nd ed.
New York: Humana Press; 2013:29 -42.
Friedman JH, Factor SA. Atypical antipsychotics in the treatment of drug-induced psy-
chosis in Parkinson’s disease. Move Disord. 2000;15(2):201–211.
Krauss JK, Pohle T, Borremans JJ. Hemichorea and hemiballism associated with contralat-
eral hemiparesis and ipsilateral basal ganglia lesions. Move Disord. 1999;14(3):497–501.
Munhoz RP, Teive HA, Eleftherohorinou H, Coin LJ, Lees AJ, Silveira-Moriyama L.
Demographic and motor features associated with the occurrence of neuropsychi-
atric and sleep complications of Parkinson’s disease. J Neurol Neurosurg Psychiatry.
2013;84(8):883–887.
152
Postuma RB, Lang AE. Hemiballism: revisiting a classic disorder. Lancet Neurol.
2003;2(11):661–668.
Ristic A, Marinkovic J, Dragasevic N, Stanisavljevic D, Kostic V. Long-term prognosis of
vascular hemiballismus. Stroke. 2002;33(8):2109–2111.
Section 2
Interventions and
Monitoring
Chapter 15
Multimodality Monitoring
Maranatha Ayodele and Kristine O’Phelan
Introduction
Advancements in in the critical care of patients with neurological conditions
recognizes that in addition to the initial insult, there is a secondary cascade of
physiological events that contribute to morbidity and mortality. Multimodality
monitoring (MMM) in neurocritical care aims to help recognize this process,
and then help tailor management to prevent or minimize any secondary injury.
This chapter will discuss a variety of invasive and non-invasive MMM techniques
aimed at monitoring brain physiologic parameters (Table 15.1), and offer a
practical guide to their integration and use in the intensive care setting.
155
The Basics: Intracranial Pressure and
Cerebral Perfusion Pressure
Intracranial pressure (ICP) monitoring allows for determination of cere-
bral perfusion pressure (CPP). These parameters are intrinsically linked to-
gether with mean arterial pressure (MAP), and this relationship is detailed
by the equation CPP = MAP –ICP. ICP and CPP are arguably the most
commonly assessed physiologic parameters in brain-injured patients. Per the
equation, knowledge of ICP and MAP can allow for the calculation of CPP.
Table 15.1 Multimodal Monitors

Physiologic Intracranial Brain Tissue Cerebral Brain Electrical
Parameter and Cerebral Oxygenation Blood Flow Metabolism Activity
Perfusion
Pressures
Invasive EVD Jugular bulb Thermal Cerebral EEG depth
Micro oximetry dilution probe microdialysis electrode
transducers Brain tissue Laser-Doppler eCOG
oxygen tension Flowmetry
Noninvasive Pupillometry NIRS TCD cEEG
ONSD qEEG
EVD = external ventricular drain; EEG = electroencephalography; eCOG = electocorticography;
NIRS = near-infrared spectroscopy; ONSD = optic nerve sheath diameter (via ultrasound);
TCD = transcranial doppler ultrasonography; cEEG = continuous electroencephalography;
qEEG = quantitative electroencephalography.
Optimization of ICP and CPP remains an essential component in the critical
Interventions and Monitoring
care management of neurocritical care patients. Invasive and noninvasive

methods of assessing ICP exist. The most commonly used invasive tech-
nique involves the use of either intraventricular fluid–coupled transducers
(considered by many to be the gold standard) or solid-state microtranducers.
Invasive Techniques
External Ventricular Drain
An external ventricular drain (EVD) is a fluid-coupled transducer that can be
placed via burr hole into the lateral ventricle with the tip positioned at the level
of the foramen of Monroe. The internal catheter is connected to an external
pressure transducer. This pressure transducer must be zeroed to the level of
Section 2
the foramen of Monroe using the tragus of the ear as an external landmark.
EVD use has advantages and disadvantages.
Advantages:
• It allows for the continuous monitoring of ICP when not draining cerebro-
spinal fluid (CSF)
• Analyses of ICP waveforms provide information about brain compliance
• Elevations in ICP may be alleviated by draining CSF
• Intrathecal medications may be administered through the catheter
156
Disadvantages:
• Requires placement through the brain parenchyma, and this may lead to
• Hemorrhage—along the EVD tract or into the ventricle, especially in
coagulopathic patients
• Infection
• There may be CSF over drainage leading to intracranial hypotension and
development of subdural hematomas
• The catheter may become occluded and prevent accurate measurements of ICP
• The fluid-filled system must be closed to accurately measure pressure
• Malposition may prevent accurate readings
Parenchymal Monitors
An alternative invasive method of ICP monitoring involves the use of
microtransducer devices. With these devices, the transducer is located at the
distal tip of the catheter and is composed of a fiberoptic cable, strain gauge
device, or pneumatic sensor that responds directly to pressure changes in
the tissue in which it is placed. These microtransducers are most commonly
placed intraparenchymally because this location yields the most accurate
data. However, they may be placed anywhere in the intracranial vault in-
cluding the epidural, subdural, subarachnoid, and intraventricular spaces.
Advantages:
• May be placed through smaller burr hole, and the small size of the catheter
allows for minimal trauma to brain tissue with fewer hemorrhagic and infec-
tious complications
• Allows for the continuous monitoring of ICP
• Analysis of ICP waveform provides information about brain compliance
Disadvantages:
• Elevations in ICP cannot be alleviated by draining CSF
• Calibration drift may occur over time as the device cannot be re-zeroed
once inserted
Noninvasive Techniques
Noninvasive methods of monitoring of ICP are much desired in neurocritical
Chapter 15
care as they provide alternatives for patients in which invasive methods may
be undesirable (i.e., ongoing coagulopathy) and are devoid of complications
such as intracranial hemorrhage and infection seen with invasive methods.
Pupillometry, the use of transcranial doppler ultrasonography (TCD) to de-
termine pulsatility index (PI), and ultrasound duplex imaging of the optic nerve
sheath diameter (ONSD) represent noninvasive methods of evaluating ICP,
though each has significant limitations.
Pupillometry
Pupillometry involves the use of a portable computer-based infrared digital
video device to assess pupillary size and reactivity. Changes in the pupillary
157
light reflex have long been observed to predict severity and prognosis in var-
ious forms of brain injury. The pupillometer device allows for a quantitative
assessment of this important reflex.
The pupillometer is placed over the patient’s eye and shines a fixed-intensity,
fixed-duration light stimulus into the eye. Pupillary images are captured in rapid
sequence. The device measures variables such as pupillary size at baseline and
after constriction, latency in milliseconds, and constriction/dilation velocity
in millimeters per second. Using these parameters, some devices can calcu-
late a score known as the Neurological Pupil Index (NPi). The NPi ranges
from 0 to 5 with scores above 3 considered normal. Several studies have ex-
amined the use of such quantitative pupillometry in patients with brain in-
jury and have found correlations predictive of elevations in ICP (Table 15.2).
Pupillometry use has its limitations in that the parameters measured may be
affected by medications, some of which are used routinely in the management
of neurocritical care patients. The device also cannot measure consensual pu-
pillary responses and cannot be used in patients with direct eye trauma, with
prior surgery to the iris, or who resist the exam with forced eye closure.
Ultrasound Modalities
Determination of PI and ONSD are two distinct ways in which ultrasound may
be used as a noninvasive determinant of ICP. PI is determined by using TCD to
insonate the middle cerebral arteries (MCA) via thin temporal bone windows.
PI is calculated as systolic flow velocity minus diastolic flow velocity divided by
mean flow velocity. While Bellner et al. found a reasonably strong correlation
between PI and invasively determined ICP, Behrens et al. and Morgalla and
Table 15.2 Pupillometry Parameters

Constriction CV <0.6 mm/sec—increased likelihood of elevated ICP in
velocity patients with mass effect, may precede ICP elevation up to 30
minutes
• Affected by medications:
• Decreased by opiates, barbiturates, and benzodiazepines in a
symmetric fashion
Neurological Pupil NPi <3—associated with increased ICP
Index • Decreasing trend of NPi values—associated with increasing
ICP and may precede ICP increase by several hours
• Affected by medications:
• Decreased by propofol (general anesthetic levels) and
Section 2
barbiturates
• Increased by dexmedetomidine
CV = constriction velocity; ICP = intracranial pressure; NPi = Neurological Pupil Index.
Magunia in their respective studies found PI to be an unreliable determinant

of ICP. In addition to imprecision, PI determination also requires trained per-
sonnel to perform the measurements, and up to 20% of patients may not have
adequate temporal bone windows to allow for insonation of the MCAs.
ONSD is measured by ultrasound duplex imaging of the optic nerve over
158
the patient’s closed eyelid. The optic nerve sheath is a continuation of the
brain’s dura matter, and its contents are contiguous with the intracranial sub-
arachnoid space. Thus, elevations in ICP may be reflected as increases in
ONSD. Measurements of ONSD have been shown to be especially valuable
in distinguishing normal versus raised ICP (ICP >20 mm Hg). Rajajee et al. and
Dubourg et al. both found an ONSD >5 mm to be predictive of increased
ICP. Additionally, in a recent large systematic review and meta-analysis by
Ohle et al. this value had a sensitivity of 95.6% and a specificity of 92.3% for
predicting elevated ICP.
Cerebral Oxygenation
Cerebral oxygenation may be monitored invasively and noninvasively. Invasive
techniques currently involve measuring jugular venous oxygen saturation
(SjVO2) and brain oxygen tissue tension (PbtO2). These two techniques allow
for global and regional determination of cerebral oxygenation, respectively.
Invasive Techniques
Jugular Bulb Oximetry
SjVO2 involves retrograde cannulation of the internal jugular vein. A small
fiber-optic catheter is inserted and positioned into the jugular bulb at the
base of the skull. Appropriate positioning at the base of the skull around the
level of the C1 vertebrae is verified on a lateral skull X-ray. Once calibrated,
SjVO2 measurements reflect the oxygen saturation of blood returning from
the brain. This represents the balance between oxygen delivery and utilization
and allows for estimation of global oxygen consumption. Normal SjVO2 values
are between 60% and 85%. SjVO2 desaturations (i.e., values less than 50%) are
concerning for global cerebral ischemia and were found by Robertson et al. to
be associated with poor outcomes in patients with severe brain injury. SjVO2
values greater than 85% are also problematic and may indicate an excess of
oxygen delivery relative to consumption.
Routine use of SjVO2 monitoring in the neurointensive care unit is challenged
by the need for frequent recalibration of the catheter once inserted. Additionally,
there are complications of catheter misplacement, infection, and risk of venous
Chapter 15
thrombosis around the catheter propagating intracranially or obstructing venous
outflow. This technique is also criticized for its low accuracy to detect regional is-
chemia. Thus, it may be best when used in patients with widespread brain injury
or in conjunction with a regional monitor such as PbtO2.
Brain Tissue Oxygen Monitoring
PbtO2, or the partial pressure of oxygen in brain tissue, is determined via
burr-hole insertion of an intraparenchymal oxygen probe into the subcor-
tical white matter of the brain. Current commercially available probes include
Licox (Integra Neurosciences) and Neurovent-PTO (Raumedic, Inc). Normal
PbtO2 measurements are usually 25 to 30 mm Hg. Values less than 15 to
159
20 mm Hg are concerning for local tissue ischemia and are the typically used
thresholds for initiating interventions. PbtO2 values are especially used in
conjunction with ICP and CPP monitoring. Optimization of CPP to alleviate
ischemia and treatment of elevated ICP are guided by PbtO2 values in this set-
ting (see Table 15.3 for clinical algorithm). Observational studies have shown
favorable outcomes for patients with subarachnoid hemorrhage (SAH) and
severe traumatic brain injury (TBI) treated with a combination of PbtO2 and
ICP/CPP guided therapy over ICP/CPP–based therapy alone. PbtO2 values
are greatly influenced by the location in which the probe is placed. The probe
provides regional measurement of PbtO2 for an area reflecting only about
14 mm3 of tissue. If placed in contusional tissue or if a hematoma develops
around the probe site, values may be greatly reduced and resistant to ther-
apeutic interventions, giving a false sense of the degree of brain ischemia.
Alternatively, if placed in normal-appearing tissue remote from focal brain in-
jury, values may not reflect the degree of regional ischemia present. These are
the limitations of this technology at this stage.
Near-Infrared Spectroscopy
Noninvasive assessment of cerebral oxygenation is possible using near-
infrared spectroscopy (NIRS). NIRS involves placement of probes emitting
near-infrared light (~700–1000 nm) along the forehead. The emitted near-
infrared light penetrates into the underlying brain tissue to an estimated
depth of about 2 to 3 cm, and the amount of light absorbed is used to
Table 15.3 Brain Tissue Oxygen Monitoring

Physiologic PbtO2 <20 mm PbtO2 <20 mm PbtO2 PbtO2
Parameter Hg Hg >20 mm Hg >20 mm Hg
ICP <20 mm Hg ICP >20 mm Hg ICP ICP <20 mm
>20 mm Hg Hg
Interventions Administer FiO2 Administer FiO2 Treat ICP No changes,
100% × 15 min to 100% × 15 min Follow continue
test the probe to test the probe PbtO2 for monitoring
↑ PaCO2 to Drain CSF changes
40–45 mm Optimize CPP
Hg range as
Administer
tolerated, carefully
fluids to
monitor both ICP
Section 2
euvolemia goal
and PbtO2
Give blood
Optimize CPP
products
Administer fluids for anemia
to euvolemia,
Administer
watch for signs
mannitol 0.25–
and symptoms of
0.5 gm/kg
fluid overload
Administer
Give blood
hypertonic saline
products
for anemia Optimize
160
sedation/
Cooling measures
analgesia
for brain
temp >37 ºC Consider
paralytics
Optimize
sedation/analgesia Cooling measures
for brain temp
Consider
>37 ºC
paralytics if
ventilator
dysynchrony
ICP = intracranial pressure; CPP = cerebral perfusion pressure; CSF = cerebrospinal fluid.
calculate cerebral oxygen saturation. This saturation is a combination of

the arteriovenous oxygen saturation of the underlying tissue. NIRS is more
commonly used in neonates who have thin skulls and open fontanelles, which
allow for better light penetration. In adults, NIRS is currently largely limited
to research uses as it has not been well validated clinically. Additionally, it
has limited reliability if the brain tissue in question is deep, there is overlying
fluid collection or hematoma, or if the overlying bone has been removed by
surgical craniectomy.
Cerebral Blood Flow Monitoring

Adequate cerebral blood flow (CBF) is essential for the delivery of substrates
that are necessary for neural tissues to maintain normal function. CBF that
is insufficient to meet metabolic demands can lead to tissue ischemia and, if
persistent, eventual infarction. Alterations in CBF have been described fol-
lowing TBI, high-grade SAH, and stroke. CBF may be measured directly or
indirectly, and invasive as well as noninvasive techniques exist.
Invasive Techniques
Thermal Diffusion and Laser-Doppler Flowmetry
Use of a Thermal Diffusion (TD) probe and Laser Doppler flowmetry (LDF)
represent invasive methods of obtaining continuous measurements of re-
gional CBF (rCBF). Currently, the only commercially available TD probe is
the QFlow 500 Probe (Hemedex, Inc, Cambridge, Massachusetts). Both TD
and LDF probes can be inserted into the brain parenchyma and are typically
Chapter 15
secured via a bolt system.
TD is currently the only invasive method that can offer direct quantitative
measurements of rCBF. The TD probe is a flexible, radio-opaque catheter
that contains two thermistors, an “active” one at the tip and a “passive”
one a short distance proximal to the tip. The passive thermistor meas-
ures a baseline tissue temperature. The active thermistor is then heated
above this baseline temperature. As blood flows through the capillary
bed around the active thermistor, the heat dissipates into the capillaries
surrounding the thermistor. The power required for the active thermistor
to maintain the surrounding tissue at the elevated temperature is directly
161
proportional to the CBF in the spherical volume of tissue surrounding the
probe. The TD probe can be connected to a monitor (Bowman Perfusion
Monitor®, Hemedex, Inc., Cambridge, Massachusetts) that then reports
quantitative rCBF in units of ml/100g/min.
LDF offers qualitative measurements of rCBF. The LDF probe is a fiberoptic
probe that emits a monochromatic laser light. The probe is connected to a
sensor that measures the movement of red blood cells based on the amount
of the Doppler shift of the reflected light and converts this into a voltage
signal. Fluctuations in this signal over time offer a qualitative measurement
of rCBF. Both TD and LDF allow for the continuous bedside monitoring of
rCBF. Such knowledge of rCBF may be used in real time to determine cere-
bral autoregulation status and thus optimize CPP in brain-injured patients as
described by Rosenthal et al.
TCD
Noninvasively, TCD is the most commonly used bedside technique, and it
can indirectly determine CBF. Additional noninvasive methods involve various
imaging techniques such as positron emission tomography, which is currently
considered the gold standard and is discussed elsewhere. TCD allows for in-
direct determination of CBF based on calculations of mean blood flow ve-
locity. Mean flow velocity in the vessel of interest is determined via insonation
through temporal or occipital windows. Mean flow velocity correlates well
with CBF as long as the angle of insonation and the diameter of the insonated
vessel remains constant (CBF = mean flow velocity × area of insonated vessel
× cosine of angle of insonation). In patients with SAH and cerebral vaso-
spasm, the mean flow velocity in the vessel must be examined in relation to
the extracranial portions to distinguish vasospasm from diffusely increased
CBF. Correlations have been shown between TCD-determined CBF and
CBF determined using gold-standard imaging techniques. TCD offers the ad-
vantage of being widely and readily available in most neurocritical care units
and allows for a more cross-sectional determination of CBF as anterior and
posterior circulation regions may be examined. However, it offers discrete
measurements in time and lacks the continuous monitoring possible with the
invasive methods mentioned earlier.
Cerebral Metabolism
Section 2
Invasive Techniques
Cerebral Microdialysis
Noninvasive imaging techniques such as magnetic resonance spectroscopy
can offer information at a single point in time about cerebral metabolism in
the injured brain. However, cerebral microdialysis (CMD) offers a unique,
bedside method for determining the brain’s extracellular chemical milieu on
an hourly basis. Thus, valuable information may be obtained and tracked re-
162
garding energy status, cellular integrity, and substrate delivery. CMD involves
intraparenchymal insertion of a small, flexible dialysis probe consisting of a
10 mm dialysis membrane and a catheter connected to a micro-infusion pump
circulating artificial CSF. Extracellular fluid is allowed to come into equilibrium
across the dialysis membrane. This technology allows for collection of a va-
riety of substances depending on their molecular weight and the dialysis mem-
brane pore size. Dialysate fluid is collected and placed in a bedside analyzer (M
Dialysis AB, Stockholm, Sweden), which measures the concentration of a va-
riety of analytes. The most commonly measured analytes used in clinical mon-
itoring are those relevant to energy metabolism and are glucose, pyruvate,
and lactate. Glycerol and glutamate have also been studied but are used less
commonly. Both the absolute values of these analytes as well as their trends
over time can be used to investigate the redox state of the local tissue. In gen-
eral, ischemia or cellular distress can cause elevations in the lactate to pyruvate
(L/P) ratio along with a decrease in the concentration of extracellular glucose.
Both low glucose and a high L/P ratio have been correlated with poor out-
come in patients with acute brain injury (see Table 15.4 for clinical algorithm).
Electrical Activity
Invasive Techniques
Intracranial monitoring of electrical activity is possible with electroencephalog-
raphy (EEG) depth electrodes and electocorticography (eCOG). eCOG is most
Table 15.4 Cerebral Metabolism
Physiologic Lactate/Pyruvate Ratio >40 Lactate/Pyruvate Ratio >40
Parameter Glucose <0.7 Glucose <0.7
ICP <20 mm Hg ICP >20 mm Hg
Interventions Optimize CPP Treat ICP
Raise PaCO2 to 40–45 mm Hg • Elevate head of bed
as tolerated by ICP • Drain CSF
Optimize: temperature, PaO2, • Osmotic therapy
hemoglobin, sedation, systemic
• Consider paralysis
glucose
Optimize: PaCO2, temperature,
Chapter 15
sedation, CPP, hemoglobin,
systemic glucose
ICP = intracranial pressure; CPP = cerebral perfusion pressure; CSF = cerebrospinal fluid.
commonly used for preoperative assessment and seizure foci localization in epi-
lepsy patients. EEG depth electrodes have been used in some centers for contin-
uous EEG monitoring in high-risk patients with acute brain injury and can detect
epileptiform activity that is not well seen on surface EEG. EEG depth electrodes
are placed through a standard twist drill burr hole. Electrocorticography grids
are placed along the surface of the brain via craniotomy. Their use as a contin-
163
uous bedside monitor is thus far limited to research studies such as COSBID,
where they are used to detect spreading cortical depression.
Continuous and quantitative electroencephalography with surface electrodes
are used to detect not only seizure activity but changes suggestive of under-
lying cellular injury in high-risk patients with acute brain injury. Their use is
discussed in detail in chapter 7.
Further Reading
Behrens A, Lenfeldt N, Ambarki K, Malm J, Eklund A, Koskinen LO. Transcranial
Doppler pulsatility index: not an accurate method to assess intracranial pressure.
Neurosurgery. 2010;66(6):1050–1057.
Bellner J, Romner B, Reinstrup P, Kristiansson KA, Ryding E, Brandt L. Transcranial
Doppler sonography pulsatility index (PI) reflects intracranial pressure (ICP). Surg
Neurol. 2004;62(1):45–51.
Bhatia A, Gupta AK. Neuromonitoring in the intensive care unit. I. Intracranial pressure
and cerebral blood flow monitoring. Intensive Care Med. 2007;33(7):1263–1271.
Chen JW, Gombart ZJ, Rogers S, Gardiner SK, Cecil S, Bullock RM. Pupillary reac-
tivity as an early indicator of increased intracranial pressure: the introduction of the
Neurological Pupil Index. Surg Neurol Int. 2011;2:82.
Dubourg J, Javouhey E, Geeraerts T, Messerer M, Kassai B. Ultrasonography of optic
nerve sheath diameter for detection of raised intracranial pressure: a systematic re-
view and meta-analysis. Intensive Care Med. 2011;37(7):1059–1068.
Ghosh A, Elwell C, Smith M. Review article: cerebral near-infrared spectroscopy in
adults: a work in progress. Anesth Analg. 2012;115(6):1373–1383.

Hutchinson P, O’Phelan K. International multidisciplinary consensus confer-
ence on multimodality monitoring: cerebral metabolism. Neurocrit Care.
2014;21(Suppl. 2):S148–S158.
Lauritzen M, Dreier JP, Fabricius M, Hartings JA, Graf R, Strong AJ. Clinical relevance of
cortical spreading depression in neurological disorders: migraine, malignant stroke,
subarachnoid and intracranial hemorrhage, and traumatic brain injury. J Cereb Blood
Flow Metab. 2011;31(1):17–35.
Morgalla MH, Magunia H. Noninvasive measurement of intracranial pressure via the
pulsatility index on transcranial doppler sonography: Is improvement possible? J Clin
Ultrasound. 2016;44(1):40–45.
Nangunoori R, Maloney-Wilensky E, Stiefel M, et al. Brain tissue oxygen-based therapy
Section 2
and outcome after severe traumatic brain injury: a systematic literature review.
Neurocrit Care. 2012;17(1):131–138.
Oddo M, Bosel J. Monitoring of brain and systemic oxygenation in neurocritical care
patients. Neurocrit Care. 2014;21(Suppl. 2):S103–S120.
Ohle R, McIsaac SM, Woo MY, Perry JJ. Sonography of the optic nerve sheath diameter
for detection of raised intracranial pressure compared to computed tomography: a
systematic review and meta-analysis. J Ultrasound Med. 2015;34(7):1285–1294.
Rajajee V, Vanaman M, Fletcher JJ, Jacobs TL. Optic nerve ultrasound for the detection
of raised intracranial pressure. Neurocrit Care. 2011;15(3):506–515.
164
Robertson CS, Gopinath SP, Goodman JC, Contant CF, Valadka AB, Narayan RK. SjvO2
monitoring in head-injured patients. J Neurotrauma. 1995;12(5):891–896.
Rosenthal G, Sanchez-Mejia RO, Phan N, Hemphill JC 3rd, Martin C, Manley GT.
Incorporating a parenchymal thermal diffusion cerebral blood flow probe in bedside
assessment of cerebral autoregulation and vasoreactivity in patients with severe trau-
matic brain injury. J Neurosurg. 2011;114(1):62–70.
Spiotta AM, Stiefel MF, Gracias VH, et al. Brain tissue oxygen- directed manage-
ment and outcome in patients with severe traumatic brain injury. J Neurosurg.
2010;113(3):571–580.
Steiner LA, Andrews PJ. Monitoring the injured brain: ICP and CBF. Br J Anaesth.
2006;97(1):26–38.
Taylor WR, Chen JW, Meltzer H, et al. Quantitative pupillometry, a new tech-
nology: normative data and preliminary observations in patients with acute head in-
jury. Technical note. J Neurosurg. 2003;98(1):205–213.
Wilson JA, Shutter LA, Hartings JA. COSBID-M3: a platform for multimodal mon-
itoring, data collection, and research in neurocritical care. Acta Neurochir Suppl.
2013;115:67–74.
Chapter 16
External Ventricular Drainage

Clinical Indications, Surgical Technique,
and Management
Nitin Agarwal and Andrew F. Ducruet
Introduction
Ventriculostomy refers to placement of a catheter into the ventricular system
in order to temporarily drain cerebrospinal fluid (CSF). This surgical proce-
dure may be performed in the setting of hydrocephalus or in patients suf-
fering from increased intracranial pressure (ICP) in a variety of neurological
165
conditions. This chapter reviews the indications, technique, and postoperative
management strategies for external ventricular drainage.
Historical Overview
The drainage of CSF dates back to 1500 BC, whereby the spillage of clear
fluid from the interior of the brain is noted in ancient Egyptian text. However,
Claude-Nicholas Le Cat clearly documented the first procedure of external
ventricular drainage. Carl Wernicke was credited by Keen with the inaugural
sterile ventricular puncture and external ventricular drain (EVD) placement
in 1881. In 1890, Keen published his own surgical technique. The entry site
he advocates, Keen’s point, is 3 cm superior and 3 cm posterior to the pinna.
Later, in his 1894 textbook, Kocher described his eponymous point. Adson
and Lillie augmented the EVD technique by adding manometry in 1927. Over
the years, EVD began to be used for ICP monitoring in patients with brain
tumors and subarachnoid hemorrhage. Most recently, research has focused
on quality improvement with regards to surgical training, infection control, and
image guidance systems.
Indications
The Monro-Kellie doctrine states that the volume of the cranial compartment,
composed of blood, brain tissue, and CSF, is fixed. An increase in any one
of these cranial constituents must therefore be balanced by a decrease of
the others. Displacement of venous blood and CSF serve as compensatory

mechanisms to a certain point. However, progressive expansion of a space-
occupying lesion, such as intracranial hematoma, inevitably leads to downward
displacement of brain tissue, which may result in herniation. Diversion of CSF
via ventriculostomy reduces the volume of CSF in the intracranial space in the
face of such an expanding mass lesion or elevated ICP, and serves as an essen-
tial therapeutic strategy for the management of several neurosurgical disorders.
Indications for ventriculostomy include the treatment of symptomatic hy-
drocephalus as well as elevated ICP caused by a variety of pathologies. An
EVD may also be necessary for ICP monitoring and treatment in the setting
of mass lesions such as subdural or epidural hematoma, tumor, or infarction
Section 2
with resulting edema. Cerebral edema may occur with liver failure, hyper-
tensive encephalopathy, and anoxic brain injury. Furthermore, obstruction of
CSF outflow secondary to mass lesion or ventricular hemorrhage resulting
in obstructive hydrocephalus is also frequently treated by ventriculostomy.
Additionally, increases in venous pressure that occur with venous sinus throm-
bosis or obstruction of the jugular veins may also lead to increases in ICP.
Decreases in cerebral perfusion pressure due to cerebral edema may also
warrant CSF diversion, particularly in the setting of increased ICP due to trau-
matic brain injury.
166
Lumbar drains represent an alternative to EVDs. For instance, with hydro-

cephalus secondary to subarachnoid hemorrhage (SAH) or intraventricular
hemorrhage (IVH), the placement of a lumbar drain may be indicated for the
temporary drainage of CSF. Moreover, some have advocated that lumbar CSF
drainage after aneurysmal SAH may help prevent cerebral vasospasm by pro-
moting CSF circulation and removing spasmogens. These benefits achieved
with lumbar drains may come at a reduced risk as compared to passing
catheters to the cerebral parenchyma. However, just as EVDs are associated
with risks elaborated upon later, so too are lumbar drains. In a study by Coplin
et al., 4.2% of 312 lumbar drains employed resulted in a case of bacterial men-
ingitis. Another common complication with lumbar drains is catheter breakage.
Fracture of a catheter may occur with excessive traction or shearing over a
Tuohy needle. Several factors in surgical technique may contribute to catheter
breakage and associated complications: faulty use of guidewire, retraction of
catheter through needle, or excessive force during catheter removal. Of note,
Guppy et al. presented a patient who developed SAH due to the rostral mi-
gration of a retained intrathecal lumbar catheter to the brain.
Surgical Technique
Following an appropriate informed consent of the patient or his or her rep-
resentative, a surgical time-out is performed. The nondominant right frontal
approach is generally utilized but may be modified based on known pathology.
The hair over the frontal region is clipped. A linear incision over Kocher’s

point is then prepped and the patient draped in a sterile fashion. Kocher’s
point is 11 cm posterior from nasion and 3 cm lateral from midline (Figure
16.1). Alternatively, the incision may be marked 1 cm anterior to coronal su-
ture in the midpupillary line. Ipsilateral medial canthus and ipsilateral tragus
should be marked as trajectory points. Given that misplaced catheters are
often too lateral or posterior, some advocate a more medial and anterior
trajectory and therefore marking the nasion and 1 cm anterior to the tragus.
Electrocardiogram leads may be placed at these locations to enhance trajec-
tory targeting by palpating through the drapes. Preoperative antibiotics, such
as cefazolin, are administered prior to skin incision. Conscious sedation and
local anesthetic are administered. A 1-to 2-cm incision is made in the antero-
Chapter 16
posterior direction. A self-retaining retractor is placed, and this may also pro-
mote hemostasis. Once stabilization of the head is achieved by an assistant or
mechanical device, a twist drill is utilized to create a burr hole perpendicular
to the skull in the planned trajectory for the EVD. The dura is then opened
Kocher’s Point
(Adults)
11–12 cm behind Mid Pupillary Line
167
nasion
3 cm off midline
Aim catheter toward
ipsilateral medial canthus
and ipsilateral tragus.
1 cm anterior
to coronal suture
Figure 16.1 Standard preoperative planning for a right frontal external ventricular drain
with appropriately demarcated Kocher’s point.
Source: Used with permission from the Neurosurgery Survival Guide.
utilizing the scalpel or trochar. The EVD is passed in the planned trajectory,
and the surgeon should be able to perceive tactile feedback when entering
into the ventricle. The catheter is advanced to a depth of 6.5 cm at the outer
table of bone. If spontaneous CSF flow is observed, the opening pressure
is noted, and the catheter is tunneled posteriorly and toward the midline.
If no CSF is obtained following three passes, a head computed tomography
(CT) scan should be obtained to check catheter position to assist with posi-
tioning for further placement attempts. The catheter should be secured at the
exit site with a roman sandal and strain release loop. The incision is then pri-
marily closed with a running nylon suture, and additional securing stitches are
placed along the strain release loop. The surgeon must check again for sponta-
neous CSF flow after closure of the incision, then connect the catheter to the
Section 2
drainage system bag and level the system to an appropriate height.
Management Strategy
Several general strategies are typically standardized with protocols following
ventriculostomy. With EVDs, the collection chamber is generally set at a spe-
cific height and left open, so that CSF drains when the ICP exceeds the set
level. ICP is transduced and measured on a frequent basis, for example every
168
hour, with a pressure transducer. As a general rule this transducer is leveled at

a height that corresponds with the external auditory meatus, approximating
the level of the foramen of Monroe, in the supine position. Given that the
collection system is gravity-dependent, all position changes must be accom-
panied by a change in EVD height. Drains are therefore typically clamped for
patient transport and bed transfer. CSF sampling is performed if meningitis or
ventriculitis is clinically suspected. Some institutions may obtain surveillance
sampling, but one must be aware that an increased frequency of sampling may
yield an increased infection risk.
The specific pathology serves as the basis for individualized EVD manage-
ment strategies. For example, a patient with tumors may require just enough
drainage to accommodate for the mass effect due to the space-occupying
lesion. Meanwhile, patients with IVH may require aggressive drainage in an
effort to clear the ventricles. Similarly, patients afflicted by conditions such
as low-pressure hydrocephalus may be drained at a lower level in order to
effectively decompress the ventricles. Two types of neurointensive care unit
patients regularly require EVD drainage and are discussed in more detail: head
trauma and subarachnoid hemorrhage. Following severe traumatic brain injury,
CSF diversion may be used during the period of maximal brain swelling. This
is typically performed with an initial drain height of 10 cmH20, which may be
lowered to the level of the midbrain to maximize CSF drainage as needed. In
contrast, in the setting of hydrocephalus secondary to aneurysmal subarach-
noid hemorrhage, the initial drain height must be set higher in order to prevent
rebleeding that may theoretically be associated with a transmural pressure
gradient across an unsecured aneurysm wall. As such, until the aneurysm is

secured, the initial drain height is set at 20 cmH20 and left open to drain, or
kept clamped and only opened in the setting of elevated ICP, depending on
institutional protocol. The EVD is then lowered after aneurysm treatment to
optimize CSF drainage. Regardless of the indication for ventriculostomy, man-
agement strategies following insertion are critical for appropriate CSF diver-
sion during acute and long-term management periods.
Weaning the Ventriculostomy

Following several days of drainage in the acute management period, the drain
Chapter 16
is typically weaned. The number of days of active drainage depends on the
pathology. For hydrocephalus related to SAH or IVH, EVD drainage generally
is continued throughout the vasospasm period or until the third and fourth
ventricles are cleared of blood. For obstructive hydrocephalus, EVD is generally
continued until the obstruction is relieved. There is wide institutional variation,
but EVD removal is generally accomplished after a brief period of weaning.
During this wean, the drain system is gradually raised and ultimately clamped. If
no clinical or radiographic evidence of hydrocephalus is noted during this wean,
the drain is removed. At our institution, we generally raise the EVD by 5 cm
169
each day to a level of 20 and if there is no change in the neurological exam,
the drain is clamped for 24 hours. We obtain a CT scan following 24 hours,
and if there is no change in ventricular size or neurological exam, the drain is
removed. In some cases, a more rapid wean or even an immediate clamp trial
are attempted, and there remains no strong evidence to support a rapid versus
a gradual wean. Klopfenstein et al. (2004) reported no advantage to a gradual,
multistep wean as compared to a rapid one with regards to preventing the
need for long-term shunt placement or prolonged intensive care unit and hos-
pital stays. If during the drain wean the patient becomes symptomatic, then the
wean is stopped, and this may indicate the need for long-term CSF diversion
with a shunt.
Complications
Misplacement
Given that EVDs are often placed in an emergent fashion by neurosurgical
trainees utilizing a freehand technique, misplacement of the ventricular cath-
eter can occur. Hsieh et al. noted 18 catheters misplaced among 129 EVD
placements. However, only 4 of these 18 required revision. While many ad-
vocate that freehand surgical technique for EVD placement remains an effi-
cacious method for urgent treatment, replacement may be required in up to
50% of free-hand cases. Thus Meral et al. have advocated for a novel device,
using transcranial ultrasound, to guide EVD placement.
Malfunction
Mechanical malfunction of a ventricular catheter occurs due to occlusion, mi-

gration, and kinking or discontinuity of the tubing. Patients with malfunctioning
catheters frequently present with symptoms of increased ICP, often in
association with radiographic findings of hydrocephalus, reduced CSF flow, or
poor ICP waveform. In cases of suspected EVD obstruction, the drain should
be inspected immediately to rule out kinking or disconnection. Catheters
clogged with blood or brain material may be flushed with sterile normal saline.
Generally this is carried out in a distal direction. If flushing is performed toward
the head through the proximal catheter, it must be done with extreme care to
avoid iatrogenic injury. When a clogged catheter cannot be flushed, the cath-
eter must be replaced. Head CT should be obtained immediately in all cases of
Section 2
suspected obstruction to evaluate ventricular size and to rule out new hema-
toma or catheter misplacement. A recent prospective study demonstrated that
of 98 patients, 41 (42%) developed at least one temporary occlusion requiring
irrigation and 19 developed at least one permanent occlusion (19%). This study
estimated the cost of occlusion at $615 per enrolled patient, and, therefore,
the authors advocated for the use of large-bore catheters to prevent occlusion.
Infection
Meningitis or ventriculitis is a rare but significant complication associated with
170
ventriculostomy. The incidence of EVD-associated meningitis or ventriculitis

ranges from 0% to 22%. Among the many factors associated with the risk
of EVD-associated infections are sterile technique, surgical site CSF leak,
tunneling of EVD catheter away from entry incision, and frequency of CSF
sampling. Strategies to prevent or reduce infections include the administration
of perioperative intravenous antibiotics and the use of antibiotic-impregnated
or silver-coated catheters. A recent study by Edwards et al. demonstrated in
100 patients requiring EVDs, the use of antibiotic-impregnated catheters may
be associated with 2.7 fewer deaths and 82 fewer hospital days due to infec-
tion. Furthermore, the decrease in infection using antibiotic-impregnated cath-
eter when placing an EVD resulted in an estimated $264,069 of net savings per
100 patients. In another study by Keong et al. the use of silver-impregnated
catheters during EVD placement resulted in a significantly decreased infection
rate (12.3%) as compared to that with plain catheters (21.4%). Given the sig-
nificant morbidity associated with meningitis or ventriculitis, many institutions
have adopted strict protocols for the insertion and management of EVDs to
minimize infection.
Hemorrhage
There remains significant risk of hemorrhage associated with placement of an
EVD. This risk is likely increased by the use of antiplatelet and anticoagulant
medications, and these should be addressed prior to placement. Catheter-
associated hemorrhage may cause obstruction of the EVD, and in some
cases these hematomas can cause mass effect and neurological symptoms.
We recommend a short-term interval follow-up scan following discovery of a

drain tract hemorrhage, particularly in comatose patients. The rate of EVD-
associated hemorrhage reported in the literature is highly variable, with rates
ranging from 9% to 33%. Some data has shown increased rates of hemorrhage
when catheters are replaced.
Neurological Deficits
Although rare, the possibility of iatrogenic neurological deficits must be
discussed during the consent process for an EVD. Neurological deficits
may also arise as a sequela of any of the aforementioned complications.
Fargen et al. reported that 2% of EVD placement associated hemorrhages
resulted in new neurological deficit. Chai et al. reported a case of coma
Chapter 16
induced by a misplaced EVD. They hypothesized their placement was too
deep resulting in mass effect onto the ventral periaqueductal grey matter
and ascending neurons from the brainstem, thereby affecting the reticular
activating system. Following adjustment of the EVD position, the patient
regained consciousness.
Conclusion
171
CSF diversion by placement of an EVD or lumbar drainage remains one
of the most important and frequently performed emergent neurosurgical
interventions. Indications for external ventricular drainage include a variety of
intracranial disorders causing symptomatic hydrocephalus or increased ICP.
Major complications include malfunction, infection, bleeding, and, rarely, neu-
rological deficits.
Further Reading
Brownlee RD, Dold ON, Myles ST. Intraventricular hemorrhage complicating ven-
tricular catheter revision: incidence and effect on shunt survival. Pediatr Neurosurg.
1995;22(6):315–320.
Chai FY, Farizal F, Jegan T. Coma due to malplaced external ventricular drain. Turk
Neurosurg. 2013;23(4):561–563.
Coplin WM, Avellino AM, Kim DK, Winn HR, Grady MS. Bacterial meningitis associated
with lumbar drains: a retrospective cohort study. J Neurol Neurosurg Psychiatry. 1999
Oct;67(4):468–473.
Dey M, Jaffe J, Stadnik A, Awad IA. External ventricular drainage for intraventricular
hemorrhage. Curr Neurol Neurosci Rep. 2012 Feb;12(1):24–33.
Fargen KM, Hoh BL, Neal D, O’Connor T, Rivera-Zengotita M, Murad GJ. The burden
and risk factors of ventriculostomy occlusion in a high-volume cerebrovascular prac-
tice: results of an ongoing prospective database. J Neurosurg. 2016;124(6):1805–1812.
Greenberg MS, Greenberg MS. Handbook of Neurosurgery, 7th ed. Tampa, FL: Greenberg
Graphics; 2010.
Guppy KH, Silverthorn JW, Akins PT. Subarachnoid hemorrhage due to retained lumbar
drain. J Neurosurg Spine. 2011 Dec;15(6):641–644.

Hsieh CT, Chen GJ, Ma HI, et al. The misplacement of external ventricular drain
by freehand method in emergent neurosurgery. Acta Neurolog Belgica. 2011
Mar;111(1):22–28.
Keong NC, Bulters DO, Richards HK, et al. The SILVER (Silver Impregnated Line
Versus EVD Randomized trial): a double-blind, prospective, randomized, controlled
trial of an intervention to reduce the rate of external ventricular drain infection.
Neurosurgery. 2012 Aug;71(2):394–403; discussion 403–394.
Klimo P Jr, Kestle JR, MacDonald JD, Schmidt RH. Marked reduction of cerebral vaso-
spasm with lumbar drainage of cerebrospinal fluid after subarachnoid hemorrhage.
J Neurosurg. 2004 Feb;100(2):215–224.
Klopfenstein JD, Kim LJ, Feiz-Erfan I, et al. Comparison of rapid and gradual weaning
Section 2
from external ventricular drainage in patients with aneurysmal subarachnoid hemor-

rhage: a prospective randomized trial. J Neurosurg. 2004 Feb;100(2):225–229.
Kompanje EJ, Delwel EJ. The first description of a device for repeated external ven-
tricular drainage in the treatment of congenital hydrocephalus, invented in 1744 by
Claude-Nicolas Le Cat. Pediatr Neurosurg. 2003 Jul;39(1):10–13.
Maniker AH, Vaynman AY, Karimi RJ, Sabit AO, Holland B. Hemorrhagic complications
of external ventricular drainage. Neurosurgery. 2006 Oct;59(4 Suppl. 2):ONS419–
424; discussion ONS424-415.
Meral FC, Persaud MA, Silva AE, et al. A novel device for guiding ventriculostomy with
172
transcranial ultrasound. J Acoust Soc Am. 2014;135(4):2211–2211.

Mokri B. The Monro-Kellie hypothesis: applications in CSF volume depletion. Neurology.
2001 Jun 26;56(12):1746–1748.
Muralidharan R. External ventricular drains: Management and complications. Surg Neurol
Int. 2015;6(Suppl. 6):S271–S274.
Olivar H, Bramhall JS, Rozet I, et al. Subarachnoid lumbar drains: a case series
of fractured catheters and a near miss. Can J Anaesthes = J can anesthes. 2007
Oct;54(10):829–834.
Srinivasan VM, O’Neill BR, Jho D, Whiting DM, Oh MY. The history of external ventric-
ular drainage. J Neurosurg. 2014 Jan;120(1):228–236.
Wiesmann M, Mayer TE. Intracranial bleeding rates associated with two methods of
external ventricular drainage. J Clin Neurosci. 2001 Mar;8(2):126–128.
Wilkins RH. Neurosurgical Classic. Xvii. J Neurosurg. 1964 Mar;21:240–244.
Yong CI, Hwang SK, Kim SH. The role of lumbar drainage to prevent shunt-dependent
hydrocephalus after coil embolization for aneurysmal subarachnoid hemorrhage in
good-grade patients. J Korean Neurosurg Soc. 2010 Dec;48(6):480–484.
Chapter 17
Neuroimaging and
Neurointerventional
Procedures
Cynthia L. Kenmuir and Ashutosh P. Jadhav
Introduction
Diagnostic neuroimaging and interventional neurology are commonly used
tools in the diagnosis and management of critically ill patients. This chapter is
divided in two parts addressing: 1) the imaging indications and findings associ-
ated with various disease states, and 2) procedural considerations in patients
173
undergoing catheter neuro-angiography.
Neuroimaging

Traumatic brain injury (TBI) is an alteration in brain function caused by ex-
ternal force. Neuroimaging is a crucial component of the initial assessment of
patients after a TBI to help guide clinical management.
Computed tomography head (CTH) without contrast is the initial
study of choice to assess for intracranial hemorrhage, mass effect, midline
shift, ventricular distortion, skull fractures, displaced bone fragments, foreign
bodies, and intracranial air. Advantages of CTH include its wide availability and
fast image collection, which minimizes image degradation in uncooperative
patients. Repeat CTH should be obtained with neurological decline but rou-
tine follow-up CTH is not necessary. Limitations include radiation exposure.
CT angiography (CTA) should be considered if there is penetrating
trauma, skull base fracture, or neck trauma. There is a higher likelihood of
vascular injury if there is cervical facet subluxation or dislocation, fracture lines
approaching an artery, or high-impact injury mechanisms. CTA can be useful
for identifying traumatic vascular injury (e.g., pseudo-aneurysm, dissection, or
uncontrolled hemorrhage).
Digital subtraction angiography (DSA) is the gold standard for
detecting vascular injury. Denver criteria can be used to describe blunt ce-
rebrovascular injury as follows: grade I-mild intimal injury or irregular intima,
grade II-dissection with raised intimal flap/intramural hematoma with luminal
narrowing >25%/intraluminal thrombosis, grade III-pseudoaneurysm, grade
IV-vessel occlusion, grade V-vessel transection.
Magnetic resonance imaging (MRI) for TBI may be considered for
suspected pathology involving the brainstem, posterior fossa, or brain paren-
chyma, if the CTH findings do not explain the clinical exam, and may assist
with prognostication. MRI T1-weighted imaging is useful for high resolution of
Section 2
anatomy and can help to detect mass effect, midline shift, and ventricular dis-
tortion. The use of contrast is rarely helpful. Gradient resonance echo (GRE)
is useful for detecting blood products (hemorrhagic diffuse axonal injury
[DAI], contusions). Susceptibility weighted imaging (SWI) is three to six times
more sensitive in detecting hemorrhage than GRE. Fluid attenuated inversion
recovery (FLAIR) is useful for nonhemorrhagic DAI and subarachnoid hemor-
rhage. Short tau inversion recovery (STIR) “suppresses” fat and is therefore
useful in differentiating blood or edema from fat as well as for identifying optic
nerve injury and vertebral body compression fractures. Diffusion weighted
174
imaging (DWI) is highly sensitive for DAI. Diffusion tensor imaging (DTI) can
be used to assess white matter integrity and connectivity. Limitations of MRI
for TBI include slow image acquisition (more useful in the subacute or chronic
setting), it is not widely available, and it has a higher cost.
Hemorrhagic Stroke
Hemorrhagic stroke accounts for 15% of strokes in adults. Hemorrhagic
stroke can be divided into categories based on the location of hemorrhage: in-
tracerebral (ICH) with or without intraventricular extension, subarachnoid
(SAH), subdural (SDH), and epidural (EDH).
ICH
The specific location of nontraumatic, nonaneurysmal ICH is often related
to the underlying mechanism. For instance, hypertensive hemorrhages most
often occur in the thalamus, basal ganglia, pons, and cerebellum, whereas ce-
rebral amyloid often causes lobar hemorrhage, and hemorrhagic conversion
of ischemic stroke will follow a vascular territory. CTH is highly sensitive for
acute blood and helps to identify hemorrhage size and location, presence
of edema, and hydrocephalus. Hematoma size can be estimated using the
“a x b x c/2” rule: measure “a” (width) and “b” (breadth) of hematoma on
the largest axial section; “c” (height) can be measured on coronal slice or
by counting the number of involved slices and multiplying by the slice thick-
ness. Additionally, vascular anomaly can be suggested on CTH by the pres-
ence of enlarged vessels or calcifications along the margins of the hemorrhage
or by hyperattenuation within a dural venous sinus or cortical vein along the
Neuroimaging & Neurointervention
Table 17.1 T1 and T2 Sequences to Determine the age of Blood
Products at the Time of MRI
Stage Time Hemoglobin T1 T2
Hyperacute <24 hours Oxyhemoglobin Iso Hyper
Acute 1–3 days Deoxyhemoglobin Iso Hypo
Early subacute 3–7 days Methemoglobin in RBCs Hyper Hypo
Late subacute >7 days Methemoglobin Free Hyper Hyper
Chronic >14 days Hemosiderin Iso, Hypo Hypo
MRI = magnetic resonance imaging; RBC = red blood cells.
presumed venous drainage path of the ICH. CTA can be acutely helpful in
identifying active extravasation (“spot sign”) and can identify underlying vas-
Chapter 17
cular pathology (aneurysm, arteriovenous malformation/fistula, venous sinus
thrombosis, vasculopathy, moyamoya). DSA is the gold standard for detecting
vascular malformations and should be especially considered for ICH in young
patients without a history of hypertension or impaired coagulation. MRI gra-
dient echo or susceptibility weighted imaging is sensitive in detecting cerebral
microbleeds, cavernomas, and cortical vein thrombosis. T2-based sequences
are helpful in detecting characteristic changes of small vessel disease, edema,
175
and abnormal flow-voids that can be suggestive of vascular malformation.
T1-based sequences with contrast are useful to detect underlying enhancing
lesions like tumors. Hemorrhage can be seen with primary central nervous
system (CNS) tumors or with metastatic lesions from breast, lung, melanoma,
renal cell carcinoma, choriocarcinoma, or papillary thyroid. Additionally, T1
and T2 sequences can be helpful to determine the age of blood products at
that time of MRI (Table 17.1).
SAH
Up to half of SAH are fatal, and those who survive often have neurologic im-
pairment. Up to 85% of nontraumatic SAH are caused by aneurysmal rupture.
In about 15% of nontraumatic SAH, there is no evidence of aneurysm on the
first angiogram, and half of those are ultimately attributed to nonaneurysmal
perimesencephalic hemorrhage.
Only 10% of people presenting with thunderclap headache are found to
have SAH, which makes proper workup essential. CTH without contrast has
a sensitivity over 95% for detecting SAH within 24 hours of hemorrhage, but
MRI brain may be more sensitive after several days. In SAH without hemor-
rhage on CTH, lumbar puncture may reveal SAH in an additional 3% to 5%.
CSF should be examined for the persistence of red blood cells in each tube
and for xanthochomia, which is a yellow color of the supernatant that can be
seen after ~12 hours due to hemoglobin breakdown. CTA is a noninvasive tool
that may be useful for detection of aneurysms and helpful to identify active ex-
travasation. DSA is the gold standard for detection of aneurysm as well as va-
sospasm and offers the additional benefit of less invasive aneurysm treatment
as described further in the following neurointerventional procedures sec-
tion. Transcranial Doppler (TCD) monitoring is recommended daily and can

be used to monitor intravascular cerebral blood flow velocities and calculate
Lindegaard ratio, both of which have been correlated to symptomatic and/or
angiographic vasospasm.
SDH
Hemorrhage into the subdural space is caused by damage to small bridging
veins. Most commonly this damage is caused by head trauma, but it can also
be seen in the elderly, infants, or neurodegenerative disorders where the sub-
dural space is wider and therefore bridging veins are more prone to damage
with changes in head velocity. CTH classically shows a crescent-shaped hemor-
rhage that crosses suture lines. While a hyperacute SDH may appear isodense
Section 2
with brain tissue, the typical acute SDH will appear hyperdense on CTH, and
chronic SDH are typically hypodense although anticoagulation may result in a
fluid-fluid level of varying densities.
EDH
The most common cause of EDH is head trauma resulting in damage to the
middle meningeal artery, which courses near the pterion, the skull point where
the frontal, temporal, parietal, and sphenoid bones merge, making it thin and
more susceptible to injury. CTH shows a convex or lentiform-shaped extra-
176
axial hemorrhage due to restriction between suture lines. Due to the high
velocity nature of arterial hemorrhage, EDH can cause increased mass effect,
increased intracranial pressure, and herniation. Interval CTH can be a useful
tool to monitor for midline shift and early signs of herniation if clinical exami-
nation cannot be reliably followed.
Ischemic Stroke
Acute ischemic stroke (AIS) accounts for 85% of strokes in adults. Workup
and treatment of AIS is largely dependent on etiology, which can be catego-
rized as follows:
• large vessel atherosclerotic disease resulting in hypoperfusion or
arteriogenic emboli
• penetrating artery disease due to lipohyalinosis
• cardiogenic embolism including atrial fibrillation, valvular disease, and ven-
tricular thrombus
• cryptogenous stroke
• other causes including dissection, drug abuse, prothombotic states,
vasculopathy, and vasospasm
Initial AIS workup starts with CTH, which may show frank hypodensity or
early signs of infarction including cortical effacement, blurring of the grey-
white junction, loss of the insular ribbon, or a hyperdense vessel sign. The
ASPECT score is a standardized scoring system for quantifying the size of
infarction and subtracts 1 point from 10 total for each of the following re-
gions of hypodensity scored only on the affected side: at the ganglionic level
caudate, lentiform nucleus, internal capsule, insular ribbon, anterior middle

cerebral artery (MCA) cortex (M1), MCA cortex lateral to the insular ribbon
(M2), posterior MCA cortex (M3), and then at the supraganglionic level an
additional three cortical areas (M4–6) immediately superior to M1–3. Large
strokes (APSECT 1–4) are associated with high risk of reperfusion bleeding
if revascularized.
DSA is the gold standard for extracranial and intracranial vascular im-
aging for AIS, but generally noninvasive studies are first obtained. CTA of
the head and neck can be used to identify and measure the degree of a
flow-limiting stenosis and is often helpful in determining the composition
of the stenosis (soft plaque, thrombus, calcified plaque). In the case of ar-
terial dissection, a true and false lumen can often be identified. Multiphase
CTA is a technique where multiple additional CT images are collected
following one bolus of intravenous (IV) dye and can be used to detect
Chapter 17
delayed vessel filling in a particular area. This technique can help discrim-
inate occluded vessels from pseudo-occlusions and help characterize col-
lateral supply in the setting of acute intracranial occlusions. Although CTA
can usually be obtained more quickly than MRI/A, it does require the use
of IV dye, which limits its application in patients with iodine-based contrast
allergy or renal impairment.
CT perfusion (CTP) is a powerful tool for AIS imaging that can also be
177
obtained quickly, though with additional IV dye and radiation exposure. CTP
raw data can be reformatted into cerebral blood volume (CBV), cerebral
blood flow (CBF), and mean transit time (MTT) or time to peak (TTP). The
ischemic core is the area with low CBV (and therefore low CBF and high MTT
and/or TTP). The ischemic penumbra, or area at risk that may be salvaged if
blood supply is restored quickly, is the area with preserved CBV but reduced
CBF with increased MTT and/or TTP.
MRI may also be used for AIS imaging, though it is not as widely available, is
more costly, and is not compatible in all patient populations (e.g., patients with
retained metal products). DWI is exquisitely sensitive for identifying the ischemic
core in AIS. The ischemic core is the area of DWI hyperintensity that correlates
to apparent diffusion coefficient hypointensity. Early AIS changes can be detected
as FLAIR hyperintensity after approximately six hours. GRE or susceptibility
weighted imaging (SWAN, SWI, T2*) is highly sensitive for detecting early blood
products. Magnetic resonance angiography (MRA) of the head time of flight
(TOF) are images collected without gadolinium and can help to identify intracra-
nial occlusions or flow-limiting stenosis. MRA neck TOF is often limited by motion
artifact and an inability to visualize the vertebral and/or internal carotid origins,
so MRA neck with gadolinium is preferred unless there is renal insufficiency. MR
perfusion can be interpreted similarly to CTP described previously.
CTA and MRA both tend to overestimate the degree of vessel stenosis
versus DSA but provide additional information concerning the blood vessel
wall and plaque composition not obtained by lumenogram (DSA). It is es-
timated that 1.9 million neurons die per minute, so there must be an ur-
gency to the rapid evaluation of AIS. As such, there are certain situations
where it may be warranted to skip directly to emergent DSA without first
obtaining noninvasive vessel imaging, perfusion imaging, or MRI. DSA is

useful for identifying the location of occlusion, measuring the degree of
stenosis, identifying collateral vascular supplies, and providing endovascular
revascularization via thrombectomy and/or stenting. Dual energy CTH can
be useful to separate hyperdensity from acute hemorrhage from that of
iodinated contrast media.
Carotid duplex ultrasound (CDUS) is a noninvasive technique useful for
identifying and tracking carotid stenosis. B-mode images can be obtained, which
help to visualize the vessel wall. Doppler measurements are also obtained, and
blood velocities can be correlated to degree of stenosis. While CDUS also
typically overestimates the degree of stenosis versus DSA, it is a safe, cost-
Section 2
effective way to manage carotid disease over time without exposure to excess
IV dye or radiation. Similarly, TCD can also be used to measure blood velocity,
which can be correlated to degree of stenosis for intracranial vessels. Notably
skull thickness may limit the operator’s ability to obtain good recording win-
dows. TCD can also be useful for prolonged monitoring to detect high intensity
transient signals (HITS) in patients with cardiac sources of embolism.
Postanoxic Injury
Anoxic brain injury occurs in the setting of widespread cardiopulmonary com-
178
promise including cardiac arrest, asphyxiation, near-drowning, drug overdose,

and carbon monoxide poisoning. Without oxygen, neurons begin to die after
about four minutes, but increased neuronal survival has been demonstrated
with use of aggressive resuscitation techniques and targeted temperature
management. Grey matter is first affected by hypoxia due to its high meta-
bolic rate, and damage is further exacerbated by excite-toxic glutamate re-
lease. Areas most commonly affected include basal ganglia, thalami, cerebral
cortices, cerebellum and hippocampi. CTH may initially show diffuse edema
with sulcal effacement, loss of grey-white differentiation, and basal ganglia
hypodensities. In more severe cases, a reversal of the normal grey/white
differentiation can be seen due to venous engorgement from venous outflow
obstruction due to increased ICP. Rarely a “white cerebellum sign” may be
seen where the brainstem and cerebellum appear relatively hyperdense due
to severe hypoattenuation of the cerebrum. A “pseudo-subarachnoid hem-
orrhage sign” may also be seen due to engorgement and dilatation of the
superficial venous structures appearing hyperdense relative to the ischemic
parenchyma. Cortical laminar necrosis may be seen as linear hyperdensities
outlining the cortex appearing several weeks after anoxic injury.
MRI can detect signs of cytotoxic edema within the first few hours after
postanoxic brain injury. DWI changes often pseudonormalize within the first
week. During the first 24 hours, T1-and T2-weighted images often appear
normal. T2 hyperintensities may be seen from 24 hours to two weeks. T1
hyperintensities corresponding to cortical laminar necrosis may be seen around
two weeks after injury. Functional MRI and positron emission tomography (PET)
may be helpful for prognostication, though larger studies are needed.
Infection

Prompt identification and treatment of CNS infection may significantly limit
morbidity and mortality. Neuroimaging can be helpful for identifying a specific
pathogen based on characteristic imaging findings, managing complications,
and prognostication of CNS infectious diseases. MRI with and without con-
trast is the test of choice for identifying CNS infections, though CTH can be
helpful in some situations.
Extra-axial Infections
Epidural and subdural collections suspicious for infection may be first iden-
tified on CTH, but additional imaging is often needed to distinguish epidural
phlegmon and subdural empyema from EDH and SDH. MRI typically shows
a fluid collection with irregular borders, interior restricted diffusion, and an
enhancing rim. An abscess is formed once the collection becomes loculated
with a fully enhancing rim.
Chapter 17
Meningitis
CTH has a low sensitivity for detecting signs of uncomplicated meningitis but
may be used to monitor for complications including edema and hydrocephalus.
Nonobstructive hydrocephalus is typically caused by inflammation of the arach-
noid villi and thus reduced CSF resorption. Obstructive hydrocephalus results
from extension of the primary infection into the ventricles with deposition of
179
purulent debris resulting in ventriculitis, which can cause focal obstruction of CSF
outflow. Enhancement of the meninges may be seen on CT or MRI with contrast.
Recently, contrast-enhanced T2 sequences were reported to have a higher sen-
sitivity for leptomeningeal enhancement than traditional contrast-enhanced T1
sequences. MRI FLAIR may also reveal hyperintense sulci resulting from a failure
of FLAIR suppression due to inflammatory proteinaceous material. Location of
meningeal involvement may also help to identify the pathogen; for instance, CNS
tuberculosis and neurosyphillis tend to affect the basal meninges.
Encephalitis
Vasogenic edema may be seen in early parenchymal infections, which appears
hypodense on CTH and hyperintense on T2-weighted MRI sequences. For
bacterial infections, as cerebritis progresses to abscess formation, a thickened
enhancing rim appears and the fluid-filled cavity fills with purulent material that
restricts diffusion and is therefore hyperintense on DWI. CNS toxoplasmosis
may appear similar to abscess on CTH, though it is usually restricted to the
grey-white junction. MRI DWI may be helpful to differentiate since restricted
diffusion is not typically seen with toxoplasmosis.
HSV encephalitis typically causes edema and sometimes hemorrhage in the
mesial temporal lobes that may be identified on CTH, though MRI FLAIR is more
sensitive early in the disease. Arboviruses commonly affect deep grey nuclei.
CTH with contrast may be helpful to identify enhancing parenchymal lesions
but may not be as useful in the postoperative period where evolving hema-
toma, seroma, and early infection may all have similar enhancing appearances
and MRI may be needed for further differentiation.
Encephalopathy
Some metabolic encephalopathies result in cytotoxic edema recognizable in

distinct anatomical patterns. For instance, Wernicke’s encephalopathy may be
associated with restricted diffusion on DWI but more commonly is associ-
ated with hypodensity on CTH that correlates to T2 hyperintensity on MRI.
Wernicke’s encephalopathy classically involves bilateral thalami, mammillary
bodies, tectum, and periaqueductal grey; however, changes have also been
reported in dentate nuclei, cerebellar vermis, inferior olivary complex, cau-
date, abducens and vestibular nuclei, red nuclei, and splenium. Metronidazole-
induced encephalopathy has been associated with bilateral T2 hyperintensity
in some similar locations including the dentate nuclei, abducens and vestibular
nuclei, red nuclei, and splenium.
Section 2
Hepatic encephalopathy is typically associated with T1 hyperintensity in the

globus pallidum but has also been reported in the substantia nigra and cere-
bellar dentate nuclei.
Neurointerventional Procedures
Indications
Catheter-based neuroangiography can be performed for diagnostic or inter-
180
ventional purposes. Catheter-based cerebral angiography can be used to identify,

follow, and/or treat vascular anomalies including cerebral aneurysms, arterio-
venous malformations (AVM), arteriovenous fistulas (AVF), and vasospasm.
Cerebral angiography is also essential in acute stroke treatment for thromb-
ectomy, angioplasty, and/or stenting. Other common uses for cerebral angi-
ography include tumor embolization, venous imaging, thrombectomy and/or
stenting, and WADA testing. Catheter-based cervical angiography is useful for
evaluating the degree of extracranial stenosis of the great vessels, vertebral ar-
teries, and carotid arteries, and has become the gold standard for the measure-
ment of extracranial carotid stenosis. Additionally, cervical angiography may
be useful for evaluation and treatment of cavernous-carotid fistulas, epistaxis,
tumor, and vascular abnormalities in the setting of neck trauma. Spinal angi-
ography can be useful for identifying the spinal arteries prior to spinal surgery
and spinal vascular malformations, including arteriovenous malformations and
fistulas, and for vessel identification and/or embolization for spinal tumors.
Treatments
In addition to high resolution vascular imaging, catheter-based neuroangiography
can be utilized for acute thrombectomy in ischemic stroke; venous thrombec-
tomy for refractory venous sinus thrombosis; embolization using coils, glue, or
flow-diverting stents (aneurysm, AVM/AVF, tumor, epistaxis); angioplasty (in-
tracranial and extracranial flow-limiting stenosis or vasospasm); application of
intra-arterial medications (calcium channel blockers for acute vasospasm, lytics
for central retinal artery occlusion); and for stent placement (most commonly
Table 17.2 Agents used in Endovascular Procedures
Medication Commercial Indication Dose Mechanism Half Life Side Effect Reversal
Name
Antithrombotic agents
Aspirin Long-term Load: 325 mg; 81– Inhibits Platelet Bleeding, gastritis Platelets
prevention of 325 mg daily Cyclooxygenase inhibition
postprocedure and decrease for 11 d
thromboembolism thromboxane A2
production
Clopidogrel Plavix Long-term Load: 600 mg; 75 Inhibits ADP- Platelet Bleeding Platelets
prevention of mg daily induced platelet- inhibition
postprocedure fibrinogen for 11 d
thromboembolism interaction
Prasugrel Effient Long-term Load: 60 mg; 10 mg antiplatelets: blocks 7 hr Bleeding Platelets
prevention of daily ADP receptors
postprocedure
thromboembolism
Cilostazol Pletal Long-term 50–100 mg BID Inhibits 11–13 hr Bleeding, Platelets
prevention of phosphodiesterase leukopenia
postprocedure III, increased cyclic
thromboembolism AMP
Eptifibatide Integrilin Short-term 180 µg/kg lV bolus Inhibits GP Iib/IIIa 90–120 Bleeding Platelets
prevention of followed by 2 µg/kg receptor min
periprocedure per minute infusion
thromboembolism for 20–24 hr
Heparin Short-term 70–100 U/kg Prevents factor X 90 min Bleeding, Protamine
prevention of IV bolus during activation thrombocytopenia sulfate, 1
periprocedure procedures mg for 100
thromboembolism (typically 5000 U), units (not to
postprocedure 18 exceed 50
U/kg/hr and titrated mg total)
(continued)
Table 17.2 Continued
Medication Commercial Indication Dose Mechanism Half Life Side Effect Reversal
Name
Bivalirudin Angiomax Short-term 0.75mg/kg bolus Direct thrombin 25 min Bleeding Recombinant
prevention of prior to procedure inhibitor activated
periprocedure followed by factor VII
thromboembolism 1.75 mg/kg/hr
for duration of
procedure and up to
4 hr post if needed
IA alteplase Activase Arterial occlusion 540 mg Converts <5 min Bleeding FFP
plasminogen to
plasmin
Sedation
Fentanyl Multiple Sedation 25–50 µg IV bolus μ-opioid receptor 3–4 hr Diarrhea, nausea, Naloxone
agonist constipation, (0.4–2 mg)
dry mouth,
somnolence, apnea
Midazolam Versed Sedation 1–2 mg IV bolus Benzodiazepene 3 hr Amensia, apnea Flumazenil
(200 μg
every 1–2
minutes; max
3 mg/hr)
Propofol Diprivan Sedation 0.5–1 mg/kg (5–50 Potentiation of 3–12 hr Hypotension,
µg/kg/min) GABA receptor apnea, dystonia,
activity, sodium infusion syndrome
channel blocker
Dexme Precedex Sedation Start at 0.6 mcg/kg/ Alpha2-adrenergic 6 min Bradycardia,
detomidine hr and titrated to 0.2 agonist hypotension
to 1 mcg/kg/hr
Other
Clevidipine Cleviprex Hypertension 1–2 mg/hr then Calcium channel 1–15 min hypotension
titrate blocker
Verapamil Calan Vasospasm 2–120 mg IA Calcium channel 3–7 hr Hypotension/
blocker bradycardia
Nimodipine Nimotop Vasospasm 30–60 mg po 0.8–3.2 Calcium channel 1–2 hr Hypotension
mg IA blocker
Atropine AtroPen Bradycardia 0.5 mg IV q 3–5 min Anticholinergic 2–3 hr Arrhythmia,
flusing, tachycardia
Amobarbital Amytal Wada test/amytal 25–100 mg IA over Barbiturate Several Bradycardia,
provocation 4–5 seconds minutes hypotension, CNS
depression
ADP = adenosine diphosphate; AMP = adenoside monophosphate; GP = glycoprotein; IV = intravenous; FFP = fresh frozen plasma; po = by mouth; CNS = central nervous system.
extracranial carotid, vertebral, subclavian, and in rare cases intracranial vessels).
A list of commonly used medications is presented in Table 17.2.

Preprocedural Preparation
Informed consent including specific risks should be obtained from the patient
or an appointed representative prior to the procedure unless medically neces-
sary in an emergent situation. Laboratory parameters including cell counts, cre-
atinine, and coagulation factors should be reviewed. When possible, patients
should not eat or drink for six hours prior to the procedure. Anticoagulants
should be held when possible. Patients with severe contrast allergies should
be pretreated with steroids. Sedation and analgesia for diagnostic procedures
are appropriate, and intravenous dexmedetomidine, propofol, midazolam and
Section 2
fentanyl are commonly used. General anesthesia may be needed for emboliza-
tion and is controversial for stroke thrombectomy.
Intraprocedural Considerations
Although most operators prefer accessing the right common femoral artery,
brachial artery access or direct carotid puncture may be required in special
circumstances.
Postprocedural Care
Manual compression or a closure device may be used for hemostasis. The patient
184
should be observed to ensure stable vital signs and the presence of peripheral
pulses, and the closure site should be monitored for hematoma development.
Complications
Diagnostic procedures have a low risk of complications. Nonneurologic
complications include renal failure, hematoma potentially requiring transfusion,
arterial dissection or occlusion potentially requiring surgical intervention, and
infection. Neurologic complications are extremely rare and include ischemic
stroke, symptomatic vasospasm, and intracranial hemorrhage. Interventional
procedures have higher complication rates that vary per procedure.
Further Reading
Osborn AG, Salzman KL, Barkovich AJ. Diagnostic imaging. In: Brain, 2nd ed. Salt Lake
City, UT: Amirsys; 2010.
Shetty VS, Reis MN, Aulino JM, et al. ACR Appropriateness Criteria head trauma. J Am
Coll Radiol. 2016 Jun;13(6):668–679.
DeLaPaz RL, Wippold FJ 2nd, Cornelius RS, et al. ACR Appropriateness Criteria® on
cerebrovascular disease. J Am Coll Radiol. 2011 Aug;8(8):532–538.
Chapter 18
Hypothermia and Fever

Control in Neurocritical Care
Kees H. Polderman
Physiology and Pathophysiology

Temperature is a key determinant, and a potential modifier, of acute brain
injury during and immediately following a period of ischemia, and especially
during reperfusion. Evidence suggests that the role of temperature control
in modifying ischemic injuries is equivalent to that of perfusion and oxygen-
ation, since every destructive cascade that can be triggered by a period of
ischemia is temperature dependent. These processes are stimulated by fever
185
and may be modulated by mild to moderate hypothermia. In addition, they
can continue for hours to several days after initial injury and can be retriggered
by new episodes of ischemia. Most of these processes are temperature de-
pendent in a more or less dose-dependent fashion; for example, in the tem-
perature range of 30o to 40oC, the metabolic rate, oxygen consumption, and
CO2 production all increase by 7% to 10% per degree Celsius. In addition,
neuroinflammation, neurotoxicity, and apoptosis are stimulated by fever and
inhibited by hypothermia. Temperature elevations also increase the perme-
ability of the blood-brain barrier, as well as intracellular acidosis and produc-
tion of toxic metabolites. Moreover, seizure tendency can be increased by
fever and mitigated by hypothermia.
Even under normal conditions the average temperature of the brain is
slightly higher than core temperature. When fever occurs, brain temperature
can rise well above the systemic temperature, especially in patients with acute
brain injuries. This is due to excess heat generated by some of the ongoing
destructive processes, including neuroinflammation, influx of excess calcium
into injured brain cells leading to hypermetabolism, free radical production,
and trapping of the heat in injured areas due to local edema formation and
vascular blockage. These processes lead to a general “overheating” of the
brain, with additional temperature elevations in injured areas. Numerous
studies have demonstrated that brain temperature exceeds core tempera-
ture by 1o to 2oC in patients with severe brain injury, with temperatures in
injured areas exceeding core temperatures by up to 4oC. The higher brain
temperature (especially in injured areas) can cause additional neurological
damage. Animal experiments have shown that induced brain injury increases
significantly when animals are externally warmed. This phenomenon is inde-

pendent of the initial severity of injury and is especially pronounced if hy-
perthermia coincides with a period of ischemia. Conversely, fever control
mitigates brain injury in animal models, while inducing hypothermia provides
additional neuroprotection.
Fever is a frequently occurring problem in patients with acute neurological
injury; in fact, the majority of these patients will develop fever in the course
of their intensive care unit (ICU) stay. A common cause is so-called central
fever, which is a direct consequence of the brain injury itself. However, brain-
injured patients are also at a high risk for infections; apart from the risk of
complications such as aspiration pneumonia due to decreased consciousness
Section 2
and diminished protective reflexes, brain injury can directly induce immune
dysfunction mediated through the vagal nerve, with efferent signals inhibiting
proinflammatory cytokine production, leading to an immunocompromised
state with increased susceptibility to infections. Therefore, patients with
acute brain injury may have central fever, infectious fever, or a combina-
tion of both, either simultaneously or sequentially. Whatever the cause,
the result is that the temperature setpoint is elevated, triggering the body’s
mechanisms to increase core temperature, and the patient develops fever.
As would be expected when studying the pathophysiological processes, hy-
186
perthermia is independently associated with increased risk for adverse out-

come in all types of acute neurologic injury. Clinical studies in ischemic stroke
(AIS), subarachnoid haemorrhage (SAH), intracranial haemorrhage (ICH),
traumatic brain injury (TBI), and cardiac arrest have demonstrated independ
ent correlations between fever and worse neurological outcome, higher
mortality, and increased length of stay. For example, patients with AIS who
develop fever have larger infarct volumes and greater neurological deficits,
increased risk of hemorrhagic transformation, and a 3.4-to 6-fold increase
in risk for adverse outcome. Fever in cardiac arrest patients increases the risk
of unfavorable outcome by a factor of 2.3 per degree Celsius temperature
increase above 37oC. In patients with ICH, the proportion of time spent at
temperature >37.5°C within the first 72 hours is independently associated
with poor outcome, and patients with fever have a significantly higher risk of
hematoma expansion. In patients with SAH and TBI, fever has been independ
ently linked to adverse outcomes.
This detrimental impact is partly explained by the pathophysiology outlined
previously, with brain temperature far exceeding core temperature and heat
trapping in injured areas. The generalized increase in metabolic rate with cor-
responding increases in minute ventilation and oxygen consumption could also
be detrimental, depending on the patient’s condition.
Not all of the processes triggered by an episode of brain ischemia are
purely harmful. For example, it has been shown that some degree of
neuroinflammation can contribute to cellular repair after an ischemic insult.
Further, in some situations the proinflammatory state that may be associated
with fever can help the body fight infections. Fever can inhibit the growth
Hypothermia and Fever Control

of certain species of bacteria while simultaneously stimulating immune cell
function and enhancing antibody and cytokine synthesis. Several studies sug-
gest that suppression of fever with antipyretics in patients with influenza can
adversely affect outcome.
However, when the processes are uncontrolled and overwhelming they
are harmful and can lead to cell destruction and death. A useful analogy
might be the process of infection and inflammation, where an immune re-
sponse is needed to combat the infection, but an extreme cascade of in-
flammation as may be seen in septic shock can overwhelm the organism
and lead to death. In this situation mitigating these processes can improve
outcome.
Thus the potential benefits of fever should be weighed against the risks.
Chapter 18
This balance may shift even within the same patient, with protective effects of
a febrile response outweighing harm in some phases of a disease, while harm
outweighs benefits in other phases. The majority of patients with acute brain
injury are likely to suffer harmful consequences of fever and will benefit from
strict fever control or even therapeutic hypothermia (TH). As the destructive
processes play out over periods of several days and can be restarted by new
episodes of ischemia, maintaining an “appropriate” core temperature should
be a key goal of care in critically ill patients, especially in those with acute neu-
187
rological injuries.
As with blood pressure and ventilation, there is an “optimal” temperature
that suits the patient’s need. Usually this is normothermia; in some situations
(such as following anoxic brain injury), a below-normal temperature may pro-
vide additional benefits, while in other situations (such as early stage influenza
and perhaps intracranial infections), a mild degree of hyperthermia could be
beneficial.
Clinical Evidence and Potential Indications

for Temperature Control
Though randomized controlled trials assessing benefits of fever control are
still lacking, based on the pathophysiological data discussed here and dozens
of observational studies showing a link between fever and adverse outcome,
maintaining normothermia is a widely accepted goal in patients with all types
of acute brain injury. Some observational and case-control studies in patients
with SAH, AIS, and TBI have reported that aggressive temperature control
using a combination of antipyretic drugs and mechanical cooling devices can
indeed improve outcomes. Whether hypothermia could further improve out-
come compared to strict fever control remains a matter of debate. Evidence
for use of fever control and/or TH for some specific types of injury is briefly
discussed next.
Neonatal Asphyxia
Several randomized controlled trials (RCTs) in infants with perinatal as-

phyxia have reported significant improvements in neurological outcome when
treated with TH (32o–34oC for periods of 48-72 hours). The benefits in
neurocognitive function persist on multiyear follow-up in middle childhood.
The estimated number needed to treat for the therapy is 6, i.e. six newborns
with post-asphyxia injury need to be treated with hypothermia to achieve one
additional case of good neurological outcome. Multiple nonrandomized trials
have reported similar benefits. Use of TH in neonatal asphyxia should be con-
sidered a standard of care.
Cardiac Arrest
Section 2
Two RCT cooling patients to 32o to 34oC and more than forty nonrandomized
before/after studies have reported significantly improved outcomes in patients
with witnessed cardiac arrest and an initial rhythm of ventricular fibrillation or
pulseless ventricular tachycardia. In contrast, a large RCT published in 2013
found no difference between strict temperature control at 36oC compared to
33.0oC. The conclusions of this study have been criticized for problems such
as delay (up to 4 hours) in randomizing patients, prolonged time (10 hours) to
target temperature, temperature fluctuations during the maintenance phase,
excessively rapid rewarming, potential selection bias limiting generalizability,
188
and other issues. Thus although there is general agreement on the need for
targeted temperature management after witnessed cardiac arrest, the precise
optimal target temperature is still being debated. Current guidelines from the
American Heart Association (AHA), the American Academy of Neurology
(AAN) and the Neurocritical Care Society (NCS) recommend using TH in
all patients who have return of spontaneous circulation following witnessed
out-of-hospital cardiac arrest with an initial rhythm of ventricular fibrillation
or pulseless ventricular tachycardia and considering TH for patients with
witnessed asystole or pulseless electrical activity arrest. The AHA guidelines
recommend strict temperature management with a relatively wide temper-
ature range of 32o to 36oC, followed by strict fever control. The AAN and
NCS guidelines recommend 32o to 34oC followed by strict fever control, with
36o as an alternative option in patients where hypothermia is deemed risky,
for example because they are actively bleeding (hypothermia can cause mild
coagulopathy).
As outlined previously, there is compelling evidence suggesting harmful effects
of fever in AIS. Data suggests that mild hypothermia applied in the hours fol-
lowing injury could significantly improve neurological outcome. Three feasi-
bility studies have used TH in combination with treatments aimed at achieving
reperfusion through administration of clot-dissolving drugs and/or mechanical
thrombectomy. Application of TH appeared feasible and safe, though one
study reported a high rate of pneumonia in the intervention group. Several
other trials with limited numbers of patients have studied TH as a treatment

for brain edema after large middle cerebral artery (MCA) strokes. Outcomes
were better compared to historical controls; however, use of TH for this in-
dication has largely been replaced by surgical decompression for malignant
MCA infarctions. The use of TH for a limited period of time in combination
with reperfusion treatment appears to be a promising approach, but no large
RCT addressing this issue has so far been performed. Strict fever management
after AIS should be a goal of care, with the use of TH limited to clinical trials.
Hypothermia may help in control of intracranial pressure and with efforts to
reduce brain edema in patients with TBI. Trials of hypothermia in this patient
population have many significant limitations, including heterogeneous patient
Chapter 18
populations, variable triggers to initiate therapy, and differences in the dura-
tion of treatment. Several studies using TH in early management after TBI
have reported improved outcomes, while one larger RCT found that there
was no difference in outcomes between strict fever control compared to TH.
A recent trial using TH to treat refractory ICP elevations in later stages of TBI
reported significantly worse outcome in the TH group. Based on these con-
flicting results, strict fever control should be the goal in patients with severe
TBI, while TH should be used only in the context of clinical trials.
189
Other Potential Indications
Under the right circumstances, hypothermia can improve myocardial contrac-
tility (see Table 18.1) and has been used to treat cardiogenic shock. Several
small case series have described the use of hypothermia to treat intracranial
hypertension and hepatic encephalopathy in patients with acute liver failure.
The appropriate target temperature and duration remain unclear, and, in most
studies, TH was used as bridge to liver transplant. Some case series and case
control studies have reported successful usage of hypothermia for adult respi-
ratory distress syndrome, generalized seizures, and spinal cord injury. A larger
study assessing TH for severe spinal cord injury is currently ongoing. The use
of TH in all of these indications is still experimental, though fever control
should be regarded as the standard of care.
Practical Aspects
Temperature can be increased by conserving heat (mainly through vaso-
constriction of arteries in the skin) and by generating heat (mainly through
shivering). Under normal circumstances, vasoconstriction begins at a core
temperature of around 36.5oC; the reduction in heat loss resulting from cuta-
neous vasoconstriction is ±25%. The effectiveness of heat conservation and
heat generation decreases with age; this is due to a less effective vascular
response, decreased ability to detect small temperature changes (leading to
Table 18.1 Effects of Mild Hypothermia (32o–34oC) on Key

Physiological Parameters
Cardiovascular • Decreases heart rate (unless there is a stress-mediated
and tachycardia related to shivering or discomfort). Hypothermia-
hemodynamic induced bradycardia does not require treatment in most
circumstances.
• Decreases cardiac output
• Increases central venous pressure
• Increases myocardial contractility unless the patient develops
tachycardia
• Increases membrane stability; care should be taken to keep core
temperature ≥30oC
Section 2
Electrolytes & • Electrolyte disorders can develop in the induction phase of

metabolism cooling and require frequent monitoring
• Decreases insulin sensitivity and reduces insulin secretion;
of note, insulin requirements are likely to decrease during
rewarming
• Increased synthesis of glycerol, free fatty acids, ketonic acids, and
lactate producing a mild metabolic acidosis
• Decreases metabolic rate by 7–10% per degree C
Ventilation • Decreases O2 consumption and CO2 production, thus requires
ventilator adjustments and frequent monitoring of blood gases
190
• Blood gas values are temperature dependent; lab analysis of

PaO2, PaCO2, and mixed venous or venous saturation should be
performed at the patients actual core temperature, or corrected
for temperature
• Temperature corrections:
• for PaO2, subtract 5 mmHg for every 1oC below 37oC
• for PaO2, subtract 2 mmHg for every 1oC below 37oC
• for pH, add 0.012 points for every 1oC below 37oC.
Infections • Impairs immune function and inhibits inflammatory responses
• Assessment for infection should include cooling system
workload (i.e., the energy expenditure of the cooling device to
maintain target temperature)
• May increase risk of wound infections, delayed healing, and skin
breakdown
Coagulation • Induces a mild bleeding diathesis due to effects on platelet
count, platelet function, the kinetics of clotting enzymes, and
plasminogen activator inhibitors
• Coagulation tests need to be performed at patients’ actual
core temperature
Drug clearance • Significantly increases the half-life of many drugs, particularly
those metabolized by the liver; this includes benzodiazepines,
propofol, fentanyl and morphine.
• Increases drug levels and/or enhances effect of medications
a slower counterregulatory response), and a lower basal metabolic rate. This

means that, in general, fever control and induction of hypothermia are easier
to achieve and maintain in older patients than in younger ones.
Heat generation through shivering is usually much more active, and
therefore more effective, at temperatures close to the normal range than
at temperatures that are several degrees below normal. In patients with a
normal hypothalamic setpoint, the shivering threshold is ±1oC below the
vasoconstriction threshold, so ±35.5oC. The shivering response peaks at
core temperatures around 35oC and decreases significantly at temperatures
below 33.5o to 34oC; in most patients shivering ceases completely at core
temperatures around 31oC, though there is a wide variability between patients
and even within the same patient based on the hypothalamic setpoint (see
later discussion).
Chapter 18
Shivering can cause multiple problems in patient management. Sustained
shivering can double the metabolic rate, thereby preventing effective temper-
ature management. In addition, it increases oxygen consumption (by 40%–
100%), the work of breathing, and heart rate; it induces a stress-like response
with tachycardia, hypertension, and elevated intracranial pressure; and it has
been linked to increased risk of morbid cardiac events and adverse outcome
in the perioperative setting. Therefore, shivering should be aggressively and
pre-emptively controlled, and shivering management should be an integral
191
part of the temperature management strategy. Some common antishivering
measures and drug regimens are listed in Table 18.2. As explained earlier, shiv-
ering will generally be most active at temperatures around 2oC below the hy-
pothalamic setpoint (1oC below the skin vasoconstriction threshold). Febrile
Table 18.2 Examples of Antishivering Measures and Drug

Regimens
Potential antishivering measures
Skin counterwarming is highly effective in most patients and can significantly reduce
the dose of antishivering drugs (including sedation) required to manage shivering
Commonly used antishivering drug regimens
Magnesium bolus 4–8 g over 10–30 minutes, magnesium drip 0.5–1 gram per hour;
target serum levels up to 1.2–2 mmol/l (3–5 mg/dl)
Buspirone 15–30 mg Q8 hr by mouth
Ondansetron 8 mg bolus (once)
Meperidine bolus 12.5–50 mg; some studies have used meperidine drips of 0.5–1.0 mg/kg
Fentanyl bolus 12.5–50 µg, drip 50–200 µg/hr
Dexmedetomidine 0.5–1 µg/kg loading dose, 0.2–1.0 µg/kg/hr maintenance dose
Addition of acetaminophen (paracetamol) and/or (in selected patients) NSAIDs can
help reduce temperature (usually by around 0.3o in central fever and 0.7oC in infectious fever)
NSAID = nonsteroidal anti-inflammatory drugs.
Table 18.3 Most Frequently Occurring Hypothermia-Induced

Changes in Laboratory Measurements
Increase in serum amylase levels
Mild-to-moderate thrombocytopenia (platelet count 30–100 × 1012)
Increase in serum lactate levels (2.5–5 mmol/l)
Hyperglycemia (risk of hypoglycemia during rewarming when insulin requirements
decrease)
Electrolyte disorders (low Mg, K, P, Ca)
Increase in liver enzymes (SGOT and SGPT)
Mild metabolic acidosis
Mild coagulopathy
Section 2
Changes in blood gasses
patients with acute brain injury are likely to have an elevated hypothalamic
setpoint, so shivering will occur at significantly higher temperatures.
Another important parameter affecting the ease and speed of cooling is
body mass; obese patients are more difficult to cool, especially with surface
cooling, due to insulating properties of adipose tissue and because of the
greater mass that needs to be cooled. Finally, an issue that often confounds
studies dealing with temperature management is that severe brain injury
192
can significantly diminish or even obviate the thermoregulatory response; it

is therefore much easier to cool patients with very severe brain injury (and
absent shivering response) than those with less severe injury. Thus “easy”
temperature control is, paradoxically, often a poor prognostic sign, whereas
increased workload of cooling devices predicts better neurologic outcome.
The most important physiological changes associated with induc-
tion of hypothermia and some management strategies are listed in Table
18.1. Associated changes seen in laboratory values are listed in Table 18.3.
Temperature elevations have the opposite effects.
Cooling Methods
Cooling technologies can be broadly divided into invasive (core cooling) and
noninvasive (surface cooling) methods.
The theoretical advantages of invasive cooling over surface cooling are
1. Some studies suggest greater speed of hypothermia/normothermia in-
duction when core cooling is used; however, it is unclear whether more
rapid induction improves outcome.
2. Possibly, invasive cooling has fewer and smaller temperature fluctuations in
the maintenance phase.
3. Some types of endovascular catheter allow continuous central (blood)
temperature measurement.
4. There is no risk of surface cooling-induced skin lesions.
5. The patient is easily accessible (i.e., no need to cover large areas of the

skin to achieve cooling).
6. Less medication may be needed to control shivering because there is more
effective shivering suppression with skin counterwarming (i.e., the entire
surface area can be warmed using warm air, leading to a significantly di-
minished shivering response). In a related issue, there may be better toler-
ance/less shivering with endovascular cooling when TH is used in awake,
nonintubated patients.
The theoretical advantages of surface cooling over invasive cooling are
1. It is easy to use and can be applied by nurses or nurse practitioners without
intervention by a physician.
2. No invasive procedure is required.
Chapter 18
3. There is no delay in the initiation of cooling.
4. Compared to endovascular cooling, there is no risk of catheter-induced
thrombus formation.
5. It can be easily applied outside the ICU setting.
The available data on safety and efficacy of different cooling technologies
is limited. Most published studies are small and evaluate only a single cooling
device or method; comparative studies have often been retrospective or
193
nonrandomized and/or have enrolled only small numbers of patients. Some
studies have suggested that invasive cooling may have more rapid and effective
temperature control and may help reduce nursing workload; in addition, these
studies all reported nonsignificant trends toward more favorable outcome
with more accurate temperature control.
Summary
In summary, temperature is a key physiological parameter in critically ill
patients. It should be regarded in the same way (and controlling it granted the
same importance) as blood pressure, heart rate, and ventilation parameters.
As with these other parameters, usually normal values are good, particularly
if the patient has acute brain injury, and as a general rule fever should be
avoided, which is easier said than done as the vast majority of these patients
will develop fever as a direct and indirect consequence of their neurologic
injury. The availability of effective mechanical cooling devices has allowed im-
proved and more accurate temperature control, and preliminary data suggests
that this can lead to further improvements in outcome.
Further Reading
Andrews PJ, Sinclair HL, Rodriguez A, Harris BA, Battison CG, Rhodes JK, Murray GD,
Eurotherm3235 Trial Collaborators. Hypothermia for intracranial hypertension after
traumatic brain injury. N Engl J Med. 2015;373:2403–2412.
Auer RN. Non-pharmacologic (physiologic) neuroprotection in the treatment of brain
ischemia. Ann N Y Acad Sci. 2001;939:271–282.

Azzopardi D, Strohm B, Marlow N, et al. TOBY Study Group. Effects of hypothermia
for perinatal asphyxia on childhood outcomes. N Engl J Med. 2014;371:140–149.
Badjatia N, Fernandez L, Schmidt JM, et al. Impact of induced normothermia on
outcome after subarachnoid hemorrhage: a case- control study. Neurosurgery.
2010;66:696–700
Bernard SA, Gray TW, Buist MD, et al.Treatment of comatose survivors of out-of-
hospital cardiac arrest with induced hypothermia. N Engl J Med. 2002;346:557–563.
Crossley S, Reid J, McLatchie R, et al. A systematic review of therapeutic hypothermia
for adult patients following traumatic brain injury. Crit Care. 2014;18:R75.
Deye N, Cariou A, Girardie P, et al. Endovascular versus external targeted tempera-
Section 2
ture management for out-of-hospital cardiac arrest patients: a randomized controlled

study. Circulation. 2015;132:182–193.
Hypothermia after Cardiac Arrest Study Group. Mild therapeutic hypothermia to im-
prove the neurologic outcome after cardiac arrest. N Engl J Med. 2002;346:549–556.
Karvellas CJ, Todd Stravitz R, Battenhouse H, Lee WM, Schilsky ML, US Acute Liver
Failure Study Group. Therapeutic hypothermia in acute liver failure: a multicenter
retrospective cohort analysis. Liver Transpl. 2015;21:4–12.
Nielsen N, Wetterslev J, Cronberg T, et al. Targeted temperature management at 33°C
versus 36°C after cardiac arrest. N Engl J Med. 2013;369:2197–2206.
194
Piironen K, Tiainen M, Mustanoja S, et al. Mild hypothermia after intravenous thrombolysis

in patients with acute stroke: a randomized controlled trial. Stroke. 2014;45:486–491.
Polderman KH. Induced hypothermia and fever control for prevention and treatment of
neurological injuries. Lancet. 2008;371:1955–1969.
Polderman KH. Mechanisms of action, physiological effects, and complications of hypo-
thermia. Crit Care Med. 2009;37:S186–202.
Polderman KH, Herold I. Therapeutic hypothermia and controlled normothermia in the
intensive care unit: practical considerations, side effects, and cooling methods. Crit
Care Med. 2009;37:1101–1120.
Chapter 19
Neuropharmacotherapy
Gretchen M. Brophy and Theresa Human
Neuropharmacotherapy Considerations
Medications used in the treatment of neurocritical care patients need to be
monitored closely and adjusted based on individual patient characteristics.
Pharmacokinetics (PK; the effects of the body on the drugs) and pharmaco-
dynamics (PD; the effects of drugs on the body) are not always predictable
due to changes in physiologic parameters in critically ill patients, age, and drug-
drug interactions. This chapter includes information on the PK/PD and drug
interactions of select medications. Physiologic changes that can impact medi-
cation PK in critically ill patients as well as the older adult patient are listed in
Table 19.1. These changes can alter medication concentrations, which will also
195
impact the medication’s duration of action, efficacy, and toxicity.
Hyperosmolar Pharmacotherapy
for Intracranial Pressure Management
Mannitol and hypertonic saline (HS) are commonly used in neurologically in-
jured patients in the acute setting to treat elevated intracranial pressure (ICP)
and cerebral edema. HS is also used in the treatment of hyponatremia. Both
agents theoretically work by producing osmotically driven fluid shifts and ap-
pear to be equally effective at equal osmolar doses.
Patient characteristics to consider when choosing agent of choice:
• Serum sodium concentrations
• Serum osmolality
• Osmolar gap
• Fluid status
• Renal function
Mannitol
Dose:
• 0.5–1 gm/kg over 5–15 minutes
• Can redose every 4–6 hours
• Duration of effect 90 minutes to 6 hours
Table 19.1 Potential Physiologic Changes that can Impact PK

Characteristics in Critically Ill and Older Adult Patients
PK Changes Changes Potential Medications
Parameter in Older in Critical Combined Impacteda
Adults Illness Effect in
Critically
Ill Older
Patients
Absorption ↓
Gastric pH ↓
Gastric pH ↑↓ Drug Itraconazole,
↑ Gastric ↑ Gastric absorption ketoconazole,
motility motility with GI sulfonamides,
changes (pKa dapsone,
↓ Absorptive ↓ Perfusion
of drug) acetaminophen,
Section 2
surface Enteral
↓ Drug and dabigatran
↓ Splanchnic nutrition
nutrient
blood flow Exocrine
absorption
pancreatic
with decreased
dysfunction
perfusion
↓ Bioavailability
of oral drugs
Distribution ↓ CO ↓CO Third spacing Aminoglycosides,
↓ Total ↑ Total body and Vd β-lactams,
body water weight and alterations daptomycin,
196
total body Significant phenytoin

↓ Lean
body mass water (fluid changes in
accumulation/ unbound
↓ Serum
volume fraction of
albumin
overload) drugs—
↑ Alpha-1 acid changes in
↓ Lean
glycoprotein effect and rate
body mass
↓Body fat, but of elimination
↓Serum
↑ % of albumin
total body ↑Alpha-1acid
weight glycoprotein
↓↑ Plasma pH
Metabolism ↓ Hepatic ↓ Enzyme Reduced Codeine,
mass activity(phase first-pass oxycodone,
↓ Enzyme I reactions are metabolism hydrocodone,
activity more profound Reduced tramadol,
(primarily than phase II effect of clopidogrel,
phase conjugative CYP2D6 dabigatran,
I [oxidative metabolism) prodrug tamoxifen,
pathways]) ↑↓ Hepatic substrates lidocaine,
blood flow verapamil,
↓ Hepatic Decreased
(early vs. propranolol,
blood flow clearance
late effects midazolam,
of high-
based on labetalol,
extraction
disease state, promethazine
drugs
iatrogenic
vaso
constriction)
PK Changes Changes Potential Medications
Parameter in Older in Critical Combined Impacteda
Adults Illness Effect in
Critically
Ill Older
Patients
Excretion ↓ Renal blood ↓ GFR ↓ Drug Isoflurane,
flow RRT elimination desflurane,
↓ GFR and increased sevoflurane,
↓ Ventilatory
risk of toxicity H2RAs,
Chapter 19
↓ Tubular capacity
↓ Elimination enoxaparin,
secretion
of inhaled antibiotics,
↓ Renal mass anticonvulsants,
anesthetic
↓ Ventilatory agents fluoroquinolones
capacity
PK = pharmacokinetics; GI = gastrointestinal; GFR = glomerular filtration rate; RRT = renal replacement
therapy.
a
Not all-inclusive.
Adapted from “Critical Illness and the Aging Population: Clinical Implications and Pharmacotherapy
Challenges.”
197
Use with caution in these situations:
• Renal failure (acute or chronic)—may cause drug accumulation
• Dehydration/hypovolemia—may worsen
• Mannitol causes wide fluid shifts by causing rapid fluid expansion followed
by diuresis. Use cautiously in volume-sensitive patients and possibly con-
sider a fluid bolus (500–1000 ml 0.9% NaCl) postdose
• Osmolar gaps >15–20 mOsm/kg or rapid increase or increasing trajectory
between calculations
Monitoring:
• Osmotic gap = measured osmolality—calculated osmolality
• Calculated osmolality: [(2 × Na) + (BUN/2.8) + (glucose/18)]
• Osmolar gap is the most accurate monitoring tool to detect presence of
unmeasured osmoles, such as mannitol
• It is useful in assessing the clearance of mannitol between doses
• Osmolar gap of greater than 15–20 mOsm/kg indicates incomplete
clearance between doses
• High osmolar gaps correlate with increased risk of reverse osmotic shift
and nephrotoxicity
• Osmolality >320 mOsm/kg is not a contraindication for continued
mannitol administration
• Laboratories
• Serum osmolality, BMP necessary to calculate osmolar gap (see pre-

vious calculation); drawn as a trough or prior to the mannitol dose
• Na, K, Mg, Phosphorus—monitored to prevent/treat imbalances due
to excessive diuresis
• Urine output should be evaluated to prevent hypotension and dehydra-
tion due to excessive diuresis
Adverse effects:
• Rebound ICP elevation (with high, repeated dosing)
• Acute kidney injury
• Dehydration
Section 2
• Hypotension
• Electrolyte imbalances
Administration pearls:
• Requires in-line filter (precipitates causing crystal formation); may re-
quire warming to dissolve crystals prior to administration
• May be given via peripheral intravenous access
Hypertonic Saline
198
Dose (concentrations listed are approximately equal osmolar to mannitol

1g/kg):
• 3% NaCl: 5 mL/kg over 5–20 minutes (range 2.5–5 mL/kg)
• 5% NaCl: 3 ml/kg over 5–20 minutes (range 2.5–5 mL/kg)
• 7.5% NaCl: 2 ml/kg over 5–20 minutes (range 1.5–2.5 ml/kg)
• 23.4% NaCl: 30 ml over 10–20 minutes (range 0.3–0.7 ml/kg)
• Duration of effect 90 minutes to 4 hours
Other dosing option:
• Continuous infusion titrated to a goal-sodium range
• Controversy exists regarding the benefit of a continuous infusion for ICP
control
Use with caution in these situations:
• Serum sodium >160 mEq/L or rapidly rising serum sodium
• Serum sodium <135 mEq/L due to potential increased risk of osmotic de-
myelination syndrome from rapidly changing sodium concentrations
• HS provides fluid expansion, so patients with decompensated heart failure
or pulmonary edema may be at increased risk of fluid overload
Monitoring:
• Laboratories
• Serum sodium every 4–6 hours; drawn as a trough prior to HS dose;
avoid prolonged hypernatremia >160 meq/L
• Cl-, CO2; HS can induce hyperchloremic acidosis when used repeatedly;
consider decreasing chloride content and replacing with sodium acetate
Adverse effects:
• Pulmonary edema
• Heart failure
• Acute kidney injury
• Coagulopathy
Chapter 19
• Hypernatremia
• Metabolic acidosis
• Thrombophlebitis
• Osmotic demyelination syndrome with rapid correction
Administration pearls:
• Central access required for 23.4% NaCl bolus and should be consid-
ered for bolus doses of >2% NaCl in nonemergent situations
• If a continuous infusion is given, central access required for >2% NaCl
199
Treatment Strategies for Status Epilepticus
and Refractory Status Epilepticus
Treatment of status epilepticus (SE) should be initiated and escalated rap-
idly until seizure activity is controlled both clinically and electrographically.
Anticonvulsants should be monitored closely for efficacy and toxicity, espe-
cially when administering first-generation anticonvulsants that have narrow
therapeutic concentration targets. Pharmacotherapy pearls for medications
used for emergent, urgent, and refractory SE treatment can be found in Tables
19.2, 19.3 and 19.4.
Table 19.2 Emergent Control Treatment Options

Drug Name Dose/Rate Adverse Effects Clinical Pearls
Diazepam 0.15 mg/kg IV (up Hypotension, Rapid redistribution
(Valium®) to 10 mg per dose); respiratory rate; can be given
may repeat in 5 min depression rectally; contains
Rate: 5 mg/min propylene glycol
Lorazepam 0.1 mg/kg IV (up May be longer-acting
(Ativan®) to 4 mg per dose); for seizure cessation
repeat in 5–10 min than diazepam;
Rate: 2 mg/min contains propylene
glycol
Midazolam 0.2 mg/kg IM up to Can also be given
(Versed®) 10 mg per dose buccally or intranasally
Table 19.3 Anticonvulsants for Urgent Control of SE

Phenytoin Load: 20 mg/kg Arrhythmias, Hypotension, esp in
(Dilantin®) Maintenance hypotension, older adults; STRONG
dose: 4–6 mg/kg/ bradycardia CYP inducer with
day divided in 2–3 many potential drug
doses interactions; total serum
level 10–20 mcg/ml;
Max rate: 50 mg/min
free level 1–2 mcg/ml;
levels may be attained 1
hr after infusion
Fosphenytoin Load: 20 mg PE/kg Paresthesias, Prodrug—converts
(Cerebryx®) Maintenance hypotension, to phenytoin 7-15
Section 2
dose: 4–6 mg/kg/ bradycardia min after infusion; less

day divided into thrombophlebitis than
2–3 doses phenytoin; same drug
interactions, monitoring,
Max rate: 150 mg
and drug level goals as
PE/min
phenytoin
Phenobarbital Load: 20 mg/kg Hypotension, Long-acting; contains
Maintenance sedation, propylene glycol;
dose: 1–3 mg/kg/ respiratory STRONG CYP
day divided into depression enzyme inducer with
many potential drug
200
1–3 doses
interactions
Rate: 50–100 mg/
min Serum level goal: 15–40
mcg/ml
Valproate sodium Load: 20–40 mg/ Hepatotoxicity, Fewer CV side effects
(Depacon®) kg IV pancreatitis, than phenytoin; CYP
Maintenance thrombocytopenia, enzyme inhibitor
dose: 10–15 mg/ hyperammonemic with many potential
kg/day divided into encephalopathy drug interactions;
2–4 doses meropenem will
significantly reduce
Rate: 3–6 mg/kg/
VPA levels and should
min
not be used with VPA;
caution use with pre-
existing pancreatitis,
mitochondrial, hepatic,
or bleeding disorders
Serum level goal: 50–
150 mcg/ml
Levetiracetam 1000–3000 mg/day Dizziness, behavior Reduce dose CrCl
(Keppra®) in 2 divided doses disturbances <30 ml/min; few drug
Rate: over 15 min interactions
Serum level goal: 12–46
mcg/ml (obtained rarely)
Lacosamide 200–400 mg IV PR prolongation, Monitor EKG in patients
(Vimpat®) every 12 hrs hypotension (rare) with underlying cardiac
Rate: over 15 min disease; reduce dose if
CrCl <30 ml; Few drug
interactions exist
Serum level goal: 2.8–18
mcg/ml (obtained rarely)
Topiramate 200–400 mg NG/ Metabolic acidosis No IV formulation
(Topamax®) PO every 6–12 hrs available
Propofol Bolus: 1–2 mg/kg IV Hypotension, Requires mechanical
(Diprovan®) RSE: 30–100 mcg/ respiratory intubation; often
kg/min depression, PRIS, vasopressors are
hyperlipidemia required, hemodynamic
monitoring, high lipid
load (increased calories;
Chapter 19
1.1 kcal/ml)
SE = status epilepticus; RSE = refractory status epilepticus; PRIS = propofol related infusion syndrome;
CV = cardiovascular; VPA = valproic acid; PR = PR interval on EKG; EKG = electrocardiogram;
IV = intravenous; NG/PO = nasogastric/oral.
Table 19.4 Refractory Status Epilepticus Treatment Options

Midazolam Bolus: 0.2mg/kg IV (to Sedation, respiratory Requires mechanical
(Versed®) max of 2 mg/kg) then depression intubation; often
Continuous vasopressors are
201
infusion: 0.05–2.9 mg/ required, rapid
kg/hr titration of dose
necessary to achieve
seizure suppression as
quickly as possible
Propofol Boluses: 1–2 mg/kg IV Hypotension, Requires mechanical
(Diprovan®) over 3–5 min then respiratory intubation; often
Continuous depression, PRIS vasopressors
infusion: 30–100 mcg/ are required,
kg/min hemodynamic
monitoring, high
lipid load (increased
calories; 1.1 kcal/ml)
Pentobarbital Bolus: 5–15 mg/kg Hypotension, Requires mechanical
(Nembutal®) IV (up to 50 mg/min) respiratory intubation,
then depression, cardiac vasopressors,
Continuous depression, infection, hemodynamic
infusion: 1–10mg/kg/ ileus monitoring; STRONG
hr (may require higher CYP enzyme inducer
doses to induce burst with many potential
suppression) drug interactions;
contains propylene
glycol
Ketamine Boluses: 1.5 mg/kg IV Hypertension, May be intubation and
(Ketalar®) (up to 4.5 mg/kg) then arrhythmias, vasopressor sparing
Continuous tachycardia,
infusion: 0.9–7.5 mg/ hypersalivation
kg/hr
IV = intravenous; PRIS = propofol related infusion syndrome.
Urgent Control Treatment Options
Patient characteristics to consider when choosing an agent of choice for

urgent control:
• Home anticonvulsant (if applicable)
• Liver/renal impairment
• Seizure type
• Drug-drug interactions
• Comorbid conditions
• Allergies/genetics
Hemostatic Agents for Anticoagulant Reversal

Section 2
Reversal of anticoagulants may be necessary in patients with life-threatening

bleeding or when an emergent surgical intervention is required. General man-
agement strategies to employ when treating a major hemorrhage include:
identify the cause and source of bleeding, maintain hemodynamic and respira-
tory stability, maintain normal body temperature, maintain blood pH and elec-
trolyte balance to facilitate coagulation, apply packing or dressing if applicable,
use local hemostatic measures or surgical intervention to control bleeding,
202
identify the anticoagulant and time from the last dose, and administer an ap-
propriate reversal agent. Reversal agents for warfarin and target selective oral
anticoagulants can be found in Tables 19.5 and 19.6.
Shiver Control during Therapeutic

Temperature Management
It is crucial to evaluate and treat shivering in patients who are being treated with
targeted temperature management because sustained shivering increases the
metabolic rate and counteracts cooling induction, consumes energy, contributes
to increased ICP, and increases brain oxygen consumption. Monitoring tools
can be used to measure the degree of shivering (Table 19.7) as well as help
guide antishivering therapy. Medications used to prevent shivering in patients
undergoing targeted temperature management can be found in Table 19.8.
Hemodynamic Agents
Neurocritical care patients may have a loss of autoregulation requiring
agents to prevent rapid or large fluctuations in blood pressure and perfu-
sion. In addition, hemorrhagic expansion can be worsened by elevated blood
pressure in the acute period. Vasoactive agents should be chosen based on
advantages and disadvantages for each patient when controlling blood pres-
sure. Characteristics of vasoactive agents that may help determine the most
appropriate drug for the patient can be found in Tables 19.9 and 19.10.
Table 19.5 Reversal of Warfarin
INR Clinical Setting Treatment Options
< 4.5 No bleeding Hold warfarin until INR in therapeutic range
Rapid reversal Hold warfarin
required (<24 Vitamin K 2.5mg PO
hrs)
If urgent reversal needed (≤12 hrs) for procedure
consider 4PCC 25 IU/kg IV or rFVIIa 1 mg IV
4.5-10 No bleeding Hold warfarin until INR in therapeutic range
Consider vitamin K 2.5 mg PO if risk factors for bleeding
Chapter 19
required (<24 Give vitamin K 5 mg PO
hrs)
If urgent reversal needed (≤ 12 hrs) for procedure
>10 No bleeding Hold warfarin until INR in therapeutic range
Give vitamin K 2.5-5 mg PO or 1–2 mg IV
Repeat every 24 hrs as necessary
required Give vitamin K 1–2 mg IV
(<24 hrs)
Repeat every 6–24 hrs as necessary
If urgent reversal needed (≤12 hrs) for procedure
203
Any INR Serious or life- 1. Hold warfarin
threatening 2. Give vitamin K 10 mg IV over 15–30 min.
bleeding
3. Option 1: If patient volume overloaded give PCC;
recheck INR 30 min after PCC administered
INR 4-factor PCC dose Max dose
2–3.9 25 units/kg 2500 units
4–6 35 units/kg 3500 units
>6 50 units/kg 5000 units

Option 2: If volume resuscitation needed give 15–
20 ml/kg FFP; recheck INR after FFP administered
INR = international normalized ratio; IV = intravenous; 4PCC = 4 factor prothrombin complex
concentrate; PO = by mouth; PCC = prothrombin complex concentrate; FFP = fresh frozen plasma.
Summary
Understanding the pharmacotherapy of neurocritical care patients can help
optimized medications management. This includes taking PK, PD, and age-
specific issues into consideration for each patient. Careful medication man-
agement may help improve efficacy and minimize toxicity, as well as decrease
drug-drug interactions in neurocritical care patients.
Table 19.6 Reversal of Target-Specific Oral Anticoagulants

Generic Half-Life Treatment Options
Factor Xa-Inhibitors
Apixaban 8–12 hrs; longer in • If ingested within 2 hrs, administer activated
(Eliquis®) renal impairment charcoal 50 g.
Rivaroxaban 5–9 hrs; • Consider time of last dose and half-life of
(Xarelto®) longer in elderly agent when deciding to reverse agent.
and renal • Recommend reversal if last dose given within
impairment 3–5 elimination half-lives of the drug to
Edoxaban 10–14 hrs; longer in ensure hemostasis.
(Savaysa®) renal impairment Option 1:
Section 2
• Administer PCC 25–50 units/kg over 10 min

• or if volume needed consider 15–20 ml/
kg FFP
Option 2:
• Andexanet alfa—recombinant, modified
factor Xa decoy that binds direct and indirect
Xa inhibitors
Direct Thrombin Inhibitor

204
Dabigatran 12-17 hrs; up to 34 • If ingested within 2 hrs, administer activated

(Pradaxa®) hrs in severe renal charcoal 50 g.
impairment • Consider time of last dose and half-life of
agent when deciding to reverse agent.
• Recommend reversal if last dose given within
3–5 elimination half-lives of the drug to
ensure hemostasis.
Option 1: Idarucizumab (Praxbind®) 5 gm IV
Option 2: Emergent Hemodialysis
PCC = prothrombin complex concentrate; FFP = fresh frozen plasma; FDA = Food and Drug
Administration; IV = intravenous.
Table 19.7 The Bedside Shivering Assessment Scale

Score Description
0 No shivering noted on palpation of the masseter, neck or chest wall and no
electrophysiological evidence of shivering (using EKG)
1 Electrophysiological evidence of shivering (using EKG), without clinical evidence
of shivering
2 Shivering localized to the neck and/or thorax only
3 Shivering involves gross movement of the upper extremities (in addition to
neck and thorax)
4 Shivering involves gross movements of the trunk, upper and lower extremities
Table 19.8 Antishivering Medications for Therapeutic
Temperature Management
Drug Dose Clinical Pearls
Acetaminophen 650–1000 mg q 4–6 hrs • No sedative effects
Route: IV, PO, PT • Caution with liver dysfunction
• Max dose 4000 mg/day
Ibuprofen 400–600 mg q 4–6 hrs • No sedative effects
Route: IV, PO, PT • Caution use with renal impairment or
recent GI bleed
Chapter 19
• Caution in high risk of bleeding due to
theoretical platelet dysfunction
Buspirone 20–30 mg q 8 hrs • May be synergistic with meperidine
Route: PO, PT • Does not decrease seizure threshold
Dexmedetomidine Bolus dose not • May be synergistic with meperidine
recommended • Short activity
Continuous • Dose limiting ADRs: Hypotension and
infusion: 0.2–2 mcg/ bradycardia
kg/hr
Magnesium Bolus: 4 gm IV q 4 hrs to • Serum Mg+ goal: 3–4 mg/dl
maintain goal serum level
or
205
Infusion: 0.5–1 mg/hr
Propofol 50–75 mcg/kg/min • Short activity
• Dose limiting ADRs: Hypotension and
bradycardia
• Must be intubated
Benzodiazepine Bolus: 2–5 mg IV PRN • Prolonged sedation with continuous
(midazolam, or infusion
lorazepam) • Lorazepam contains propylene glycol.
Infusion: 1–10 mg/hr
Caution when using high dose or
prolonged infusions
Fentanyl 25–150 mcg/hr • Short acting (t ½ = 3–4 hrs)
• Common ADR: Constipation
Remifentanyl 0.1–1 mcg/kg/min • Short acting (t ½ = 5–10 min)
• Common ADR: Constipation
Meperidine 25–100 mg IV q 4–6 hrs • Most effective antishiver drug
PRN • May have additive effects with
dexmedetomidine
• Accumulation occurs in renal
dysfunction
• Decreases seizure threshold; caution
with frequent dosing
Dantrolene 1–2.5 mg/kg IV q 6 hrs • Impacts degree of shiver (gain), not
(doses >100 mg q 6 shiver threshold
hrs are generally not • Good adjunctive therapy
recommended)
(continued)

Drug Dose Clinical Pearls
Neuromuscular Vecuronium: • Last-line therapy
blocker: Bolus: 0.05–0.1 mg/kg • Patient must be receive adequate
Vecuronium (duration 30–45 min sedation and analgesia prior to
Cisatracurium Infusion: 0.05–1.5 paralytic administration
mcg/kg/min titrate to • Must be intubated
TOF of 1–2 • EEG recommended during paralysis
Cisatracurium: • Vecuronium may have prolonged
Bolus: 0.1–0.2 mg/kg effect with renal and hepatic
(duration 45–60 min) dysfunction
Infusion: 2–10 mcg/
Section 2
kg/min
PO = by mouth; PT = per feeding tube; IV = intravenous; GI = gastrointestinal; ADR = adverse drug
reaction; PRN = as needed; TOF = train of four; EEG = electroencephalography.
Table 19.9 Antihypertensive Agents for Acute Blood Pressure

Management
Agent Dosing Clinical Pearls
206
Vasodilators
Nicardipine Initial dose: 2.5 mg/ • Onset: 5–15 min
hr and titrate up by • Duration of activity: 0.5–8 hrs
2.5mg/hr every 15
min until goal BP • Use caution with rapid titration as dose
achieved or max 15 stacking may occur and prolonged
mg/hr hypotension
• Adverse effects: Reflex tachycardia, nausea,
vomiting, headache, flushing
• Available in peripheral and central IV
concentrations
• Utilize the peripheral concentration (0.1
mg/ml) with caution in patients with
volume overload (i.e., pulmonary edema)
due to the high volume delivered
• Hepatic metabolism; Inhibits 2D6, 2C9,
2C19, 3A4; substrate of 3A4
Clevidipine 1–2 mg/hr initially, • Onset: 2 min
may increase dose • Duration of activity 5–15 min
every 90 sec to a
maximum of 32 • Preferred agent in patients with labile BP or
mg/hr need for rapid control of BP
• Contraindications: soy/egg product allergy
(formulated in a lipid compound), severe
aortic stenosis
• Caution: reflex tachycardia
• Metabolized by blood and tissue esterases
Sodium 0.3–0.5 mcg/kg/min • Onset: <2 min
nitroprusside initially, may increase • Duration of activity: 1–2 min
by 0.5 mcg/kg/min
every few minutes to • Caution in patients with coronary artery
achieve desired effect, disease due to coronary steal
maximum 3 mcg/ • Avoid in patients with elevated ICP
kg/min and AKI
• Sodium nitroprusside induced elevations
in ICP have never been observed in
Chapter 19
human subjects, although the potential
should be considered.
• Sodium nitroprusside should be avoided
in patients with AKI due to the risk of
thiocyanate toxicity
• If doses between 3–10 mcg/kg/min are
indicated, the patient should be monitored
for signs of cyanide toxicity (metabolic
acidosis, decreased oxygen saturation,
bradycardia, confusion and/or convulsions).
• Although not routinely administered,
sodium thiosulfate has been coadministered
207
with sodium nitroprusside using a 10:1
ratio of sodium thiosulfate to sodium
nitroprusside to prevent cyanide toxicity
• Adverse effects: Cyanide/
thiocyanate toxicity, nausea, vomiting,
methemoglobinemia
Hydralazine 10–20 mg every • Onset: 5–20 min
4–6 hrs • Duration of activity: 2–12 hrs (variable)
• Adverse effects: Reflex tachycardia,
headache, flushing
• Ensure adequate volume resuscitation to
avoid hypotension
Adrenergic Agents
Esmolol 250–400 mcg/kg/min • Onset: 1–2 min
• Duration of activity: 10–30 min
• Contraindicated in bradycardia, heart
block, cardiogenic shock, decompensated
heart failure
• Avoid loading doses
• Adverse effects: Reflex tachycardia,
headache, flushing
(continued)

Labetalol 20–80 mg every 10 • Onset: 2–5 min
min up to 300 mg • Duration of activity: 2–4 hrs
• Use caution with rapid IV titration as
dose stacking may occur and prolonged
hypotension
• Continuous infusion difficult to titrate due
to long duration of activity. Infusion should
be reserved for refractory hypertension
uncontrolled by bolus doses and/or other
medications. Titrate with caution to avoid
Section 2
prolonged hypotension.
• Adverse effects: Bronchospasm, HF
exacerbation, bradycardia/heart block
BP = blood pressure; IV = intravenous; ICP = intracranial pressure; AKI = acute kidney injury;
HF = heart failure.
Table 19.10 Vasopressor Agents for Hemodynamic Support

Medication Adrenergic Initial Dosing Clinical Pearls
208
Receptor
Activation
Vasopressors
Norepinephrine α, β1 2–5 mcg/min or • Great for increasing SVR and
0.05–1.5 mcg/ MAP while preserving CO
kg/min • First-line agent for
septic shock
• May increase oxygen
consumption
• Risk of dysrhythmias and
myocardial ischemia
• May decrease intestinal
perfusion and increase lactate
levels
Dopamine α, β1, Dopa: 1–3 mcg/ • Effective at multiple receptors
dopamine kg/min • Highest risk of dysrhythmias
β: 3–10 mcg/kg/ (esp. at higher doses)
min
α: 10–20 mcg/
kg/min
Epinephrine α, β1, β2 0.05–2 mcg/kg/ • Risk of dysrhythmias and
min myocardial ischemia
• Adverse effects:
Tachyarrhythmias,
hyperglycemia, lactic acidosis,
hypokalemia
Medication Adrenergic Initial Dosing Clinical Pearls
Receptor
Activation
Phenylephrine α 10–100 mcg/min • Least arrythmogenic
or • May decrease CO
0.1–1 mcg/kg/min • May cause reflex bradycardia
• Not indicated for septic
shock
Vasopressin 0 0.04 units/min • May reduce the dose of other
Chapter 19
vasopressors when added in
refractory hypotension
• Not indicated for shock
• May decrease splanchnic
perfusion and increase gut
ischemia
• Can be utilized for diabetes
insipidus
Inotropes
Dobutamine α, β1, β2 2.5–10 mcg/kg/ • Good to improve CO in
209
min decompensated heart failure
• May decrease SVR and
provoke hypotension
• Tolerance may occur with
prolonged administration
Milrinone 0 0.25–0.75 mcg/ • Useful if adrenergic receptors
kg/min are downregulated or
Reduce dose if desensitized in the setting of
CrCl <50 ml/min chronic HF
• Elimination reduced
in patients with renal
dysfunction
• Adverse effects: hypotension,
arrhythmias,
thrombocytopenia
SVR = systemic vascular resistance; MAP = mean arterial pressure; CO = cardiac output; HF = heart failure.
Further Reading
Brophy GM, Human T. Emergency Neurological Life Support: Pharmacotherapy.
Neurocrit Care. 2017;27(Suppl 1):51. https://doi.org/10.1007/s12028-017-0456-x.
Brophy GM, Bell R, Claassen J, et al. Guidelines for the evaluation and management of
status epilepticus. Neurocrit Care. 2012 Aug;17(1):3–23.
Brophy GM FC, Harpe SE, Patricia W. Critical illness and the aging population: clinical
implications and pharmacotherapy challenges. J Neurol Disord. 2015;3(1):1–8.
Section 3
General Management
and Special Populations
Chapter 20
General ICU Care of the

Neurological Patient
Samuel A. Tisherman and Sara Hefton
Introduction
Management of critically ill patients with acute neurologic conditions requires
a comprehensive approach to support extracerebral organs and promote neu-
rologic recovery. In general, the approach to management mirrors that in the
general intensive care unit (ICU) population. This chapter reviews general ICU
management topics important for patients with acute neurologic conditions.
213
Pulmonary
Airway management in critically ill patients with acute neurologic conditions
should be handled in much the same way as that for the general ICU patient.
The indication for intubation is often airway protection, though difficulty with
oxygenation and ventilation can also occur in these patients. The decision is
usually made by clinical assessment of the patient rather than blood gases. For
patients with neuromuscular diseases, however, serial measurements of nega-
tive inspiratory force or vital capacity can be useful.
Once a decision is made to intubate the patient, the next consideration is
whether or not the patient may have a difficult airway. Management of the
difficult airway is beyond the scope of this chapter, but the critical point is that
a patient with a known or suspected difficult airway should be treated differ-
ently from other patients. For a standard intubation, rapid sequence induction
(RSI), including administration of an anesthetic agent followed immediately by
a neuromuscular blocking agent, is usually appropriate. Preoxygenation is very
important to prevent hypoxemia during apnea.
Some neurointensivists use preinduction agents, such as lidocaine or fen-
tanyl, to help decrease the patient’s hemodynamic response to intubation.
The degree to which these agents help is unclear. There are several choices
of anesthetic agent for induction. Etomidate is commonly used as it has rapid
onset and does not directly affect hemodynamics. Etomidate, even a single
dose, does have an effect upon adrenal function, though the clinical impor-
tance of this finding remains unclear. In patients with seizure disorders, there
is a concern that etomidate may decrease the seizure threshold. Propofol and
General Managemen
thiopental can be useful as these agents can decrease intracranial pressure

(ICP), but they also can cause hemodynamic compromise. For patients who
are already hypotensive, ketamine can be a good choice. Previous concerns
about the potential increase in ICP caused by ketamine were unfounded.
Many intensivists practice a modified approach to RSI by performing bag-
valve-mask (BVM) ventilation after induction and before the endotracheal
tube is in place. Cricoid pressure (Sellick maneuver) has generally been aban-
doned as a technique for decreasing aspiration during intubation as its efficacy
Section 3
is questionable, and, if not done correctly, it can impair intubation. Elevating

the head of the bed or positioning the patient in reverse Trendelenburg po-
sition may help decrease the risk of aspiration. The sniffing position is also
helpful for improving visualization of the glottis but may not be possible in
patients with cervical spine injury requiring in-line stabilization.
The choice of neuromuscular blocker for RSI is either a depolarizing (suc-
cinylcholine) or nondepolarizing (e.g., rocuronium) agent. Contraindications
to the use of succinylcholine include a family or personal history of malig-
nant hyperthermia, hyperkalemia, or renal failure with potential hyperkalemia;
chronic neuromuscular diseases (e.g., stroke with paralysis, spinal cord injury,
myopathy, myasthenia gravis, or demyelinating disease, including Guillain-Barré
syndrome); or severe tissue trauma or burns. If any of these contraindications
214
exist, rocuronium is a good alternative because of its rapid onset and short
duration of action.
Direct laryngoscopy has long been the standard approach to visualizing the
airway. Videolaryngoscopy can improve first-pass success rates. If visualization
is poor, use of a Bougie can be extremely helpful. If intubation is unsuccessful,
adequate BVM ventilation can buy time for the induction agents to wear off
and to obtain additional equipment and expert assistance. If BVM is difficult,
use of a supraglottic airway, such as the laryngeal mask airway can provide
additional time. When other approaches fail, however, a surgical airway may
be needed.
Mechanical ventilation is used to improve oxygenation and provide ade-
quate ventilation. Noninvasive positive pressure ventilation, with alternating
high and low pressures, can be helpful for patients who have respiratory issues
that can be reversed in a relatively short period of time, but the patients must
be able to maintain airway patency. This approach may not be appropriate for
many patients with acute neurologic conditions.
As ventilators have advanced in recent years, there have been an increasing
number of ventilator modes. For the most part, there is no evidence that
one mode is better than another. Though there are hybrids, modes are either
volume or pressure controlled. In an assist-control mode, the patient receives
a controlled breath each time the ventilator is triggered above a minimum
rate. In synchronized intermittent mandatory ventilation mode, the patient
only receives a set number of breaths from the ventilator and breathes spon-
taneously in between. In pressure support mode, the patient must initiate a
breath to receive support. Pressure support is frequently used during venti-
General ICU Care

lator weaning.
For oxygenation, FiO2 can be increased as needed, though both neurologic
and pulmonary oxygen toxicity can be an issue above an FiO2 of 60%. Positive
end-expiratory pressure (PEEP) can be used to improve alveolar recruit-
ment, but high levels of PEEP can cause hemodynamic compromise by de-
creasing venous return and theoretically may increase ICP. The typical PaCO2
goal is 35 to 40 mmHg to prevent increased ICP from cerebral vasodilation.
Chapter 20
Hyperventilation with PaCO2 25–35 mm Hg to decrease elevated ICP should
only be used temporarily, as a bridge to a surgical solution.
A major challenge is that mechanical ventilation can cause ventilator-induced
lung injury, including barotrauma, volutrauma, and atelectrauma. A lung-
protective strategy includes using a tidal volume of 6–8 ml/kg, maintaining
plateau pressures <30 cm H2O, and using sufficient PEEP to recruit alveoli and
prevent atelectrauma. The “open lung” approach to ventilator management,
including PEEP and lung recruitment maneuvers, may optimize oxygenation
while protecting the lungs.
For patients with acute neurologic conditions who require intubation either
for acute respiratory failure or airway protection, an early tracheostomy may
be appropriate, though the optimal timing and impact on ventilator-associated
pneumonia, length of stay, and mortality remain unclear.
215
Cardiovascular
Hypotension, even if brief, can have a significant impact upon patients with
acute neurologic conditions. Shock can be classified as hypovolemic, cardio-
genic (e.g., chronic heart failure, valvular heart disease), obstructive (e.g., ten-
sion pneumothorax, cardiac tamponade, acute pulmonary embolism), and
distributive (e.g., sepsis, adrenal insufficiency, neurogenic, and medications).
While evaluating the patient, a bolus of fluid or starting a vasopressor may
be appropriate. Transthoracic echocardiography can be extremely helpful for
determining the cause of shock and titrating appropriate fluid resuscitation
and vasoactive agents (e.g., vasopressors [norepinephrine, phenylephrine] or
inotropes [dobutamine, milrinone]).
Control of hypertension may be critical for the management of some acute
neurologic conditions. The preferred agents are nicardipine and labetalol.
Because of its short half-life, clevidipine has some advantages over nicardipine,
but its use may be limited by cost considerations. Hydralazine and nitroprusside
are typically avoided because of undesirable effects on cerebral circulation.
Myocardial ischemia is a common issue in patients with acute neurologic
conditions because coronary artery disease shares many of the same risk
factors (e.g., hypertension, diabetes, and smoking). One of the first challenges
when dealing with suspected ischemia in neurological patients is the determi-
nation of which came first. Patients with primary cardiac events, particularly
myocardial infarction, can secondarily have neurologic events, such as a
General Managemen
stroke. The cardiac pathophysiology is likely acute coronary occlusion from

a ruptured plaque. In contrast, patients with severe neurologic insults, such as
subarachnoid hemorrhage or traumatic brain injury, can develop myocardial
ischemia. The pathophysiology is poorly understood, but it is likely driven by
catecholamines, with resulting demand ischemia. Because of patients’ neuro-
logic conditions, symptomatology and physical exam findings are often not
helpful. Management of these two different situations is significantly different.
In the former, reperfusion of the myocardium is a high priority, though some-
Section 3
times the decision-making is challenging because of the concomitant neuro-

logic status. In the latter situation, the main goal is decreasing demand, not
focusing on coronary vascular occlusion.
The first step in evaluating the potential for myocardial ischemia is to obtain
an electrocardiogram (ECG) to see if there is evidence of an acute coronary
syndrome. Serum troponin measurements are helpful. It is worth keeping in
mind that minor ECG changes and/or troponin elevations are likely related to
demand, not coronary occlusion.
If an acute coronary syndrome is suspected, early administration of
antiplatelet therapy (if safe), beta blockade, and a statin are beneficial.
Nitroglycerin or morphine can be used to treat ongoing chest pain. Early
cardiac catheterization and percutaneous coronary intervention should be
216
considered.
For cases of demand ischemia, the mainstay of therapy is beta blockade.
Decreasing blood pressure (if high) may be helpful, if this is safe from a neu-
rologic standpoint.
Stress-induced cardiomyopathy (Takotsubo) can occur in association with
the stress of neurologic critical illness. The presentation can mimic an acute
coronary syndrome, but the patient will have patent coronary vessels on cath-
eterization. The classic echocardiographic finding is transient ballooning of the
left ventricular apex. Management is supportive.
Congestive heart failure with pulmonary edema is a common concomitant
disease in older patients. The differential diagnosis includes neurogenic pul-
monary edema or acute respiratory distress syndrome. Echocardiography can
help, as can serial measurements of brain natriuretic peptide. Initial treatment
is diuresis (or ultrafiltration if the patient requires dialysis), though care should
be taken not to decrease cerebral perfusion with aggressive fluid removal.
Inotropic support may be necessary. Dobutamine works well but may cause
tachyarrhythmias. Milrinone is another good choice.
A very common arrhythmia in the ICU is atrial fibrillation. If the arrhythmia
is causing instability, cardioversion is indicated. If not, the mainstay of treat-
ment is rate control. The best drug for rate control is unclear. Beta blockers
and calcium channel blockers can be very effective but run the risk of causing
hypotension. Amiodarone can both slow the rate and promote conversion
but, even in the short run, may have significant toxicities. Digoxin can also be
effective, but obtaining an effect may take more time. For the patient with
chronic or recurrent atrial fibrillation, the next question is often whether to
General ICU Care

anticoagulate the patient. Weighing the risks and benefits of anticoagulation
may be complicated by the patient’s neurologic condition, including risk of
falls. The patient’s CHADS2 score, which can predict the risk of stroke, can
be helpful.
Sepsis
Chapter 20
All critically ill patients are at significant risk for developing infections and re-
sultant sepsis. The new onset of a fever (often defined as a temperature of
>38.3oC) in a patient in the ICU should prompt a thorough clinical evaluation
of the patient, seeking infectious and noninfectious causes. Though acute neu-
rologic conditions often cause fever, this should be considered a diagnosis of
exclusion after clinically ruling out infectious causes. Common causes of infec-
tion in the ICU include pneumonia (usually ventilator-associated), urinary tract
infection, surgical site infection, Clostridium difficile colitis, and central line as-
sociated blood stream infection. For patients with neurologic disease, partic-
ularly postoperative neurosurgical patients or any patient who has undergone
ICP monitoring, ventriculitis or meningitis need to be considered and sampling
of cerebral spinal fluid may be indicated. Noninfectious causes of fever include
217
drugs and, very rarely, venous thromboembolic disease (VTE). Workup of
fever should be tailored to the patient’s clinical assessment.
Sepsis represents a dysregulated host response to infection leading to life-
threatening organ dysfunction. The latest clinical scoring system for identi-
fying patients with sepsis is the quick Sequential (Sepsis-specific) Organ Failure
Assessment (qSOFA) score which can be applied to both ICU and non-ICU
patients. Patients who have two or more of the following criteria would meet
the definition of sepsis: respiratory rate of >22/min, altered mentation, or
systolic blood pressure of <100 mm Hg. Patients in septic shock require
vasopressors for maintaining an adequate blood pressure (mean arterial pres-
sure of 65 mm Hg) and have a lactate >2 mmol/L despite fluid resuscitation.
The Surviving Sepsis Campaign has recommended that septic patients re-
ceive specific interventions in a timely fashion in order to optimize outcomes.
These bundles of care have been listed in Box 20.1.
Hematologic
Appropriate goals for transfusions and correction of coagulopathy vary based
upon the patient’s primary disease process and comorbid conditions. One
needs to weigh the risks, benefits, and costs of blood component therapies.
For patients who are actively bleeding, packed red blood cells should be
transfused based upon hemodynamic parameters, not hemoglobin (Hb)
or hematocrit levels. Extrapolating from the findings in trauma patients,
General Managemen
Box 20.1 Surviving Sepsis Campaign Bundles of Care
To be completed within three hours of presentation

• Measure lactate level
• Obtain blood cultures prior to administering antibiotics
• Administer broad spectrum antibiotics
• Administer 30 ml/kg crystalloid for hypotension or lactate >4 mmol/L
To be completed within six hours of presentation

Section 3
• Apply vasopressors (for hypotension that does not respond to initial fluid resuscitation)
to maintain mean arterial pressure >65 mmHg
• In the event of persistent hypotension after initial fluid resuscitation, reassess volume
status and tissue perfusion
• Remeasure lactate level if initial lactate elevated
a “hemostatic resuscitation” approach in which fresh frozen plasma and

platelets are administered proactively to maintain adequate clotting ability
during the resuscitation is likely to be beneficial, though the exact target for
ratios of these blood products is unclear. Optimally, the blood component
therapy is guided by repeated clotting studies. Use of thromboelastography
218
or thromboelastometry can be extremely helpful.

For the stable patient who is anemic, the optimal transfusion trigger in the
general ICU population is a Hb of 7 g/dl. For patients with an acute coronary
syndrome, maintaining Hb above 8 g/dl may be appropriate. For patients with
acute neurologic conditions, there is a concern that, because of the brain’s high
level of metabolism and sensitivity to decreased oxygen delivery, a higher Hb
level is similarly appropriate. There is currently little data to substantiate this
assertion. So far, there has not been clinical evidence of improved outcomes
with a more liberal transfusion policy.
For patients who have intracranial bleeding or who require neurosurgical
intervention, anticoagulated patients should be treated with discontinuation
of the medication and reversal of the drug as soon as possible (Table 20.1).
Exceptions include intracerebral hemorrhages in the setting of cerebral ve-
nous sinus thrombosis or if the duration of time since the last drug admin-
istration was more than 3 to 5 terminal half-lives of the drug. Coagulopathy
from hepatic failure may be improved with recombinant factor VIIa; however,
correction must be weighed against the risk of thrombosis.
Renal
Acute kidney injury (AKI) is common in the ICU, often secondary to shock,
nephrotoxins (e.g., radiographic contrast agents, or other medications), or
sepsis. The workup typically begins with determining if the cause is prerenal
Table 20.1 Reversal of Anticoagulants
Mechanism Examples Reversal Not FDA Approved
Vitamin K antagonist Warfarin • Vitamin K 10mg IV
• 3- or 4-factor PCC
• If PCC not available, may use FFP
Direct Factor Xa Inhibitors Rivaroxaban, apixaban, edoxaban • Activated charcoal in intubated patient within 2 hours Aripazine (synthetic small
of ingestion molecule)
• Activated PCC or 4-factor PCC, 50 units/kg within
3–5 terminal half-lives of drug or in liver failure
• Andexanet alpha (recombinant modified factor Xa
protein)
Direct thrombin inhibitors Dabigatran, bivalirudin, desirudin, • Activated charcoal in intubated patient within 2 hours Aripazine (dabigatran)
argatroban, lepirudin of ingestion
• Idarucizumab 5 g IV (dabigatran only) if within 3–5
half-lives of drug
• Activated PCC or 4-factor PCC 50 units/kg if 3–5
half-lives or renal failure
• If idarucizumab not available in dabigatran use, may
use hemodialysis
Unfractionated Heparin • Reverse anticoagulation dose heparin with protamine Aripazine
sulfate 1 mg for every 100 units heparin in prior 2–3
hours, maximum dose 50 mg
LMWH Enoxaparin, dalteparin, • Protamine administration. For enoxaparin protamine Aripazine
Nadroparin, dose is based on time since last dose, 1 mg per
Tinzaparin. 1 mg if <8 hours prior, 0.5 mg protamine for 1 mg
enoxaparin if 8–12 hours prior. Other agents should
be reversed with 1 mg per 100 anti-Xa units of
LMWH in past 3–5 drug half-lives, up to 50 mg
• If protamine is contraindicated, rFVIIa 90 mcg/kg IV once
• Andexanet alpha
(continued )
Mechanism Examples Reversal Not FDA Approved
Heparinoid Danaparoid • rFVIIa 90 mcg/kg IV once
Pentasaccharides Fondaparinux, idraparinus, • aPCC (20 IU/kg)
idrabiotaparinux • If aPCC not available, rFVIIa (90 mcg/kg)
• Andexanet alpha
Thrombolytics Urokinase plasminogen activator, • Cryoprecipitate 10 units if dose within 24 hours
streptokinase, recombinant tissue • If cryoprecipitate not available, may consider
plasminogen activator, reteplase, antifibrinolytic agent (tranexamic acid 10–15 mg/kg
tenecteplase IV or ε-aminocaproic acid 4–5 g IV)
Anti-platelet agents: Aspirin, ibuprofen, naproxen, • DDAVP 0.4 mcg/kg IV
Cyclo-oxygenase inhibitors, dipyridamole, clopidogrel, prasugrel, • May give 1 unit platelets if neurosurgical intervention
adenosine diphosphate receptor ticagrelor, ticlopidine, cilostazol, planned
inhibitors, phoshodiesterase anagrelide, abciximab, eptifibatide,
inhibitors, glycoprotein IIB/ tirofiban, vorapaxar
IIIA antagonists, thromboxane
receptor antagonists, protease-
activated receptor-1 antagonists
IV = intravenous; PCC = prothrombin complex concentrate; FFP = fresh frozen plasma; LMWH = low molecular weight heparin; rFVIIa = recombinant factor VIIa; aPCC = activated
prothrombin complex concentrate; DDAVP = desmopressin acetate.
Reversal options are listed but note that the direct factor Xa inhibitors, direct thrombin inhibitors, and LMWH are not truly reversible with PCC.
or renal. Postrenal kidney injury can occur, but it usually resolves with place-
General ICU Care

ment of a Foley catheter. Clinical assessment of the patient’s volume status
is an important early step as hypovolemia is such a common cause of AKI. If
adequate hydration is assured, then urine and blood can be analyzed to cal-
culate the fractional excretion of sodium (FENa <1% suggests prerenal and
>2% suggests renal). At the same time, all potential nephrotoxic agents should
be stopped. Even with moderately severe AKI, patients may still respond to
high-dose diuretics, which may be worth trying when the patient seems fluid
Chapter 20
overloaded. Electrolytes (particularly potassium) and fluid status need to be
monitored closely.
Indications for urgent renal replacement therapy (RRT) include refrac-
tory metabolic acidosis, hyperkalemia (typically with ECG changes, rapid
rise despite medical therapy, or >6.5 mEq/L), fluid overload, and uremia as
manifested by pericarditis or unexplained mental status deterioration. The
choice between intermittent and continuous RRT can sometimes seem arbi-
trary. In general, continuous therapy is employed if the patient is hemodynam-
ically unstable (i.e., on vasopressors) or if there is neurological decline with
intermittent dialysis. For patients with severe AKI without any of the urgent
indications for dialysis, the optimal timing for initiating dialysis remains unclear.
As patients develop AKI, the dosing of medications that are cleared by the
kidneys needs to be adjusted. Following levels when possible can be invalu-
221
able. Once the patient starts on either intermittent or continuous RRT, dosing
will again need to be adjusted. Input from a critical care pharmacist can be
extremely helpful.
Electrolytes
Electrolyte abnormalities are common in the general ICU population. For the
most part, the management for the patient with neurological disease should
be the same as that for the general ICU population. The main exception to
this is sodium. Hyponatremia can be caused by the syndrome of inappropriate
antidiuretic hormone or cerebral salt wasting. The danger with hyponatremia
is increased cerebral edema. When mild, hyponatremia can be managed with
observation, free water restriction, or provision of enteral salt supplements.
For more severe cases, administration of hypertonic saline is indicated.
Hypernatremia can occur secondary to the patient’s condition (e.g.,
diabetes insipidus) or treatment with hypertonic saline, frequently for the
treatment of intracranial hypertension. Management of diabetes insipidus
depends upon the neurologic condition and the severity of hypernatremia.
When associated with nonlethal traumatic brain injury, it may be mild and
transient. Observation is appropriate. Administration of hypotonic solutions
runs the risk of increasing cerebral edema. When more severe, or associated
with the development of brain death, treatment may include administration
of desmopressin.
General Managemen
Endocrine
Glycemic control for critically ill patients has been a controversial topic in recent
years following findings that very tight glycemic control could improve patient
survival. Such an approach has been shown to increase the risk of hypogly-
cemia. Subsequent studies have not demonstrated benefit. Recommendations
now are typically to maintain blood sugar in the range of 70 to 180 mg/dl.
Adrenal insufficiency related to critical illness remains a controversial topic.
There has been evidence for alterations in metabolism of adrenal hormones in
Section 3
critically ill patients. The significance and implications for these changes are unclear.
For the most part, there is little evidence that testing for adrenal insufficiency and
empirically administering steroids are beneficial in the general ICU population.
One exception is for sepsis. The Surviving Sepsis Campaign recommends con-
sideration of “stress dose” steroids (200 mg hydrocortisone over 24 hours) for
patients in septic shock refractory to fluid resuscitation and vasopressors.
For neurologic diseases, steroid replacement may be indicated for patients
with direct injury to, or ischemia of, the pituitary gland. There may also be
disease-specific indications for steroids.
Venous Thromboembolism Prophylaxis

222
Patients in the ICU are at increased risk for VTE events because of immobility,
conditions that cause hypercoagulability, and vascular injury from the use of
central lines. Use of mechanical devices (intermittent pneumatic compression)
offers some protection with little risk and should therefore be applied unless
there is a contraindication. If safe, however, use of low molecular weight hep-
arin offers significantly greater protection from VTE, though the benefit of
appropriately dosed subcutaneous heparin may be similar. Pharmacological
prophylaxis and mobility activities should be started as soon possible based
on clinical safety considerations. Surveillance duplex studies to screen for DVT
are generally not recommended.
Nutrition
All critically ill patients are at high risk for nutritional deficiencies without ap-
propriate nutritional support. Patients with neurologic illness are no excep-
tion. Many are quite catabolic, with increased metabolic needs. If possible,
enteral nutrition should be initiated within 24 to 48 hours of ICU admission,
with a goal of reaching full nutritional support within 72 hours using high-
nitrogen formulas or adding protein supplements. There is rarely a need for
parenteral nutrition in this patient population. Though patients with neurologic
diseases may have difficulty swallowing, this is mainly an issue for oral feeding.
When fed with a feeding tube, standard aspiration prevention techniques
should be utilized, such as elevation of the head of the bed and use of con-
General ICU Care

tinuous feedings. Most patients do well with gastric feedings. Monitoring of
gastric residuals is not clearly beneficial. If they are monitored, tube feedings
should generally not be stopped unless the residual is greater than 500 ml
over six hours. If patients demonstrate intolerance to gastric feeding, either
by vomiting or gastric residuals >500 ml, use of a prokinetic agent, such as
metoclopramide or erythromycin, may help. If not, switching to postpyloric
feeding would be the next step. If the patient is not going to be able to tolerate
Chapter 20
sufficient oral feeding for several weeks, there may be a role for placement of
an endoscopic or operative feeding tube.
Stress Ulcer Prophylaxis

Patients who require mechanical ventilation for more than 48 hours or are
coagulopathic are at high risk for developing stress ulcerations that can lead
to clinically significant hemorrhage. Debate continues as to whether a his-
tamine2 (H2) blocker or a proton pump inhibitor (PPI) is the better option
for prophylaxis in terms of efficacy and/or cost. With increased gastric pH
by either agent, there is an increased risk for health care–associated pneu-
monia or development of Clostridium difficile colitis. Either agent is currently
223
appropriate. Also, gastric feeding alone may confer some level of benefit,
but it is not clear that this benefit is equivalent to that provided by either an
H2-blocker or PPI.
Further Reading
Dellinger RP, Levy MM, Rhodes A, et al. Surviving sepsis campaign: international
guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med.
2013;41580–637.
Holcomb JB, Tilley BC, Baraniuk S, et al. Transfusion of plasma, platelets, and red blood
cells in a 1:1:1 vs a 1:1:2 ratio and mortality in patients with severe trauma. The
PROPPR randomized clinical trial. JAMA. 2015;313(5):471–482.
Kacmarek RM, Villar J, Sulemanji D, et al. Open lung approach for the acute respi-
ratory distress syndrome: a pilot, randomized controlled trial. Crit Care Med.
2016;44:32–42.
Levitov A, Frankel HL, Blaivis M, et al. Guidelines for the appropriate use of bedside
general and cardiac ultrasonography in the evaluation of critically ill patients—part
II: cardiac ultrasonography. Crit Care Med. 2016;44:1206–1227.
Rhodes A, Evans LE, Alhazzani W, et al. Surviving Sepsis Campaign: international
guidelines for management of sepsis and septic shock: 2016. Crit Care Med.
2017;45:486–552.
Taylor BE, McClave SA, Martindale RG, et al: Guidelines for the provision and assess-
ment of nutrition support therapy in the adult critically ill patient: Society of Critical
Care Medicine (SCCM) and American Society for Parenteral and Enteral Nutrition
(ASPEN). Crit Care Med. 2016;44:390–438.
Chapter 21
Managing the Postoperative

Neurosurgical Patient
Daniel Ripepi and Colleen Moran
Introduction
Managing the postoperative neurosurgical patient involves timely recognition
and management of specific issues that arise in the immediate postoperative
period. The likelihood that a specific complication will arise for a given patient
is influenced by the nature of the procedure, the anesthetic techniques used,
and the patient’s preoperative comorbidities. The risk of some complications
can be reduced with appropriate preoperative assessment and medical opti-
224
mization. The management and treatment of postoperative complications is

equally important, and often management techniques used are unique among
neurosurgical patients. Management strategies along with rationale for these
postoperative concerns will all be discussed.
Postoperative Nausea and Vomiting

Postoperative nausea and vomiting (PONV) is one of the most common
complications of general anesthesia. Many studies have shown that there are
certain groups of patients whom are particularly at risk for PONV. Studies
comparing surgery types have shown patients undergoing thoracic, integu-
mentary, musculoskeletal, and superficial surgeries have significantly less re-
quirement for PONV treatment postoperatively when compared to patients
undergoing neurological, head or neck, and abdominal surgeries. Breast, axilla,
and endoscopic procedures had similar high rates of requiring PONV treat-
ment. Other studies have shown that a 10-year increase in age decreased the
likelihood of PONV by 13%, the risk for men was one-third that for women,
a 30-minute increase in the duration of anesthesia increased the likelihood of
PONV by 59%, general anesthesia increased the likelihood of PONV 11 times
compared with other types of anesthesia, and patients with plastic and ortho-
pedic shoulder surgery had a six-fold increase in the risk for PONV.
In summary, the factors that increase likelihood of PONV include female,
nonsmoker, history of PONV or motion sickness, anesthesia type, use of
nitrous oxide, anesthesia duration, intraoperative and postoperative opioid
Table 21.1 Medications for Treating Postoperative Nausea and Vomiting
Drug Class Common Example Mode of Action Preventive or Comments Contraindications
Rescue
Serotonin Receptor Ondansetron 5HT-3 Antagonism Both Nonsedating, may cause QT QT prolongation
Antagonist prolongation
Glucocorticoids Dexamethasone Unknown Preventive May affect wound healing
and glucose control
Anticholinergic Scopolamine Unknown Preventive Sedating Narrow angle glaucoma
Neurokinin Receptor Aprepitant Block neurokinin’s effect Preventive Expensive
Antagonist (proemetic agent) at
receptor site
Phenothiazine Prochlorperazine Dopamine receptor Rescue Side effects: acute Bone marrow
antagonist dystonic reactions, depression, blood
pseudoparkinsonism, dyscrasias, Parkinson
neuroleptic malignant disease, severe liver or
syndrome (at high doses) cardiovascular disease
Phenothiazine Promethazine Antihistamine Rescue Sedating
administration, and certain surgeries including neurological, head or neck, ab-
General Management
dominal, breast, axilla, and endoscopic procedures.

PONV can be treated prophylactically preoperatively and intraoperatively as
well as with rescue therapy postoperatively if the patient develops symptoms.
Prophylactic PONV treatments including ondansetron, dexamethasone, and
droperidol each reduce the risk of postoperative nausea and vomiting by
about 26%. Propofol and nitrogen (total intravenous [IV] anesthetic) were
similar to that observed with each of these antiemetics. All the interventions
act one another and the patient’s baseline risk. New medications for the pre-
Section 3
vention of PONV include neurokinin receptor antagonists such as aprepitant,

which is an oral medication taken preoperatively. See Table 21.1 for examples
of PONV treatments.
Airway Management in the Postoperative

Neurosurgical Patient
Airway and pulmonary management of the patient with neurological disease
contain an array of challenges, whose composition varies with the pathology at
hand. The managing physician should be familiar with the patient’s neurological
and cerebrovascular pathophysiology and the implications for management of
226
his or her airway and respiratory status. Extubation of the neurological pa-
tient also demands consideration of airway patency as well as respiratory me-
chanics. It is the goal of many neurosurgical procedures for the anesthesia team
to reverse the anesthetic agents and extubate the patient at the end of the pro-
cedure. This is because the diagnosis of complications relies on rapid neurolog-
ical examination after early awakening, and an awake patient is the easiest way
to obtain a neurologic exam. An early diagnosis of postoperative neurological
complications can limit potentially devastating consequences.
Many neurosurgical patients have ventilation impairment. Elevated PaCO2
or reduced PaO2 will increase cerebral blood flow or cerebral blood volume,
which can lead to a rise in intracranial pressure (ICP). ICP will also be a
function of cranial vault compliance and any intracranial mass effect secondary
to the neurologic disease.
Managing the airway postoperatively is of utmost importance in neu-
rosurgical patient. If the patient is not intubated, assessment of the airway
and review of previous intubation notes will be the starting point for airway
management. Particular attention should be paid to examination of the oro-
pharynx, teeth, mouth opening, and tongue, and it should be noted that the
airway exam postoperatively can be significantly different compared to the
preoperative assessment secondary to swelling of tongue, lips, and pharyngeal
soft tissue, as well as neuromuscular tone.
Care should be taken to assess the patient for respiratory distress or airway
obstruction in the postoperative period. All patients should have standard
pulse-oximeter, blood pressure, and electrocardiogram monitors. The care
Postoperative Management
team should assess for signs of increased respiratory effort such as nasal
flaring, intercostal retractions, and abdominal muscle use, as these can also be
signs of threatened airway obstruction. Initial options for patients with these
signs/symptoms include supplemental oxygen, simple jaw thrust, placement
of an oro-or nasopharyngeal or supraglottic airway if not contraindicated,
facemask ventilation, endotracheal intubation, cricothyrotomy, or tracheos-
tomy. Some of the indications for intubation for the neurological patient are
listed in Box 21.1, chief among which is coma.
It should be noted that certain neurosurgical patients can have pathology
which makes intubation more challenging, necessitating modification of normal
Chapter 21
techniques for intubation. Direct laryngoscopy has advantages of familiarity
and speed and also has several disadvantages, one being that it may require an
additional operator to provide in-line manual stabilization if a patient requires
cervical stabilization or has a cervical collar/ traction. Videolaryngoscopy
reduces but does not remove cervical spine movement, and although a good
view of the glottis can be rendered, the tracheal tube cannot always be guided
through the vocal cords. Fiber optic intubation is the most reliable method
of avoiding further distraction injury to the vocal cords but requires signifi-
cantly more investment in both training and equipment. Fiber optic intubation
can also be performed on an awake patient after topical anesthesia, and this
227
may be useful for patients in whom it is preferable to avoid induction with
standard anesthetics. Furthermore, the choice of medications for intubation
should be tailored to meet individual patient needs and optimize neurological
assessments.
Box 21.1 Indications for Intubation for the Neurological Patient
Immediate (life-threatening hypoxia likely)

Persistent apnea
Persistent airway obstruction despite airway insertion
Inability to bag/mask ventilate
Urgent
Glasgow Coma Scale <8
Protection of the lower respiratory tract from aspiration
Anticipated occlusion by hematoma or edema
Relative
Control of ICP by controlling PaCO2
Therapeutic ventilation for hypoxemia/hypercarbia in: pulmonary contusion/edema/infec-
tion, flail chest, cervical cord injury
Therapeutic and diagnostic procedures in combative or uncooperative patients
High metabolic demand from work of breathing
Developing specific extubation criteria for the neurological patient has
General Management
proved to be somewhat problematic. Successful extubation requires both res-

olution of underlying pulmonary processes and the ability to maintain airway
patency. Attempts to elucidate patients’ pulmonary function can be taken
from spontaneous breathing trials or trends in rapid shallow breathing index,
and by following institution-specific guidelines for weaning and discontinuation
of ventilatory support. The neurological patient is likely to have a higher inci-
dence of problems with airway patency.
Examples of specific problems in neurosurgical patients include facial and
Section 3
laryngeal edema from prolonged surgical cases in the prone position for spine
surgeries, posterior fossa surgery affecting protective airway reflexes, and in-
ability to use positive-pressure facemask ventilation in patients who have un-
dergone endonasal skull base surgeries. The use of corticosteroids to reduce
laryngeal edema has recently come into favor, with current data suggesting
that methylprednisolone (20–40 mg every four to six hours) should be con-
sidered 12 to 24 hours before a planned extubation in a patient at high risk
for postextubation stridor. Neurological and neurosurgical patients with
lesions involving the posterior fossa may have similar issues in maintaining
airway patency because of the impact on airway reflexes and musculature.
Another patient group with whom particular attention should be paid is those
who have had skull base surgery, especially via the endonasal approach. Careful
228
attention should be paid to emergence and extubation in these patients to pre-

vent excessive coughing and gagging, which may displace the reconstruction
materials and thus cause a CSF leak. It is also important to understand the con-
traindication for the passage of nasal tubes (e.g., nasogastric tube) and positive-
pressure mask ventilation. If positive-pressure ventilation is necessary, options
include a laryngeal mask airway or endotracheal tube, but not mask ventilation.
Postoperative Fever
Although the list of causes of postoperative fever is extensive, the initial focus
for most patients should be limited to the more common infectious and nonin-
fectious causes for that particular patient. Temperature elevations are frequent
in neurosurgical patients, especially in the postoperative period, and often cause
diagnostic confusion. Common causes include upper respiratory tract infection,
urinary tract infection, IV line sepsis, ventriculoperitoneal shunt infections, wound
infection, deep venous thrombosis, drug fever, myocardial infarction, and cen-
tral fever. The approach to a patient with temperature elevation usually starts
with lab tests and cultures of the potential infectious sources of the temperature,
while concurrently working through a differential of noninfectious causes.
Meningitis is a frequent and serious cause of fever after neurosurgery.
Classic symptoms and signs of meningeal inflammation, such as headache,
photophobia, and nuchal rigidity, are usually not helpful because they can
be caused either by infection or by hemolyzed blood from the surgery
irritating the meninges. Microscopic and analytical examination of the cere-
brospinal fluid (CSF) is indicated in patients with fever, because when com-
bined with specific clinical findings, characteristics of the CSF can help to
distinguish patients with infections from those with chemical meningitis. The
spinal fluid profiles in bacterial and chemical meningitis are similar. The fol-
lowing CSF profile is more indicative of chemical meningitis: fever less than
39.4ºC (102.9ºF), CSF WBC less than 7,500/µL, and CSF glucose above 10
mg/dL. It should be noted that chemical meningitis is infrequent after sur-
gery involving the spine and sinuses.
Neurosurgical procedures that impact the hypothalamus, stroke, and trau-
matic brain injury can lead to disorders of thermoregulation and cause fever.
Chapter 21
Postoperative hyperthermia may result following resection of tumors of the
hypothalamic floor. It should not be blindly attributed to hypothalamic dysfunc-
tion as surgical causes could also be implicated. Even though fever may cause
diagnostic confusion (central fever vs. infectious source), due to the potentially
devastating effects of pyrexia in patients with cerebral diseases, an attempt to
correct fever appears warranted in all patients with acute cerebral damage. This
is done in order to obtain a better functional recovery and to limit further insult
to the brain. In addition to antipyretics, some common and innovative methods
to control body temperature include intravascular cooling devices, surface
cooling devices, and cold IV fluid. Maintenance of normothermia is a desirable
229
therapeutic goal in managing patients with damaged or at-risk brain tissue.
Deep vein thrombosis occurs more frequently after neurosurgery than after
many other types of surgery. Not only is the patient likely to have limited
mobility before and after surgery, but prophylactic anticoagulation is often
delayed for neurosurgical patients because of concern for CNS hemorrhage.
Diagnosis is often based on a combination of a clinical exam as well as Doppler
examination of the deep veins. Treatment options are often debated, taking
into consideration the need to treat the deep vein thrombosis as well as mini-
mize the risk of hemorrhage associated with anticoagulation.
Postoperative Pain Control

Surgery and the associated tissue injury and inflammation are almost al-
ways associated with postoperative pain. Recent evidence indicates that
postcraniotomy pain is reported as moderate to severe in up to 80% of
patients and may persist for several days postoperatively. There is concern
that pain is often undertreated in neurosurgical patients as they may be unable
to communicate their pain level, and there is concern that the side effect of
pain medications will include oversedation and blunted response to hypox-
emia and hypercarbia.
The pain medication requirement will be variable; however, the extent and
type of the surgery will influence this requirement. Appropriate treatment of
pain postoperatively will include infiltration of a local anesthetic at the incisional
site at the end of surgery, nonsteroidal anti-inflammatory medications, ace
General Management
taminophen, opioids, and other nonnarcotic pain medications. Previously, co-

deine had been an analgesic of choice due to its favorable side effect profile.
However, recent studies have shown that if titrated properly opioids such as
fentanyl and morphine do not increase serious side effects compared to co-
deine, and they avoid the unpredictable effects of codeine that can occur due
to the need for metabolism of this prodrug.
Sodium Balance
Section 3
Sodium homeostasis is critical in neurosurgical patients. Disorders of sodium

metabolism including diabetes insipidus (DI), syndrome of inappropriate anti-
diuretic hormone secretion (SIADH), and cerebral salt wasting syndrome need
to be promptly diagnosed and treated. Changes in urine output and tonicity
should directly lead to appropriate diagnosis and intervention. A triphasic re-
sponse (polyuria–antidiuresis–polyuria) commonly occurs after neurosurgery.
Urine output, specific gravity, and/or osmolality should be monitored every
four to six hours in the early postoperative period of high risk patients to pre-
vent dysnatraemia.
In the absence of hyperglycemia or mannitol, polyuria (3 L/day or 40
230
mL/kg) with a low urine osmolality (<250 mOsm/kg) is usually due to cen-
tral DI but can also be nephrogenic DI (especially after subfrontal surgery).
Treatment for central DI may require (temporary) treatment with DDAVP,
while monitoring serum sodium and urine osmolality. A rise in urine osmolality
(100% if complete, 15%–50% if partial) after DDAVP [10 μg nasally or 4 μg
IV]) confirms central DI. Hyponatremia is usually caused by SIADH, but ce-
rebral salt wasting and secondary adrenal insufficiency should also be consid-
ered. The presence of a metabolic alkalosis and a high FEuric acid (usually >12%)
may help to differentiate SIADH from the other causes. Hyponatremia due to
SIADH may be treated with fluid restriction and hypertonic saline.
Blood Pressure Management and Fluid

Therapy
Cardiovascular disturbances are common in patients who undergo neurosur-
gery and can be secondary to a recent medical or surgical therapy, neurogenic,
or from a patient’s medical conditions that interact with his or her central
nervous system (CNS) pathology. Potential iatrogenic induced cardiovascular
problems include diuretic and steroid induced hypovolemia, bradycardia with
low cardiac output caused by surgical stimulation of the vagal nucleus in the
brainstem, and a Cushing’s reflex caused by increased ICP. Changes in blood
pressure are common postoperative complications. Typically, the surgeon
will have specific blood pressure parameters to decrease the risk of cerebral
edema and hematoma formation while preserving cerebral perfusion.
The differential diagnosis for hypotension is broad and includes neurologic
compromise, increased ICP, hypovolemia, heart failure, sepsis, drug effects,
anaphylaxis, and pulmonary embolus. The goal for most neurosurgical patients
is to preserve cerebral blood flow by augmenting blood pressure. Hypotension
or blood pressure augmentation is often addressed with use of either norepi-
nephrine or phenylephrine based on patient characteristics. Patients with low
myocardial reserve may require an inotrope, such as dopamine or epineph-
rine, and patients with sepsis may require norepinephrine and possibly vaso-
pressin. These medications are useful in treating the blood pressure while the
Chapter 21
etiology of the hypotension is diagnosed and/or treated.
Similarly, hypertension is often a common concern in the postoperative
neurosurgical intensive care unit. Common causes of hypertension in the
postoperative neurosurgical patient include pain, underlying hypertensive
disease, rebound from having home antihypertensives held during the hos-
pital admission, and reflexive response to neurologic injury. Calcium channel
blockers, beta blockers, hydralazine, diuretics, pain medication, and many
other medications can be useful in treating the blood pressure while the eti-
ology of the hypertension is diagnosed and treated. Nitroglycerin and sodium
nitroprusside are cerebral vasodilators and typically avoided as these agents
231
may increase cerebral blood volume.
Fluid therapy in the postoperative neurosurgical patient can be complicated
based on the goals for each individual patient. Patients with elevated ICP, for
example, may require an oncotic agent such as mannitol or hypertonic saline to
reduce their ICP. In general, the goal of fluid management is to maintain blood
pressure, intravascular volume, and cerebral perfusion pressure while min-
imizing cerebral edema. Appropriate fluid balance reduces the complications
which can result from inadequate or excessive fluid administration. The patient’s
fluid requirement is most often approximated by considering daily body fluid
requirements, fluid intake from oral and intravenous sources, and ongoing body
fluid losses. When initially evaluating the postoperative neurosurgical patient for
fluid status, much information can be gained from the report of intraoperative
blood loss, intake, and output. Furthermore, fluid status can be evaluated
by considering pulse, pulse pressure variation, weight change, thirst, mucous
membranes, intake and output, and urine output. If mannitol or diuretics are not
used, low urine output (0.5 mg/kg/hr) may indicate inadequate fluid replace-
ment. Newer tools can assess stroke volume variation to evaluate fluid status by
evaluating stoke volume variation (e.g., FloTrac-Vigileo, PiCCO, LiDCO).
Further Reading
Apfel CC, Korttila K, Abdalla M, et al. A factorial trial of six interventions for the preven-
tion of postoperative nausea and vomiting. N Engl J Med. 2004;350(24):2441–2451.
Chatzisotiriou AS, Selviaridis PK, Kontopoulos VA, Kontopoulos AV, Patsalas IA.
General Management
Delayed persistent hyperthermia after resection of a craniopharyngioma. Pediatr

Neurosurg. 2004;40(4):196–202.
Cormio M, Citerio G, Portella G, Patruno A, Pesenti A. Treatment of fever in neurosur-
gical patients. Minerva Anestesiol. 2003;69(4):214–222.
Cunha BA, Tu RP. Fever in the neurosurgical patient. Heart Lung. 1988;17(6 Pt.
1):608–611.
Forgacs P, Geyer CA, Freidberg SR. Characterization of chemical meningitis after neu
rological surgery. Clin Infect Dis. 2001;32(2):179–185.
Section 3
Hoorn EJ, Zietse R. Water balance disorders after neurosurgery: the triphasic response
revisited. NDT Plus. 2010;3(1):42–44.
Nemergut EC, Durieux ME, Missaghi NB, Himmelseher S. Pain management after crani-
otomy. Best Pract Res Clin Anaesthesiol. 2007;21(4):557–573.
Roberts RJ, Welch SM, Devlin JW. Corticosteroids for prevention of postextubation
laryngeal edema in adults. Ann Pharmacother. 2008;42(5):686–691.
Ruiz JR, Kee SS, Frenzel JC, et al. The effect of an anatomically classified proce-
dure on antiemetic administration in the postanesthesia care unit. Anesth Analg.
2010;110(2):403–409.
Sinclair DR, Chung F, Mezei G. Can postoperative nausea and vomiting be predicted?
Anesthesiology. 1999;91(1):109–118.
Souter MJ, Manno EM. Ventilatory management and extubation criteria of the neurolog-
232
ical/neurosurgical patient. Neurohospitalist. 2013;3(1):39–45.

Chapter 22
Pediatric Neurocritical Care

Dennis W. Simon and Ericka L. Fink
Epidemiology and Outcomes of Pediatric

Neurocritical Care Patients
Brain injury is the leading cause of death and disability in children. Overall, 11%
of children admitted to the hospital are given a neurologic diagnosis, giving a
national estimate of 273,900 admissions per year for the treatment of chil-
dren with neurologic complaint. Nearly one-third of these children will re-
quire intensive care unit (ICU)-level care (31% vs. 11x% without neurological
diagnosis) and, on average, will have longer hospital stay, higher costs, and in-
creased mortality (4.8% vs. 1.5% without neurologic diagnosis). Furthermore,
233
children surviving neurological conditions such as stroke, traumatic brain injury
(TBI), status epilepticus, cardiac arrest, and central nervous system (CNS) in-
flammation/infection have an elevated risk of long-term disability across mul-
tiple domains.
Despite the scope of the problem, effective, targeted therapies for children
largely remain elusive beyond supportive care. There are several challenges
to performing prospective trials in children, including difficulty in recruit-
ment, relative rarity of conditions being studied, and heterogeneity of the
study subject (i.e., age and development). However, simply translating adult
data to children undervalues important age and developmental diversities and
mechanistic and etiological differences inherent to the particular condition.
Despite these challenges, multicenter randomized, controlled trials of hypo-
thermia as a neuroprotective therapy for children with severe TBI and out-of-
hospital cardiac arrest have been successfully performed, though none were
able to demonstrate efficacy. Recently, a multicenter stroke trial evaluating
tissue plasminogen activator was stopped for futility in enrollment. Several
studies have enlightened us on optimizing supportive care targets to improve
outcome including invasively monitoring and targeting specific cerebral per-
fusion pressure levels in children with meningitis and encephalitis or severe
TBI. Clinical guidelines have been published for many pediatric neurocritical
care conditions, and though they often lack high-quality supportive data,
investigators have found that compliance results in better outcomes (Table
22.1). Pediatric neurocritical care service teams using various care models
Table 22.1 Recommendations for Management of Pediatric Neurocritical Care Conditions
Level of
Evidence
Post-cardiac 1. After OHCA, maintain five days of continuous normothermia (36°–37.5°C) or two days of initial continuous hypothermia (32°– Class IIa
arrest 34°C) followed by three days of continuous normothermia.
carepatients 2. Continuous measurement of temperature during this period. Class I
who remain
3. After IHCA, there is insufficient evidence to recommend cooling over normothermia.
comatose
4. Rewarm at 0.5°C per 2 hours or slower unless the patient requires rapid rewarming for clinical reasons.
5. Treat fever (≥38°C) aggressively after ROSC. Class I
6. Target a PaCO2 after ROSC that is appropriate to the specific patient condition; limit exposure to severe hypercapnia or Class IIb
hypocapnea.
7. Use lowest FiO2 to maintain arterial oxy-hemoglobin saturation ≥94% with goal of avoiding hyperoxia. Class IIb
8. Parenteral fluids and/or inotropes or vasoactive drugs should be used to maintain SBP >5th percentile for age. Class I
9. Reliability of any one variable for prognostication in children after cardiac arrest has not been established. Practitioners should Class I
consider multiple factors when predicting outcomes in infants and children who achieve ROSC.
Traumatic brain Consider the following:
injury 1. Use of ICP monitoring in infants and children with severe TBI (GCS ≤8). Level III
2. CSF drainage through an EVD in management of elevated ICP. Level III
3. Treatment of ICP at a threshold of 20 mm Hg. Level III
4. A CPP threshold of 40–50 mm Hg. There may be age-specific thresholds with infants at the lower end and adolescents at the Level III
upper end of this range.
5. If PbtO2 monitoring is performed, maintenance of PbtO2 ≥25 mm Hg. Level III
6. Hypertonic saline for severe TBI with elevated ICP. Effective bolus dose is 6.5–10 mL/kg. Used as an infusion, effective dose Level II (bolus)
range between 0.1 and 1 mL/kg/hr. Minimum dose to maintain ICP <20 mm Hg should be used. Serum osmolarity should be Level III (infusion)
maintained <360 mOsm/kg.
7. High-dose barbiturate therapy in hemodynamically stable patients with refractory intracranial hypertension despite maximal Level III
medical and surgical management.
8. Decompressive craniectomy with duraplasty in patients showing early signs of deterioration or herniation or are developing
elevated ICP refractory to medical management. Level III
9. Avoidance of prophylactic severe hyperventilation to a PaCO2 < 30 mmHg in the initial 48 hrs after injury.
10. Use of prophylactic phenytoin to reduce incidence of early posttraumatic seizures. Level III
11. Use of corticosteroids is not recommended to improve outcome or reduce ICP. Level III
Level II
Stroke Supportive Care
1. Ensure optimal hydration, correct hypoxemia, correct systemic hypotension, control of fever, and normalization of serum glucose Class I
levels.
2. In the absence of additional data, hypothermia should not be used in children with stroke. Class III
Perinatal Stroke
1. Correct coagulopathy or thrombocytopenia. Class I
2. If hydrocephalus develops, patients should undergo ventricular drainage and later shunting if significant hydrocephalus persists. Class I
3. Use rehabilitation and physical therapy to reduce neurologic dysfunction. Class II
4. Consider anticoagulation with LMWH or UFH in select neonates with severe thrombophilic disorder, multiple cerebral or systemic Class II
emboli, or clinical or radiological evidence of propagating CVST despite supportive therapy.
5. Thrombolytic agents are not recommended in neonates. Class III
Acute Ischemic Stroke in Infants and Older Children
1. In children with SCD and acute infarction, exchange transfusion to reduce sickle Hb to <30% total hemoglobin is reasonable. Class II
2. Age-appropriate rehabilitation and therapy programs are indicated for children. Class I
3. Consider administration of UFH or LMWH in children for up to 1 week after an ischemic stroke. Class IIb
4. Until there are additional published safety and efficacy data, tPA is generally not recommended for children with AIS outside a Class III
clinical trial. However, there was no consensus about the use of tPA in older adolescents who otherwise meet standard adult tPA
eligibility criteria.
5. Note: Use of endovascular thrombolysis or thrombectomy not reviewed in current pediatric guidelines.
(continued)
Level of
Evidence
Nontraumatic Hemorrhagic Stroke in Infants and Older Children
1.Stabilizing measures include optimizing respiratory effort, controlling systemic hypertension, controlling epileptic seizures, and Class I
managing increased ICP.
2. Individuals with SAH may benefit from measures to control cerebral vasospasm. Class IIb
Cerebral Venous Sinus Thrombosis in Infants and Older Children
1. Supportive measures for children with CVST include appropriate hydration, control of epileptic seizures, and treatment of Class I
elevated ICP.
2. Children with CVST and suspected bacterial infection should receive appropriate antibiotics. Class I
3. Consider monitoring intracranial pressure during the acute phase of CVST. Class IIb
4. Consider continuous EEG monitoring for individuals who are unconscious and/or mechanically ventilated. Class IIb
5. It is reasonable to institute either UFH or LMWH, whether or not there is secondary hemorrhage, followed by warfarin therapy Class IIa
for 3 to 6 months.
6. Consider administration of a thrombolytic agent. Class IIb
OHCA = out-of-hospital cardiac arrest; ICHA = in-hospital cardiac arrest; ROSC = return of spontaneous circulation; SBP = systolic blood pressure; ICP = intracranial pressure; TBI = traumatic
brain injury; GCS = Glasgow Coma Scale; CSF = cerebrospinal fluid; EVD = externalized ventricular drain; CPP = cerebral perfusion pressure; PbtO2 = brain tissue oxygen tension; LMWH = low
molecular weight heparin; AIS = arterial ischemic stroke; tPA = tissue plasminogen activator; UFH = unfractionated heparin; SCD = sickle cell disease; CVST = cerebral venous sinus thrombosis;
SAH = subarachnoid hemorrhage; EEG = electroencephalogram.
Class I: Evidence or general agreement that a procedure or treatment is useful and effective.
Class II: Conflicting evidence or divergence of opinion exists.
Class IIa: Weight of evidence or opinion favors utility or efficacy.
Class IIb: Weight of evidence or opinion is less well established.
Class III: Evidence or general agreement that the procedure or treatment is either not useful or effective or in some cases may be harmful.
have been implemented at some hospitals with the objective to improve clin-

ical care and outcomes by developing best practice.
General Approach to the Pediatric

Neurocritical Care Patient
Children with brain injury require a multimodal assessment by the intensivist
incorporating history, targeted physical exam, laboratory, neuroimaging, and
neurophysiologic monitoring (Figure 22.1). A relatively low threshold should
be applied toward ICU admission for observation and serial assessments to
Chapter 22
monitor disease progression and response to therapy. Care plans should be
anticipatory to prevent secondary organ injury. Appropriate consultation with
neurology, neurological surgery, neuroradiology, and rehabilitation providers
should be timely, and optimal care is team-oriented.
Several aspects of monitoring and treatment are shared across neurocritical
care conditions. For example, core body temperature should always be mon-
itored closely and fever (>38°C) avoided in patients immediately following
cardiac arrest, TBI, and stroke. On the other hand, indications for intracranial
pressure (ICP) monitoring are typically more patient and disease specific.
Although similarities between adult and pediatric neurocritical care exist,
237
the pediatrician’s oft-repeated adage “children are not simply little adults” is
perhaps nowhere more true than in the management of children with neuro-
logic injury. Clinicians managing a child with neurologic conditions must con-
sider age-appropriate epidemiology, history, physical exam, laboratory and
imaging findings, and available pediatric data when developing their assess-
ment and plan. Here we touch on a few of the expected differences for the
child requiring neurointensive care with disease-specific differences covered in
more detail in subsequent sections.
Physical Exam
Despite validation as a presenting severity of illness score for trauma patients,
the most commonly used test to describe and track neurologic status is the
Glasgow Coma Scale (GCS). The GCS score, however, can be under-or
overestimated by neurologically active medications, young age/ develop-
mental status (i.e., inability to follow commands), and endotracheal intubation
tube (i.e., loss of verbal GCS component). A modified pediatric GCS has been
developed for use in younger children that is particularly useful for children
<2 years old.
Prior to ossification of the fontanelles (anterior 9–18 months, posterior
2–3 months), the finding of a bulging fontanelle while the child’s head is el-
evated to 45° may indicate elevated ICP, an enlarged fontanelle may suggest
hydrocephalus or genetic disorder, and a sunken fontanelle may be due to
dehydration or malnutrition. An open fontanelle also allows for ultrasound to
General Management
238Section 3
Figure 22.1 General approach to the pediatric neurocritical care patient.
be used for imaging of the brain and is frequently used to monitor germinal
matrix hemorrhage or ventricular size in patients with hydrocephalus. Again,
this demonstrates how patient age must be factored into determining normal
versus abnormal exam findings.
Imaging
As children develop, the brain and skull grow, myelination occurs, and sutures
fuse. As a result, the “normal” appearance of the brain on imaging will change
over time. The availability of pediatric neuroradiology expertise can be critical

in interpreting neuroimaging. Brain computed tomography (CT) may be the
preferred imaging modality to acutely diagnose trauma and other lesions, but
due to concerns regarding the effects of ionizing radiation on brain develop-
ment and long-term cancer risk, other modalities such as magnetic resonance
imaging (MRI) or ultrasound, if appropriate to the clinical context, should be
considered.
Therapeutic Interventions
Several therapeutic agents used commonly in adult neurocritical care have
limited or no data in pediatric patients (e.g., thrombolytic therapy for stroke).
Chapter 22
In patients requiring long-term sedation, particularly children <3 years old,
propofol is not recommended due to the increased risk of propofol infusion
syndrome (bradycardia, metabolic acidosis, rhabdomyolysis, hyperlipide
mia). In addition, in our youngest patients (<2 weeks old), we typically avoid
benzodiazepines due to possible stimulatory effects of gamma-aminobutyric
acid (GABA) receptor in this population.
Monitoring the Pediatric Neurocritical

Care Patient
239
Monitoring children with neurological conditions provide key tools for early
detection and treatment of physiological derangements, assessment of
therapy efficacy, and optimization of outcome. Choice of monitoring should
be tailored to the condition and individual patient (Figure 22.1).
First, standards for monitoring of blood pressure, oxygenation and venti-
lation status, and temperature should be established. For example, children
requiring mechanical ventilation should be candidates for invasive arterial cath-
eter placement, regular blood gas assessment, and continuous central tem-
perature monitoring in order to meet standard supportive targets to prevent
secondary neurologic insult.
Electroencephalography (EEG) is an important tool that requires trained
personnel to apply and interpret. Brief (30 minutes) EEG should be consid-
ered for children with neurologic conditions and focal or diffuse neurological
examination findings, but if seizures are suspected the clinician should priori-
tize empiric treatment over delaying treatment until EEG results are available.
Continuous EEG is frequently used in children with alterations of conscious-
ness at risk of nonconvulsive status epilepticus and for those requiring neu-
romuscular blockade as the physical examination is severely limited. Of note,
normal EEG patterns will change significantly as children age and should be
interpreted by clinicians with expertise in pediatric neurophysiology.
Brain CT takes only minutes to complete and is critical in diagnosing mass
lesions, hemorrhage, and bone fractures but requires ionizing radiation,
limiting its use for repetitive evaluation. MRI is superior in early diagnosis of
General Management
ischemic stroke and can be helpful with long term prognostication but typical
evaluations last an hour or longer (excluding rapid stroke protocols) and can
be located far outside the ICU environment.
TBI guidelines recommend consideration for ICP monitoring and treat-
ment of intracranial hypertension (< 20 mmHg) for children presenting
with severe TBI (GCS ≤ 8) and stroke guidelines recommend consid-
eration of intracranial monitoring in certain situations (Table 22.1). ICP
monitoring is not currently recommended following cardiac arrest after
Section 3
failure of efficacy in observational studies in drowning patients but may

benefit from contemporary study. Monitoring is typically approached
using intraparenchymal (monitoring only) or intraventricular (monitoring
and treatment) catheters or both in order to continuously monitor and
drain cerebrospinal fluid. Observational studies using brain tissue oxygen
(PbtO2) catheters such as the Licox® report variable utility, and prospec-
tive data are needed.
Other devices that measure cerebral blood flow and metabolism such as
near-infrared spectroscopy, microdialysis, and transcranial Doppler ultra-
sound may offer utility to the neurointensivist, but more study is needed.
Cardiac Arrest
240
The incidence of out-of-hospital pediatric cardiac arrest is 8 per 100,000

person-years with infants having nine times the average pediatric incidence
rate. Cardiac arrests occur at a rate of 0.94 cardiac arrests per 100 ICU
admissions. Corresponding mortality rates are 90% for out-of-hospital and
50% for in-hospital cardiac arrests, with deaths mostly due to neurological
failure. Most events begin with asphyxia or shock (i.e., respiratory failure,
hypovolemia) followed by hypoxemia, bradycardia, and finally global ischemia,
having important pathophysiological differences compared to ventricular
arrhythmia-initiated events. Out- of-
hospital events are frequently unwit-
nessed, with asystole as initial rhythm.
Understanding best practices for the provision of cardiopulmonary re-
suscitation may lead to positive impacts on outcome. There are no proven
neuroprotective therapies to improve outcome postresuscitation. Recently,
the multicenter Therapeutic Hypothermia after Pediatric Cardiac Arrest
trial for out-of-hospital events found no statistical survival with favorable
outcome benefit for children receiving hypothermia versus controlled nor-
mothermia. Postresuscitation guidelines support five days of controlled nor-
mothermia (36°C to 37.5°C) or two days of therapeutic hypothermia (32°C
to 34°C) followed by three days of controlled normothermia. Other sup-
portive recommendations based on observational data in children include
targeting normotension, normoxia, and normocarbia (Table 22.1). Use of
extracorporeal-cardiopulmonary resuscitation is increasing and recommended
for consideration for in-hospital pediatric cardiac arrests due to cardiac eti-

ology in centers with necessary resources.
No one test accurately predicts long-term outcome after pediatric car-
diac arrest, but physical examination findings including pupillary reactivity and
motor exam; clinical details such as initial rhythm, first blood pH, and lactate;
and tests such as EEG, serum brain biomarkers (e.g., S100b), and brain MRI
findings can aid clinicians and families in prognostication; however, all need
prospective validation.
Traumatic Brain and Spinal Cord Injury
Chapter 22
Pediatric TBI accounts for over 50,000 hospitalizations per year, is a leading
cause of morbidity and permanent disability from trauma, and is one of the
leading causes of death in children. Evidence-based guidelines have been
published and updated for the medical management of severe TBI in infants,
children, and adolescents (Table 22.1). For children with GCS ≤8, our pro-
tocol is to intubate, perform central venous and arterial catheterization, and
place an external ventricular drain (EVD) for cerebrospinal fluid (CSF) diver-
sion and a parenchymal ICP sensor for continuous ICP monitoring. In older
children who can have a cranial fixation “bolt” placed (generally >2 years),
241
we place PbtO2 and temperature monitors. We employ a tiered management
approach for patients who develop elevated ICP:
• Tier 1: Endotracheal intubation, central venous catheter, arterial cath-
eter. Head of bed elevated to 30 degrees, head midline. Sedation with
continuous opiate infusion (fentanyl) and muscle relaxation with contin-
uous cisatracurium infusion. In addition, we use continuous EEG moni-
toring and antiseizure prophylaxis with phenytoin. Target: ICP ≤20 mm
Hg, cerebral perfusion pressure (CPP) > 40 mm Hg (or greater in older
children), PbtO2 >25 mm Hg, PaCO2 ≈ 32, Hemoglobin >9 g/dL,
Platelets > 100,000, international normalized ratio (INR) <1.5.
• Tier 2: Hyperosmolar therapy (3% saline) 3–5 mL/kg bolus. If repeat
boluses required, start infusion 0.5–1 mL/kg/hr. Maintain serum sodium
concentration <160 mEq/L.
• Tier 3: Pentobarbital bolus 5 mg/kg. If repeat boluses required, start infu-
sion 1 mg/kg/hr. Titrate to burst suppression on continuous EEG.
• Tier 4: Consider decompressive craniectomy. Consider hypothermia (32°–
34°C) as temporizing measure for refractory ICP elevation while awaiting
surgery.
Abusive head trauma (AHT) is the most common cause of severe TBI in
infants <1 year old in the United States. Overall incidence is estimated at 29.7
per 100,000 children <1 year of age. Infants may present with vague symptoms
such as vomiting without diarrhea and unreliable history by caregivers, thus
requiring a high index of suspicion to make a timely diagnosis. Management
should be based on the TBI guidelines, but frequently children with AHT also
General Management
present with hypoxic-ischemic injury due to seizure, apnea, and cardiac arrest.
Children with AHT have higher risk of seizure than non-AHT TBI patients
(>70% vs. 30%) and a high incidence of seizures refractory to first-line therapy.
Physicians are mandatory reporters if child abuse is suspected.
Spinal cord injury (SCI) in children up to nine years old is more likely to be
associated with high cervical spine injury and simultaneous TBI. In addition,
in younger children with relatively large head size, less muscle development,
and ligamentous laxity, a condition known as spinal cord injury without ra-
Section 3
diographic abnormality (SCIWORA), in which significant SCI occurs without

apparent abnormalities on cervical spine imaging, is possible. As a result, SCI
should be considered in all cases of pediatric TBI, and advanced imaging of the
cervical spine should be considered for cervical spine clearance.
Status Epilepticus
The definition of status epilepticus that is recommended to drive therapeutic
intervention is seizure activity lasting more than five minutes either contin-
uously or intermittently without recovery of consciousness between events.
In addition to etiologies commonly encountered in adults, several additional
242
causes should be considered in children with status epilepticus, including

simple and complex febrile seizure or status epilepticus (≤6 years old), genetic
syndromes, metabolic conditions, accidental ingestion, and AHT. Workup typ-
ically includes complete blood count and metabolic panel and blood glucose
level. Head CT, lumbar puncture, and other infectious studies should be con-
sidered in clinical context. Status epilepticus may be more difficult to detect in
infants, who rarely develop generalized tonic-clonic seizures and may present
with nonspecific and subtle findings such as vital sign changes, automatisms,
eye deviation, or tremor. In patients intubated using muscle relaxants with ac-
tive seizures, it is appropriate to assume ongoing seizure activity and provide
second-line antiepileptic drugs or seizure prophylaxis until EEG is available.
Management:
• ABCs. Supplemental oxygen. Finger-stick glucose and sodium.
• First line: lorazepam 0.1 mg/kg intravenous (IV) up to 4 mg. Repeat once if
needed. Neonates: consider phenobarbital 20 mg/kg.
• Second line: fosphenytoin 20 mgPE/kg IV.
• For refractory status epilepticus, consider midazolam or pentobarbital
infusion.
• Midazolam: Give 0.1 mg/kg bolus (maximum 4 mg) and start infusion at
0.1 mg/kg/hr. Repeat bolus and increase by 0.1 mg/kg/hr every 10 to
15 minutes as needed. Typical maximum infusion = 1 mg/kg/hr.
• Pentobarbital: Give 5 mg/kg bolus and start infusion at 1 mg/kg/hr.
Increase infusion by 0.5 to 1 mg/kg/hr every 15 minutes as needed.
Typical maximum infusion = 3 mg/kg/hr.
• Patients on high-dose midazolam or pentobarbital infusion often require

invasive blood pressure monitoring and vasoactive infusion to maintain
normotension.
Stroke and Arteriovenous Malformations

The incidence of pediatric stroke is 2.3 per 100,000 children per year.
Unlike in adults, pediatric strokes due to ischemic and hemorrhagic events
are approximately equal. Although an underlying cause is not found in up to
one-third of cases, risk factors for arterial ischemic stroke (AIS) include hy-
percoagulable states, genetic syndromes, vasculopathy, vasculitis, and trauma.
Chapter 22
It is important to properly investigate for an underlying disease as treatment
recommendations for anticoagulation vary by type. Tissue plasminogen ac-
tivator and interarterial treatment have been reported in children with AIS
but have no prospective data to support their use. Having an arteriopathy, bi-
lateral hemispheres affected, and altered consciousness at presentation were
prognostic of poor outcome.
Hemorrhagic stroke in children are commonly due to arteriovenous
malformations, intracranial aneurysms, and hemophilia. While brain CT ac-
curately detects hemorrhage, the gold standard for diagnosing the under-
243
lying cause and differentiating the lesion from a mass is cerebral angiography.
Recommended neurocritical care after diagnosing etiology and correcting co-
agulation parameters includes supporting oxygenation and ventilation, con-
trolling systemic hypertension and seizures, and managing increased ICP.
Sinus venous thrombosis presents similarly to other stroke syndromes with
seizures and headache and frequently presents with hemorrhage and hydro-
cephalus (40% and 10%, respectively, in one series) on imaging. Risk factors
include dehydration, infection, perinatal status, and hypercoagulable/hemato-
logical diseases. Hemorrhagic infarcts occurred in 40% of the patients and hy-
drocephalus in 10%. Transverse sinus thrombosis is the most common location.
Strong treatment recommendations in addition to those common to other sup-
portive measures include hydration and antibiotics when bacterial infection is
suspected. Noninvasive and invasive interventions to stabilize or eradicate the
clot have been performed, but there are no high-quality prospective data.
Central Nervous System Infection and

Inflammation
Young children, particularly those <3 months of age, with severe CNS infec-
tion may present with nonspecific symptoms. Meningismus is an uncommon
finding on exam. A high index of suspicion is required and lower threshold for
performing lumbar puncture (LP) to diagnose CNS infection in this age group.
Prior to performing LP, we frequently obtain a head CT to evaluate for space-
occupying lesion or signs of elevated ICP and check a standard coagulation
panel to evaluate for coagulopathy. In patients with depressed mental status
General Management
or focal neurologic deficits, we often delay LP (~17% of patients with ele-

vated ICP may have a normal CT scan) and cover with broad-spectrum
antimicrobials empirically.
Choice and dosing of antibiotics for suspected meningitis should be age-
appropriate. In children <3 months of age, we use ampicillin + cefotaxime
to cover group B streptococcus, E. coli, and L. monocytogenes. In children
>3 months of age, Pneumoccocus, H. influenza, and L. monocytogenes are
more common, and we cover empirically with vancomycin + third-generation
Section 3
cephalosporin. The vancomycin trough should be 15 to 20 μg/mL and the

cephalosporin should be given with meningitic dosing. In cases of acute
encephalitis, despite an extensive workup the etiology is often not identified.
Testing of CSF by polymerase chain reaction for herpes simplex virus (HSV)
and enterovirus is often performed, with acyclovir added empirically until
HSV testing returns negative. Treatment should focus on early initiation of
antibiotics and antivirals. There is limited data on management of neurologic
sequelae.
Management:
• Supportive care
• Continuous EEG. Aggressive management of clinical or electrographic sei-
244
zure. Consider seizure prophylaxis with fosphenytoin or levetiracetam.

• Empiric management of suspected or documented elevated ICP
• Mannitol 250– 500 mg/ kg q6 hr— place Foley. Avoid hypovolemia.
Monitor serum osmolality and hold if >320 mOsm/kg.
• Hypertonic saline—desired sodium range 135 to 150 mEq/L. Deliver
centrally.
• Mechanical ventilation. Maintain PaCO2 35–40 mm Hg. Maintain PaO2
>90 mm Hg.
• In children who develop ventriculitis with hydrocephalus, EVD for CSF
diversion may be required.
Acute disseminated encephalomyelitis (ADEM) is an inflammatory demye-
linating disorder of the CNS that typically occurs following infection or, less
commonly, vaccination. ADEM is more common in young children, the mean
age is seven years, and the incidence of 0.4/100,000 has been increasing in
recent years. Symptoms often start with fever, headache, and motor deficits
and may progress to encephalopathy and coma. MRI typically shows large,
multifocal, white matter T2 hyperintensities.
Management:
• First line: Methylprednisolone 20–30 mg/kg/day for three to five days.
Long taper.
• Second line: intravenous immunoglobulin (IVIG) 2 g/ kg divided over
five days.
Neuromuscular Disorders
Children with acute neuromuscular weakness often require ICU admission for
respiratory monitoring, mechanical respiratory support, management of auto-
nomic dysfunction, and general supportive care. Three acute neuromuscular
disorders that frequently require pediatric neurointensive care include
• Guillain-Barré syndrome––Most common acute neuromuscular disease
admitted to pediatric ICU. Typically presents with symmetric progres-
sive paralysis ascending from the lower extremities. May have severe au-
tonomic instability or respiratory or bulbar compromise. CSF typically
Chapter 22
demonstrates elevated protein and low leukocytes. Treatment: IVIG 2
g/kg divided over five days.
• Myasthenia gravis— Caused by antibodies to acetylcholine receptor.
Typically presents in juvenile period with weakness and fatigability. May
progress to respiratory failure. Treatment: cholinesterase inhibitors,
corticosteroids, IVIG, or plasmapheresis.
• Botulism –Classic presentation is nontoxic appearing previously healthy
child with weakness, poor feeding, and constipation. May require mechan-
ical ventilation due to respiratory muscle weakness and loss of cough or
gag. Treatment with botulism immune globulin in single dose of 50 mg/kg
245
IV and supportive care.
Further Reading
Abend NS, Wagenman KL, Blake TP, et al. Electrographic status epilepticus and
neurobehavioral outcomes in critically ill children. Epilepsy Behav. 2015;49:238–244.
Abend NS, Licht DJ. Predicting outcome in children with hypoxic ischemic encephalop-
athy. Pediatr Crit Care Med. 2008;9:32–39.
Adelson PD, Wisniewski SR, Beca J, et al. Comparison of hypothermia and nor-
mothermia after severe traumatic brain injury in children (Cool Kids): a phase 3,
randomised controlled trial. Lancet Neurol. 2013;12:546–553.
Al-Jarallah A, Al-Rifai MT, Riela AR, Roach ES. Nontraumatic brain hemorrhage in chil-
dren: etiology and presentation. J Child Neurol. 2000;15:284–289.
Allen BB, Chiu YL, Gerber LM, Ghajar J, Greenfield JP. Age-specific cerebral perfusion
pressure thresholds and survival in children and adolescents with severe traumatic
brain injury. Pediatr Crit Care Med. 2014;15:62–70.
Atkins DL, Everson-Stewart S, Sears GK, et al. Epidemiology and outcomes from
out-of-hospital cardiac arrest in children: the Resuscitation Outcomes Consortium
Epistry-Cardiac Arrest. Circulation. 2009;119:1484–1491.
Au AK, Carcillo JA, Clark RSB, Bell MJ. Brain injuries and neurological system failure are
the most common proximate cause of death in children admitted to a pediatric in-
tensive care unit. Pediatr Crit Care Med. 2011;12:566–571.
Bell MJ, Carpenter J, Au AK, et al. Development of a pediatric neurocritical care service.
General Management
Neurocrit Care. 2009;10:4–10.

Fink EL, Berger RP, Clark RSB, et al. Serum biomarkers of brain injury to classify out-
come after pediatric cardiac arrest. Crit Care Med. 2014;42:664–674.
Kochanek PM, Carney N, Adelson PD, et al. Guidelines for the acute medical manage-
ment of severe traumatic brain injury in infants, children, and adolescents—second
edition. Pediatr Crit Care Med. 2012;13(Suppl. 1):S1–S82.
Kumar R, Singhi S, Singhi P, Jayashree M, Bansal A, Bhatti A. Randomized controlled trial
comparing cerebral perfusion pressure-targeted therapy versus intracranial pressure-
targeted therapy for raised intracranial pressure due to acute CNS infections in chil-
Section 3
dren. Crit Care Med. 2014;42:1775–1787.

Moler FW, Silverstein FS, Holubkov R, et al. Therapeutic hypothermia after out-of-
hospital cardiac arrest in children. N Engl J Med. 2015;372:1898–1908.
Moreau JF, Fink EL, Hartman ME, et al. Hospitalizations of children with neurologic
disorders in the United States. Pediatr Crit Care Med. 2013;14:801–810.
Roach ES, Golomb MR, Adams R, et al. Management of stroke in infants and children: a
scientific statement from a Special Writing Group of the American Heart Association
Stroke Council and the Council on Cardiovascular Disease in the Young. Stroke.
2008;39:2644–2691.
Stippler M, Ortiz V, Adelson PD, et al. Brain tissue oxygen monitoring after severe trau-
matic brain injury in children: relationship to outcome and association with other
clinical parameters. J Neurosurg Pediatr. 2012;10:383–391.
246
Chapter 23
Neurocritical Care
in Pregnancy
Krystle Shafer and Marie R. Baldisseri
Introduction
The physiologic changes of pregnancy result in increased risk of many cere-
brovascular disorders. Additionally, the consideration of the developing fetus
results in alterations in diagnostic testing and therapies to treat such conditions.
It is important for neurologists and critical care clinicians to familiarize them-
selves with the care and approach of the pregnant patient for two lives and
not just one. If the mother is clinically deteriorating despite aggressive medical
247
management of the underlying neurological condition, consideration of early
delivery should be discussed with neonatology, anesthesiology, and obstetrical
clinicians. For women who are less than 32 weeks gestation and delivery is
being considered, it is recommended to provide glucocorticoids prior to de-
livery to help with fetal lung maturity.
Thrombotic Events: Ischemic Stroke and

Cerebral Sinus Thrombosis
Physiologic Changes
Pregnancy is associated with a significant increase in clotting factors. Pregnant
women have between a 20% to 1000% increase in their factors VII, VIII, IX, X,
XII, a 400% increase in their von Willebrand factor, and a two-to three-fold
increase in the fibrinogen level. These factors are maximally increased at time
of delivery, indicating that the third trimester and especially the postpartum
period, is arguably the time of maximum vulnerability for these patients. Thus,
the risk of a thrombotic event such as pulmonary embolism, deep venous
thrombosis, cerebral sinus thrombosis, and ischemic stroke are much higher
as compared to nonpregnant patients.
Signs and Symptoms
Pregnant patients who have suffered an ischemic stroke will present with sim-
ilar symptomatology as their nonpregnant counterparts with focal neurologic
deficits based on the areas of ischemia.
Cerebral sinus thrombosis will also present in the traditional fashion. These
General Management
patients have clinical manifestations that include encephalopathy (seizures and

mental status changes), focal deficits, and/or headache. Headache is usually
the earliest and most frequent symptom and in some patients may be the only
symptom.
Diagnostic Testing Consideration
Magnetic resonance imaging (MRI) is considered safe during pregnancy, as it
poses no radiation risk to the fetus. Gadolinium crosses the placenta and has a
Section 3
longer half-life to the fetus as compared to iodinated contrast. Animal data using
gadolinium has suggested teratogenicity when used in high doses. Because of
these data, it is generally not recommended to use gadolinium during pregnancy.
A computed tomography (CT) scan of the head, using a fetus shield during the
scan, exposes the fetus to only <10 mrad of radiation. This is considered a safe
level, as it is not until levels of 5,000 mrad are reached that there becomes a con-
cern for fetal anomalies, growth restriction, intellectual disability, and future pe-
diatric oncogenic risks. Iodinated contrast materials, when given intravenously,
will cross the placenta and can theoretically cause transient affects on fetal thy-
roid function, but there has not been any animal or clinical data suggesting any
sequelae reported from brief exposures. Oral contrast agents do not cross the
placenta and will not cause any fetal harm. Thus, iodinated contrast materials
248
may be used in pregnancy when indicated.

Considering these factors for pregnant patients who present with symptoms
of an ischemic stroke, it is appropriate to consider CT/CT angiography (CTA)
and MRI as per the usual standard of care. Cerebral angiography may also be
beneficial for specific patients. For patients presenting with symptoms con-
cerning for cerebral sinus thrombosis, a MR venography (without the use of
gadolinium) would be the best first-line diagnostic test.
Stroke Treatment
Tissue Plasminogen Activator
Recombinant tissue plasminogen-activator is considered a pregnancy Food
and Drug Administration (FDA) category C medication as there have been
numerous animal studies suggesting adverse effects to the fetus. It is impor-
tant to recognize that many of the drugs that are routinely used in pregnant
patients are category C drugs because of the lack of reliable drug studies in
humans. There are very few human studies regarding tissue plasminogen ac-
tivator (tPA), and the fears may be overstated as the high molecular weight
of tPA prevents it from crossing the placenta. In one small case review of 26
patients, only 3 had pregnancy complications with 1 patient suffering a spon-
taneous miscarriage and another suffering an intrauterine hematoma. None
of the live children born suffered any permanent deficit. Because the data
regarding tPA use in pregnancy is sparse and of poor quality, if tPA is given,
elective pregnancy termination is an option for the patient to consider. The
indications and contraindications for the use of tPA remain the same as the
Neurocritical Care in Pregnancy

general population.
For patients who are >24 weeks gestation, because the fetus is viable, the
consideration of stroke treatment is slightly different. While tPA remains an
option, there needs to be a significant discussion since there is the potential to
cause fetal demise. For those patients with a large vessel occlusion, it is rea-
sonable to consider proceeding directly to endovascular thrombectomy in lieu
of tPA if interventional therapy is available.
Endovascular Thrombectomy
As detailed in Chapter 3, there is strong level I evidence for endovascular
thrombectomy in appropriately selected ischemic stroke patients with a
large vessel occlusion presenting within 24 hours since last seen normal.
While the initial studies that proved the effectiveness of revascularization
Chapter 23
therapy for ischemic stroke did not evaluate its use in pregnant patients,
pregnancy should not be an exclusion for this highly efficacious therapy.
Radiation exposure to the developing fetus can be minimized with appro-
priate shielding. Both unfractionated heparin and low molecular weight
heparin can be safely used as neither of these agents crosses the placenta.
Agents such as clopidogrel, warfarin, and rivaroxaban should not be given to
pregnant patients routinely.
249
Decompressive Hemicraniectomy
As detailed in Chapter 3, decompressive craniectomy can be life saving for
ischemic stroke patients with malignant cerebral edema. The 2007 pooled
analysis of 3 randomized controlled trials showed a significant mortality
benefit (78% medical management, 29% surgical management—absolute
risk reduction 50%) of decompressive surgery for patients 60 years old
or younger with large hemispheric infarction who developed a decrease
in their level of arousal within 48 hours. In these trials, pregnancy was gen-
erally an exclusion criteria. However, there is no reason to hypothesize that
decompressive craniectomy would be any different in pregnancy, the clinician’s
first priority is treating their patient, and the best way to save the fetus is by
providing standard of care therapy to the mother. Thus, if the patient is ap-
propriate for this procedure, it should be strongly considered regardless of
pregnancy.
Aspirin
Low-dose aspirin has been used in pregnancy without causing fetal harm.
Risk of Recurrence (Ischemic Stroke)
For women who suffered an ischemic stroke either during pregnancy or im-
mediately afterwards in the post-partum period, the risk of recurrence is low,
at 1.8% in future pregnancies. However, there is still risk of recurrence even
without subsequent pregnancies. The overall risk of recurrence within one
year is 1% and within five years is 2.3%.
General Management
Cerebral Venous Sinus Thrombosis

Anticoagulation
The mainstay of treatment for these patients is full-dose anticoagulation with
low molecular weight heparin or unfractionated heparin. Small studies have
demonstrated improved outcomes with low molecular weight heparin as
compared to unfractionated heparin, but unfractionated heparin is sometimes
preferred if there is a need for acute reversibility in those patients at high risk
Section 3
of hemorrhage or need for surgical intervention. As mentioned previously,

low molecular weight heparin does not cross the placenta and thus is safe to
use. The duration of therapy should continue for a minimum of six weeks
postpartum for a total minimal duration of therapy of six months.
Endovascular Therapy
The use of mechanical thrombectomy or direct thrombolysis as salvage therapy
for the unstable patient has conflicting data in all patients with cerebral venous
sinus thrombosis, regardless of pregnancy. There is no data specifically looking
at this intervention in the pregnant patient. Thus, this intervention should be
reserved only for patients who have failed anticoagulation therapy and are
experiencing progressive decline with expected poor prognostic outcome and
250
have access to experienced interventionists.

Risk of Recurrence in Future Pregnancies
An analysis of 441 women revealed no recurrent cases of cerebral venous
thrombosis in two women who developed recurrent ischemic stroke during
pregnancy. Thus, this condition is not a contraindication to future pregnancies.
If there is a high concern, low-dose aspirin and/or low molecular weight hep-
arin can be utilized for prophylaxis in future pregnancies.
Reversible Cerebral Vasoconstriction

Syndrome
Reversible Cerebral Vasoconstriction Syndrome has a high prevalence in preg-
nant women and can occur along a spectrum with eclampsia or as an isolated
syndrome. Please review Chapter 6 for a full discussion on the diagnosis and
management of this condition.
There is no current evidence that is consistent with an increased risk of
aneurysmal subarachnoid hemorrhage (SAH) in pregnancy. Those with
vascular lesions that rupture during pregnancy are most likely to expe-

rience this during their third trimester. It is important to recognize this
condition as a potential cause of headache and hypertension in the preg-
nant patient as initial symptoms can mimic preeclampsia. SAH will not
have the same laboratory abnormalities as those patients suffering from
preeclampsia.
Diagnosis with CT, CTA, and/or cerebral angiography should be utilized
the same as in nonpregnant patients, with the only difference being to shield
the fetus during testing. Case reports have demonstrated endovascular
coiling of ruptured aneurysms to be a safe and effective treatment modality
in pregnancy. Blood pressure control, prevention of dehydration, sodium
control, and treatment of hydrocephalus with a ventricular drain should
all be performed in the pregnant SAH patient. The decision to treat with
Chapter 23
nimodipine to prevent potential cerebral vasospasm should be seriously
considered since it is a FDA category C medication. Nimodipine crosses the
placenta, and animal studies have displayed adverse events related to this
medication administration. For patients who develop cerebral vasospasm,
intra-arterial therapy with verapamil has been safely utilized with successful
results.
Although some clinicians will recommend vaginal delivery, others may rec-
ommend prophylactic cesarean delivery to prevent hemodynamic fluctuations
251
and increased ICP during labor and delivery. Prophylactic cesarean delivery is
preferred in the patient with an unsecured aneurysm.
Preeclampsia and Eclampsia

Definition and Pathophysiology
Preeclampsia refers to the acute onset of hypertension (>140/90) and end-
organ dysfunction that occurs after 20 weeks of gestation. Eclampsia refers
to the development of seizures in the setting of preeclampsia. The proposed
pathophysiology is abnormal development of the placenta vascular system
resulting in placenta underperfusion and hypoxia with release of factors that
cause widespread maternal endothelian dysfunction.
Diagnosis and Clinical Symptoms
Patients are diagnosed with preeclampsia if they develop hypertension and
proteinuria (no longer considered an absolute criteria for preeclampsia).
Hypertension is defined as a blood pressure of ≥ 40/90 on two occasions
at least four hours apart or a single value of ≥160/110. Proteinuria is de-
fined as a dipstick value ≥+1 or ≥0.3 g in a 24-hour collection specimen or
a protein to creatinine ratio ≥0.3 mg/mg. In the patient without proteinuria
but hypertension, it is still possible to be diagnosed with preeclampsia if
she develops any of the following: low platelet count (<100,000), serum
creatinine greater than 1.1 mg/dL, liver transaminases twice the upper limit
General Management
of normal, pulmonary edema, visual symptoms (blurred visions, flashing

lights, etc.), or cerebral symptoms (such as headache). These patients
commonly develop HELLP syndrome (hemolysis, elevated liver function
tests, and low platelets).
In addition to the aforementioned symptoms noted in the diagnostic cri-
teria, patients often complain of nausea, vomiting, right upper quadrant
pain, and/or epigastric pain, or they may be asymptomatic. Physical ex-
amination can reveal ankle clonus and reproducible right upper quadrant
Section 3
abdominal pain. Some patients with severe symptoms may experience dis-
seminated intravascular coagulopathy, acute renal failure, placental abrup-
tion, and/or death.
Patients who develop seizures and thus eclampsia most often have
tonic-clonic activity. It is important to note that women who develop gen-
eralized tonic-clonic seizures in the setting of preeclampsia without per-
sistent neurologic deficits do not require any further diagnostic evaluation.
Headaches often precede seizure activity by a day, and seizure activity has
been reported to occur in the postpartum period as well. However in
those women with persistent and refractory seizures and other neurologic
symptoms, it is recommended to obtain a diagnostic CT scan of the brain
or MRI of the brain to rule out other potential etiologies of seizures. In
252
patients who develop symptoms after delivery, the majority of symptoms

will occur within one week after delivery. The majority of patients suffer
from only one seizure episode although some patients will have recurrent
seizure activity and can develop partial seizure activity. Additional physical
exam findings in these patients can include symmetric or asymmetric in-
creased deep tendon reflexes, cranial nerve deficits, memory deficits, and
altered mental status.
Treatment
The definitive management of preeclampsia and especially eclampsia is de-
livery, with severe symptoms as an indication for emergent delivery regard-
less of gestational age. For women with a gestational age of <34 weeks,
betamethasone is recommended to help promote fetal lung maturity prior
to delivery. Fluids should be kept to a minimum due to the risk of devel-
oping pulmonary edema. Blood pressure is best managed with labetalol,
hydralazine, or nifedipine, all of which are safe in pregnancy. Blood pres-
sure management, however, does not decrease the risk of developing
eclampsia. Magnesium sulfate however does reduce this risk by 58%. For
women who develop eclampsia, it is recommended to start and/or con-
tinue magnesium sulfate as opposed to starting a traditional anticonvulsant
or benzodiazepines. Those treated with magnesium have a 67% lower risk
of recurrent seizures as compared to phenytoin and a 52% lower risk as
compared to diazepam. Magnesium is provided in a loading intravenous
dose of 4 to 6 grams over 15 to 20 minutes followed by a maintenance
infusion at 2 gms/hour. Those who have recurrent seizures should receive

an additional intravenous bolus with magnesium (2–4 grams intravenously)
and/or intravenous benzodiazepines. The optimal duration of magnesium
therapy is unclear at this time but typically continues at least through the
postpartum period.
Risk of Recurrence in Future Pregnancies
There is an increased risk of preeclampsia recurrence in subsequent pregnan-
cies, reported as 13.8%.
Seizure Management in the Epileptic

Pregnant Patient
Chapter 23
While pregnancy in itself does not increase seizure risk, use of a
single antiepileptic medication on average doubles the rate of major
fetal malformations, and polytherapy can triple this risk. All antiepi-
leptic medications have the risk of teratogenicity, and thus women who
are seizure-free for a prolonged period may choose to taper off these
medications prior to conception. Other strategies prior to a planned preg-
nancy include tapering to monotherapy, folic acid supplementation, and
253
avoiding valproate, which has higher neurodevelopmental teratogenicity
as compared to alternative agents. Avoidance of phenytoin and phenobar-
bital is also recommended (but with lower quality evidence as compared
to valproate) to prevent reduced cognitive outcomes. For those with un-
planned pregnancies, it is recommended to continue their current medica-
tion regimen without alterations.
Pregnant patients who develop status epilepticus should be treated with
the same algorithm approach as nonpregnant patients. Benzodiazepines
should be given for abortive therapy followed by loading with an AED. As
noted earlier, loading with valproate should not be used as first-line AED
therapy due to its high teratogenicity; levetiracetam is a better first-line
agent in this population. For those patients with severe symptoms, intu-
bation with continuous electroencephalography monitoring and infusion
therapies with propofol and/or midazolam can and should be utilized in
the pregnant patient.
Management of Elevated Intracranial

Pressure (ICP) During Pregnancy
Increased ICP may occur in these patients due to hemorrhagic lesions, infarcts,
and/or edema. Temporizing measures such as mannitol or hypertonic saline
can be considered. Hypertonic saline is arguably safer to use in pregnant
patients as mannitol is considered a category C medication, and there have
been no animal or human studies to investigate its safety. Measures such as hy-
General Management
perventilation should be used with extreme caution as women in their second

and third trimesters normally hyperventilate with a PaCO2 between 30–35;
thus, this intervention will likely not be as useful in this patient population.
As mentioned above, in patients with impending herniation due to a
unilaterial hemispheric lesion, a decompressive craniectomy can be lifesaving.
Pregnancy and Brain Death

Section 3
While a majority of hospitals have developed protocols regarding brain death,

93% do not provide any guidance about fetal management after brain death,
and 99% do not clarify who is responsible for making decisions for the fetus.
This is a complex social, legal, ethical, and economic situation, and one that
arguably should include a hospital’s ethical and legal teams. While state laws
address keeping the mother alive through life support in the event of a severe
brain injury, most are not clear about maintaining cardiopulmonary support
after brain death.
If the decision is made to keep the mother on supportive care after dec-
laration of brain death, it is important that this care is provided in a center
with 24-hour capability for emergent caesarian delivery and neonatology
254
intensive care.
Further Reading
Al-Safi Z, Imudia A, Filetti L, Hobson D, Bahado-Singh R, Awonuga A. Delayed post-
partum preeclampsia and eclampsia: demographics, clinical course, and complication.
Obstet Gynecol. 2011;118(5):1102–1107.
Altman D, Carroli G, Duley L, Farrell B, Moodley J, Neilson J, Smith D. Do women with
pre-eclampsia, and their babies, benefit from magnesium sulphate? The Magpie Trial:
a randomised placebo-controlled trial. Lancet. 2002;359(9321):1877–1890.
American College of Obstetricians and Gynecologists’ Committee Opinion No. 656:
Guidelines for diagnostic imaging during pregnancy and lactation. Obstet Gynecol.
2016; 127(2):e75–e80.
Bujshnell C, McCullough LD, Awad IA, et al. Guidelines for the prevention of stroke in
women. Stroke. 2014;45:1545–1588.
Edlow J, Caplan L, O’Brien K, Tibbles C. Diagnosis of acute neurological emergencies in
pregnant and post-partum women. Lancet Neurol. 2013;12(2):175–85.
Harden C, Meador KJ, Pennell PB, et al. Practice parameter update: management issues
for women with epilepsy-focus on pregnancy (an evidence-based review): teratogen-
esis and perinatal outcomes. Neurology. 2009;73(2):133–141.
Holmes L, Harvey E, Coull B, et al. The teratogenicity of anticonvulsant drugs. N Engl J
Med. 2001;344(15):1132–1138.
Lamy C, Hamon J, Coste J, Mas J. Ischemic stroke in young women: risk of recurrence
during subsequent pregnancies. Neurology. 2000;55(2):269–274.
Lewis A, Varelas P, Greer D. Pregnancy and brain death: lack of guidance in U.S. hospital

policies. Am J Perinatol. 2016;33(14):1382–1387.
Misra U, Kalita J, Chandra S, Kumar B, Bansal V. Low molecular weight heparin versus
unfractionated heparin in cerebral venous sinus thrombosis: a randomized controlled
trial. Eur J Neurol. 2012;19(7):1030–1036.
Robba C, Bacigaluppi S, Bragazzi NL. et al. Aneurysmal subarachnoid hemorrhage
in pregnancy-case series, review, and pooled data analysis. World Neurosurg.
2016;88:383–398.
Selim M, Molina C. The use of tissue plasminogen-activator in pregnancy. Stroke.
2013;44:868–869.
255 Chapter 23
Chapter 24
Management of Intracranial
Hypertension in Fulminant
Hepatic Failure
Sajid Kadir and Raghavan Murugan
Introduction
Fulminant hepatic failure (FHF) is defined by acute onset of encephalopathy
and coagulopathy (international normalized ratio [INR] greater than 1.5)
within 26 weeks of the appearance of symptoms in patients with acute liver
injury and no previous history of underlying liver disease. There are an esti-
256
257
mated 2,000 cases of FHF in the United States yearly with mortality ranging
from 50% to 90% despite intensive care treatment. In the United States,
acetaminophen toxicity and idiosyncratic drug reactions are the most frequent
causes of FHF. Viral hepatitis remains the most common identifiable cause of
FHF in the developing world. Complications of FHF include cerebral edema,
sepsis, acute respiratory distress syndrome, hypoglycemia, coagulopathy, gas-
trointestinal bleeding, pancreatitis, and acute kidney injury. This chapter fo-
cuses on the management of intracranial hypertension in patients with FHF.
Pathophysiology
Cerebral edema and intracranial hypertension complicate approximately 75%
to 80% of patients with FHF and grade III or IV encephalopathy, in whom it
remains a leading cause of death. Although several mechanisms have been
thought to contribute to the development of intracranial hypertension in
patients with FHF, they are only partially understood. Typically, early in the
course of illness, high-grade encephalopathy is coupled with reduction in
cerebral metabolic rate and cerebral blood flow (CBF). As encephalopathy
progresses, there is cerebral vasodilatation resulting in increased CBF and ce-
rebral edema. Finally, in the preterminal phase of illness, marked reduction in
CBF occurs due to cerebral herniation.
Currently two hypotheses prevail that are thought to play a major role in
the pathophysiology of intracranial hypertension: the ammonia/glutamine
hypothesis and the cerebral vasodilatation hypothesis. Ammonia is highly
Fulminant Hepatic Failure

neurotoxic and is normally metabolized in the liver to urea and glutamine.
FHF is associated with excessive circulating ammonia levels, with ammonia
crossing the blood brain barrier. Within the brain, ammonia combines with
excitatory neurotransmitter glutamate to produce glutamine catalyzed by the
enzyme glutamine synthetase. Osmotic and neurotoxic effects of glutamine
in astrocytes account for the development of cerebral edema. Ammonia has
been shown to induce membrane depolarization, calcium influx, glutamate re-
lease, activation of lipases, proteases, and production of free radicals, which
cause neuronal protein nitration and mitochondrial damage. During FHF, arte-
rial ammonia level above 200 μg/dL in grade III and grade IV encephalopathy
Chapter 24
is a strong predictor of brain herniation.
There is mounting evidence that cerebral vasodilation and increased CBF
play a major role in the development of high intracranial pressure in patients
with FHF. Failure of CBF autoregulation with consequent development of ce-
rebral hyperemia, edema, and intracranial hypertension is typically seen late
in the course of encephalopathy. The cerebral vasodilatation in patients with
FHF may result from substances produced within the brain itself, suggesting
that cerebral circulatory alterations in FHF result from metabolic changes in
astrocytes and/or neurons. Other mechanisms have also been implicated, in-
cluding increased production of nitric oxide, prostaglandins, and eicosanoids
257
that contribute to high CBF.
Clinical Manifestations
The initial symptoms of FHF can include loss of appetite, malaise, nausea,
and altered sleep wake cycle. Changes in mentation such as confusion and
agitation are common initial symptoms and can predate the precipitous de-
cline to coma. Typical time course in acetaminophen overdose is encephalop-
athy on day 3 after ingestion with rapid progression to coma within 24 to 48
hours. Encephalopathy and coagulopathy (manifested by prolonged INR) are
key features in FHF. The grades of hepatic encephalopathy are listed in Box
24.1. Clinical signs of intracranial hypertension are marked in grade III and IV
Box 24.1 Hepatic Encephalopathy Grading System

Grade Clinical features
I Changes in behavior, mild confusion, slurred speech, abnormal sleep-wake
cycle
II Lethargy, moderate confusion
III Marked confusion, incoherent speech
IV Coma and unresponsive to pain
encephalopathy. These include systemic hypertension, bradycardia, pupillary
General Management
abnormalities, decerebrate posturing, epileptiform activity, and abnormal

respiratory patterns. Thus, patients presenting with mild encephalopathy and
coagulopathy, along with rising aspartate transaminase and alanine transami-
nase should be transferred immediately to an intensive care unit at a center
with the capability of performing liver transplantation.
Diagnostic Testing
Section 3
Intracranial Pressure Monitoring

There is a paucity of data with regard to monitoring ICP in FHF patients.
Accompanying the lack of evidence is the risk of bleeding in coagulopathic
patients. ICP monitoring can potentially identify intracranial hypertension,
but it has never been shown to decrease mortality. Four types of catheters
have been used to measure ICP: epidural, subdural, parenchymal, and in-
traventricular catheters. Epidural catheters have lower complication rates
(3.8%) compared to subdural bolts (20%) or parenchymal catheters (22%).
Intraventricular catheters may be associated with fatal hemorrhage. The
major complications from ICP monitoring are infection and bleeding. The
use of ICP monitoring in this patient population remains controversial,
258
and practices vary among US centers. If used, any existing coagulopathy

should be corrected prior to catheter placement to attempt to reduce the
chance of intracranial hemorrhage, and the ICP goal should be less than
20 mm Hg.
Monitoring Cerebral Blood Flow
As there are no robust tools to directly measure CBF, indirect measurements
have been proposed. The most common measurement is cerebral perfu-
sion pressure (CPP), which is the difference between mean arterial pressure
(MAP) and ICP. This is an indirect measurement of CBF, and the typical goal
is to maintain a CPP of greater than 50 to 60 mm Hg. If an ICP monitor is not
used, then a goal to maintain MAP greater than 70 to 80 mm Hg has been
suggested to maintain an adequate CPP in the setting of presumed ICP ele-
vation. It must be stated that this theoretical assumption may be inaccurate
in FHF due to variability in cerebrovascular resistance, CBF, and medication
effects.
Transcranial Doppler (TCD) noninvasively measures the velocity of blood
flow in the proximal cerebral circulation, usually in the middle cerebral artery.
Blood flow velocity can be used as a crude tool to detect elevated ICP based
on characteristic changes in waveforms as it relates to changes in blood
flow and cerebral resistance. Although TCDs can give the clinician an idea
about CBF and resistance as well as ICP, meaningful data regarding its utility,
frequency of monitoring, and how TCD data should be used in FHF patients
to alter therapy is lacking.
Other monitoring techniques that have been considered include near in-

frared spectroscopy and novel imaging techniques that measure brain water.
None of these tools have been validated.
Jugular Venous Oximetry
Jugular venous oximetry involves measuring venous oxygenation in the supe-
rior jugular bulb of the internal jugular vein. This can be used as a surrogate
of cerebral oxygen consumption and delivery. This may be particularly useful
in FHF patients where ICP monitor placement is risky since intracranial hyper-
tension, cerebral edema, and disorders of CBF are global rather than focal
processes. The arterio-jugulovenous oxygen difference (AVjDO2), which is the
Chapter 24
arterial oxygen levels minus the jugular bulb oxygen levels, changes in response
to changes in CBF. Normal AVjDO2 is 4 to 6 mL/100 mL. AVjDO2 is meas
ured simply by sending serum samples drawn at the same time from both the
arterial line and the jugular bulb catheter for blood gas analysis to determine
the respective oxygen values. An early indicator of cerebral hyperemia is a de-
crease in the transcranial oxygen content difference (arterial oxygen content −
jugular bulb oxygen content) to less than 4 mL/100 mL. A widened AVjDO2
difference greater than 6 mL/100 ml indicates cerebral ischemia. Similarly, de-
creased jugular bulb venous oxygen saturation of less than 50% can indicate
cerebral ischemia, while venous oxygen saturation greater than 85% indicates
259
either decreased cerebral metabolic demands or cerebral hyperemia.
Computed Tomography
Computed tomography (CT) of the head is insensitive to intracranial hyperten-
sion in the early stages and thus not useful. However, it can detect cerebral edema
in FHF patients with advanced-stage hepatic encephalopathy, showing diffuse
narrowing of sulci and compression and effacement ventricles and cisterns. While
CT scans do not provide a reliable assessment of ICP, they are useful to rule out
other possible causes of altered mental status such as intracranial hemorrhage.
Electroencephalography Monitoring
Overt or subclinical seizure activity can contribute to cerebral hypoxia and
the development or worsening of cerebral edema. Electroencephalography
(EEG) is vital in detecting subclinical seizure activity in patients with grade
III or IV hepatic encephalopathy. If a routine EEG shows any risk factors for
seizures, continuous EEG is recommended as prolonged monitoring may allow
identification of subclinical events.
Management
General Measures
The treatment of encephalopathy associated with FHF is directed at limiting
gut ammonia production. Elevated arterial ammonia level correlates with
the development of encephalopathy. The risk of intracranial hypertension is
General Management
greatest with a sustained ammonia level of 150 to 200 μmol per liter. Lactulose
(30 g every one to two hours) can be used in the treatment of patients with
grade I or II encephalopathy even though clinical trials have not demonstrated
a survival benefit. Lactulose can cause dehydration by inducing profuse diar-
rhea, hypernatremia, and ileus. Rifaximin is a nonabsorbable antibiotic that has
been used in combination with lactulose. Modalities such as continuous high-
volume hemofiltration can achieve clinically significant reductions in circulating
ammonia, without a proven survival benefit.
Section 3
General measures and treatment for patients with intracranial hyperten-

sion are summarized in Box 24.2. Treatment should start with the basics of
normalization of vital signs and laboratory values as able, positioning, and se-
dation. As in other conditions, treatments become more aggressive as patient
worsens, with grade III/IV encephalopathy warranting more aggressive moni-
toring and interventions. Figure 24.1 shows a suggested algorithm for manage-
ment of intracranial hypertension based on AVjDO2.
Box 24.2 Management of Cerebral Edema in Patients

with Intracranial Hypertension
260
General measures
• Early endotracheal intubation for airway protection in patients with grade III or IV
encephalopathy
• Maintain head position at 30° upright angle to optimize jugular venous outflow
• Avoid frequent stimulation and suctioning to prevent acute surges in intracranial pressure
• Avoid hypovolemia and hypervolemia
• Avoid hypertension
• Avoid hypoxemia, hypoglycemia, and hypercapnia
• Maintain ICP <20 mmHg
• Maintain CPP >50 mmHg
• Monitor and maintain SvJO2 between 50% to 85%
• Continuous EEG monitoring
• CRRT for oliguric acute kidney injury
Treatment of intracranial hypertension

• Propofol sedation to reduce cerebral metabolism
• Hypertonic saline bolus to maintain serum sodium between 145 and 155 mEq/L
• Hyperventilation titrated to PaCO2 between 30 and 35 mmHg
• Induced moderate therapeutic hypothermia to 32°–33°C
• Mannitol bolus 1 g/kg body weight to maintain osmolality >320 mOsm/L
• Pentobarbital coma titrated to burst suppression of 5 to 10 cycles/second
Unproven therapies
• Total hepatectomy as a bridge to transplant
Arterio-jugulovenous oxygen difference (AvjDO2)
Normal 4–6 mL/100 mL
Cerebral hyperemia with Cerebral hypoperfusion
Chapter 24
decreased AVjDO2 with increased AVjDO2
Avoid arterial hypotension

Hyperventilation (Goal
Mannitol 0.5–1 gm/kg bolus
PaCO2 of 30–35 mm Hg)
Hypertonic saline, bolus of
Hypothermia of 34–35° C 30 ml of 23.4% or continuous
261
Barbiturate coma, EEG to infusion of 3%.
burst suppression pattern Target sodium
of 145–155 mEg/L
Figure 24.1 Suggested algorithm for AVjDO2-based management of intracranial

hypertension in patients with fulminant hepatic failure
Propofol Sedation
Propofol has beneficial properties in patients with FHF. It reduces cerebral
metabolism by approximately 40% in a dose-dependent manner. It can result
in a lower ICP, potentiation of GABA inhibition, and inhibition of excitatory
NMDA glutamate receptors. Many physicians favor the use of propofol in
patients with FHF for these reasons.
Hyperventilation
Hyperventilation is a temporizing measure to lower ICP in critical elevations,
but due to the resultant cerebral vasoconstriction it should not be maintained
for more than four to six hours to avoid cerebral ischemia. Hyperventilation
reduces CBF by 2% to 3% for every mm Hg reduction in PaCO2. It is im-
portant to identify the need for a long-term management plan (see later dis-
cussion) and follow the A-V oxygen content difference. Excessive cerebral
vasoconstriction can be detected by widening of the cerebral A-V oxygen
content difference.
Hyperosmolar Therapy
General Management
Hypertonic Saline
Hypertonic saline (HTS) reduces ICP and may confer benefit in patients with
FHF. HTS improves regional cerebral perfusion and improves microvascular
blood flow. HTS has been shown to achieve a reduction in ICP in patients
with FHF and grade III or IV encephalopathy. In FHF patients at highest risk for
cerebral edema (serum ammonia greater than 150 μM, grade III/IV hepatic
encephalopathy, acute renal failure, shock requiring vasopressors to maintain
mean arterial pressure), the prophylactic induction of hypernatremia with
Section 3
hypertonic saline to a sodium level of 145 to 155 mEq/L is recommended.

Dosing strategies for hypertonic saline include 30 ml of 23.4% saline boluses
repeated every three to four hours or an infusion of HTS (3% or 7.5% at
20–50 cc/hour) to achieve the targeted serum sodium level. Serum sodium
levels should be monitored every four to six hours to confirm maintenance
of target range. Among patients with acute kidney injury receiving continuous
renal replacement therapy (CRRT), increasing the dialysate sodium concentra-
tion may also achieve target serum sodium concentration.
Mannitol
Mannitol has also been shown to reduce high ICP in patients with FHF and can
be used in management of elevated ICP. It has not been subject to random-
262
ized trials. The typical dose is 0.5 to 1.0 g/kg intravenously. Serum osmolality
should be measured every six hours. Mannitol boluses may be repeated if ICP
remains greater than 25 mm Hg and the osmolar gap is less than 20 mOsm/
kg. An inherent adverse effect is hypovolemia due to mannitol’s diuretic effect.
Prophylactic administration of mannitol is not recommended. Some clinicians
prefer HTS over mannitol in the setting of AKI given the fact that mannitol has
a diuretic effect and may worsen renal injury by causing intravascular volume
depletion.
Hypothermia
Therapeutic hypothermia has been used for years to treat cerebral edema
and intracranial hypertension in several patient populations; however, defini-
tive evidence for benefit is lacking in patients with FHF. Significant reductions
in CSF ammonia levels have been observed with moderate induced hypo-
thermia. Therapeutic hypothermia (core body temperature of 34°–35°C)
may be considered for intracranial hypertension refractory to osmotic agents.
This has been shown to facilitate the bridge to liver transplantation.
Barbiturates
Barbiturate coma can be induced by using pentobarbital, with a bolus of 3 to
5 mg/kg followed by a continuous infusion of 0.2 to 1.0 mg/kg/hour, guided
by continuous EEG monitoring to achieve 5-to 10-second EEG burst sup-
pression. This can reduce cerebral metabolism and CBF in patients with ele-
vated ICP refractory to the previously mentioned therapies. Pentobarbital is
a negative inotrope, and patients frequently require inotropic and vasopressor

support.
Prognosis
Predictive models attempt to identify patients who are likely to benefit from
liver transplantation. Currently available prognostic scoring systems do not
adequately determine candidacy for liver transplantation. King’s College
Criteria are the most widely used for selecting patients for liver transplanta-
tion. The Model for End-Stage Liver Disease (MELD) score has been applied
to patients with acute liver failure. The Sequential Organ Failure Assessment
(SOFA score), the Cliche criteria, and the Acute Liver Failure Study Group
Chapter 24
(ALFSG) index also predict mortality in patients with acute liver failure.
Transplantation
The decision to list for liver transplantation depends on the probability of
spontaneous hepatic recovery. Factors taken into consideration include degree
of encephalopathy, patient’s age, etiology of FHF, and any contraindications to
transplantation. Current rates of survival after transplantation are 79% at one
year and 72% at five years.
Conclusion
263
The complexity of FHF requires the physician to employ a combination of
therapeutic modalities to manage intracranial hypertension. It is therefore cru-
cial for the critical care physician to be familiar with the most effective strate-
gies to manage intracranial hypertension in patients with FHF.
Further Reading
Blei AT. Pathophysiology of brain edema in fulminant hepatic failure , revisited. Metab
Brain Dis. 2001;16:85–94.
Blei AT, Olafsson S, Webster S, Levy R. Complications of intracranial pressure moni-
toring in fulminant. Lancet. 1993 Jan 16;341(8838):157–158
Clemmesen JO, Larsen FS, Kondrup J, et al. Cerebral herniation in patients with
acute liver failure is correlated with arterial ammonia concentration. Hepatology.
1999;29:648–653.
De Riva N, Budohoski KP. Transcranial Doppler pulsatility index: what it is and what it
isn’t. Neurocrit Care. 2012 Aug;17(1):58–66
Hoofnagle JH, Carithers RL Jr, Shapiro C, Ascher N. Fulminant hepatic failure: summary
of a workshop. Hepatology. 1995 Jan;21(1):240–252.
Jalan R, Damink SWMO, Deutz NE, et al. Moderate hypothermia prevents cerebral
hyperemia and increase in intracranial pressure in patients undergoing liver transplan-
tation for acute liver failure. Transplantation. 2003;75:2034–2039.
Larsen FS, Ejlersen E, Clemmesen JO, Kirkegaard P, Hansen BA. Preservation of cere-
General Management
bral oxidative metabolism in fulminant hepatic failure: an autoregulation study. Liver

Transpl Surg. 1996;2:348–353.
Lee WM, Larson AM, R, Stravitz T. AASLD position paper: the management of acute
liver failure. American Association for the Study of Liver Diseases; 2011.
Rutherford A, King LY, Hynan LS, Vedvyas C, Lin W, Lee WM, Chung RT, ALF Study
Group. Development of an accurate index for predicting outcomes of patients with
acute liver failure. Gastroenterology. 2012 Nov;143(5):1237–1243.
Stephan H, Sonntag H, Schenk HD, Kohlhausen S. Effect of Disoprivan (propofol) on
the circulation and oxygen consumption of the brain and CO2 reactivity of brain
Section 3
vessels in the human. Anaesthesist. 1987;36:60–65.

264
Chapter 25
Prognostication and Ethics

David Y. Hwang and Douglas B. White
Prognostication
Importance
Accurate prognostication is no less important than diagnosis and therapy in
neurocritical care. Oftentimes, early impressions of neurologic prognosis—
whether accurate or inaccurate— heavily influence crucial decisions that
clinicians, patients, and patient surrogates make and can impact outcome.
Despite its importance, accurate prognostication presents multiple challenges
for the neurointensivist, including the appropriate use of available outcome
prediction calculators and optimal communication of prognosis to patient
265
surrogates.
General Challenges
Wide Variability in Limiting Life-Sustaining Therapies
Emerging evidence has shown widespread variability in how frequently dif-
ferent centers defer aggressive care in common neurocritical care situations,
a finding which is likely related to variability in how centers prognosticate
outcome.
• A recent study of six Canadian level 1 trauma centers showed that,
among patients with severe traumatic brain injury, the percentage of
deaths associated with limitation of life-sustaining therapy ranged from
45% to 87% (p < 0.001), with one-half of these deaths occurring within
the first three days of care.
• Similarly, a retrospective analysis of data collected by the Agency for
Healthcare Research and Quality showed that variation in feeding tube
insertion rates for patients hospitalized with ischemic stroke, across US
hospitals with large stroke volumes, is high (Figure 25.1).
Clinical Nihilism and “Self-Fulfilling Prophesies”
In practice, there is often considerable pressure when managing
neuroemergencies to identify patients with poor prognosis early on and
avoid aggressive treatment that may be potentially inappropriate or unethical.
Patients who present with severe neurologic conditions are thus at very high
General Management
30
Stroke patients with feeding tube insertion (%)
20
Section 3
10
0
0 200 400 600 800 1000 1200 1400 1600
Hospitals ordered by frequency of feeding tube use (n = 1540)
Figure 25.1 Rank order of frequency of feeding tube placement in acute ischemic stroke
hospitalizations from 2008–2011.
266
The x-axis includes 1,540 individual hospitals with at least 30 acute ischemic stroke hospitalizations lasting
greater than three days with each year of the National Inpatient Sample from 2008–2011. Reprinted with
permission from George et al. Current practices in feeding tube placement for US acute ischemic stroke
inpatients. Neurology. 2014;83:874–882.
risk for having their care influenced by a “self-fulfilling prophesy,” or a pre-

diction that becomes true because its perceived inevitability evokes behavior
from clinicians that is actually in turn responsible for the predicted outcome.
This inherent risk is compounded by the fact that outcome prediction models
based on observational cohorts, in which a sizable number of patients died
after decisions to limit life-sustaining therapy, may themselves be biased in this
fashion. When prognostic uncertainty exists and future quality of life that may
be acceptable to a patient is not initially clear, it is advisable to initially defer
early limitations of care and placement of do not resuscitate orders, especially
when neurocritically ill patients first arrive in the emergency departments.
Other Pitfalls in Clinician Accuracy
When one is formulating prognosis in neurocritical care, it is best to be as self-
aware as possible regarding how personal and/or institutional factors, aside
from clinical nihilism, may bias one’s prediction.
• Being overconfident in one’s prediction for a patient may lead to insuf-
ficient acknowledgment of prognostic uncertainty, while having limited
experience in a patient’s clinical condition may lead to too much uncer-
tainty in the prognosis.
• Overestimating how a future functional and/or cognitive deficit will affect

a patient’s quality of life (i.e., a “focusing illusion” bias) may contribute to
prognostic pessimism when discussing possible outcomes with patients’
families, while focusing mostly on the chances for survival may lead to
overly optimistic predictions.
Fundamental Uncertainty
Even with possible biases in outcome prediction aside, a degree of uncertainty
is simply inherent in neurologic prognostication. In general, prognosis should
be expressed as a range of outcomes in discussions with patients’ families.
While uncertainty can often make decisions regarding goals of care challenging
with families, honestly acknowledging the uncertainty with a patient’s family
Chapter 25
has been shown to be preferable and can build clinician-family trust.
Outcome Prediction Tools
Multiple models for predicting patient outcome in common neurocritical care
situations have been derived from cohort studies. Table 25.1 highlights a few
commonly known prediction tools.
Limitations of Available Tools
It is critical that the prognostic information obtained from outcome prediction
tools such as those listed in Table 25.1 be applied to individual patients with
267
great caution.
• Many of the tools were derived from patient cohorts without standard-
ized protocols for withdrawal of care practices, a fact that could influ-
ence mortality rates, especially for more severely affected patients.
• Very few of these tools have been compared with early subjective clinical
judgment of clinicians with regards to their accuracy. A study of outcome
prediction for 121 patients with intracerebral hemorrhage found that subjec-
tive physician and nursing predictions were more accurate than ICH score
predictions with regards to correlation with three-month functional outcome.
Optimal Communication of Prognosis and Risk
Families’ Perspectives on Prognostic Information
Despite a clinician’s best efforts to formulate an accurate prognosis for a pa-
tient in the face of uncertainty, optimal discussions regarding prognosis with
patients’ families involves recognition that a physician’s formulation alone
may constitute a small fraction of a family member’s ultimate impression of a
patient’s prognosis.
• A mixed-methods study of 179 surrogate decision-makers in multiple
intensive care units (ICUs) at a single center revealed that only 2% of
surrogates reported basing their own view of patient prognosis on phy-
sician estimates alone, while less than half (47%) incorporated physician
estimates to some degree.
• Other information that family members reported incorporating into their
own prognostic impressions included the patient’s physical appearance
(64%); their own intrinsic optimism, intuition, or faith (36%); the patient’s
Table 25.1 Outcome Prediction Tools in Neurocritical Care
Disease Name of Tools Variables Included Outcomes for which Tool Is Designed
Ischemic stroke iScore Age, sex, stroke severity, stroke subtype, cardiac risk 30-day and 1-year mortality risk
factors, preadmission disability, admission glucose, prior Risk of severe functional disability at 30 days
cancer, prior dialysis
Intracerebral ICH Score Age, Glasgow Coma Scale, ICH volume, presence of Originally 30-day mortality risk; now validated for 12-
hemorrhage intraventricular hemorrhage, infratentorial ICH location month functional outcome
Aneurysmal Hunt-Hess Scale; Hunt-Hess: Neurologic exam Mortality risk and functional outcome at end of
subarachnoid WFNS grade WFNS: Glasgow Coma Scale hospitalization
hemorrhage
Traumatic brain CRASH; IMPACT CRASH: Country, age, Glasgow Coma Scale, pupil CRASH: 14-day mortality risk; risk of unfavorable
injury reactivity, extra-cranial injury outcome at six months
IMPACT: Age, motor score, pupil reactivity, vital IMPACT: Risk of mortality and unfavorable outcome
signs, CT scan characteristics, admission glucose and at six months
hemoglobin
ICH = intracerebral hemorrhage; WFNS = World Federation of Neurosurgical Societies; CT = computed tomography.
“will to live” (27%); the patient’s survival of prior illnesses (28%); and the

power of their own bedside presence (13%).
Clinicians’ Approaches to Prognostic Discussions
Compounding the challenge of patients’ families using multiple sources of in-
formation for formulating prognosis is the fact that ICU physicians have been
shown to have wide variability with regards to the actual language and ap-
proach used in prognostic discussions with families.
• A study of 50 audiotaped physician-family ICU conferences in which
patient prognosis was discussed revealed that only 20% of physicians
gave quantitative predictions during their discussions.
Chapter 25
• Only 28% of physicians cited statistical information from published re-
search that they then also tailored to a patient’s specific clinical situation.
Furthermore, only 14% of physicians explicitly checked whether families
understood the information that they were given.
Recommendations for Prognosis and Risk Discussions
Among others, Fagerlin et al. have published best-practice recommendations
for discussions regarding risk with patients and, by extension, surrogate
decision-makers. The evidence for the recommendations in Box 25.1 has been
adapted from studies of cancer patients, but many of the concepts may apply
to ICU situations where communicating “high-stakes” risk (e.g., likelihood for
269
poor functional outcome or death) and any potential benefit of life-sustaining
treatment as accurately as possible is critical.
Ethics
Shared Decision-Making
The ethical principles of respect for persons and respect for autonomy guides
the practice of shared decision-making in neurocritical care. Because patients
often lack decision-making capacity, a large part of a neurointensivist’s routine
approach to shared decision-making involves working with surrogate decision-
makers and incorporating patients’ advance care planning documentation into
decisions when possible.
Clinicians make countless decisions for a typical neuroscience ICU patient
on a daily basis, but for decisions that are major and heavily value-laden as
opposed to largely grounded in biomedical facts, a model of collaborative
shared decision-making is most relevant. Figure 25.2 illustrates how effective
shared decision-making occurs along a continuum of patient (or proxy) au-
tonomy to clinician paternalism depending on the situation, after the decisions
at stake have been clearly defined and all involved interdisciplinary ICU team
members have provided their perspectives.
Advanced Care Planning
Table 25.2 summarizes the most common types of US advanced care planning
documentation encountered in neurocritical care practice.
General Management
Box 25.1 Summary of General Recommendations for Risk

Communication in Medicine
1. Ask for permission before conveying news of a poor prognosis.
2. Use plain language to make written and verbal materials more understandable.
3. When possible, present data using absolute risks (as opposed to relative risks)
especially when discussing the impact of a potential treatment plan.
4. Present information in pictographs if including graphical information.
5. Present data using frequencies (as opposed to percentages)—percentages are more
abstract and are associated with less risk perception.
Section 3
6. Be aware that the order in which risks and benefits are presented for a treatment plan
can affect risk perceptions. When risks are presented after benefits, the listener gen-
erally perceives risks as more worrisome and more common.
7. Recognize that comparative risk information (e.g., what the “average” person’s risk is,
results from prognostic outcome tools, etc.) is persuasive and not just informative.
8. Consider presenting only the information that is most critical to decision-making, even
at the expense of completeness.
9. Repeatedly draw attention to the time interval over which a risk occurs (e.g., duration
that poor functional outcome is expected before possible recovery, etc.).
10. Provide emotional support that is appropriate for the listener’s reaction(s).
11. Check for understanding of the information that has been communicated at the end of
the discussion.
Adapted from Fagerlin A, Zikmund-Fisher BJ, Ubel PA. Helping patients decide: ten steps to better risk
270
communication. J Natl Cancer Inst. 2011;103:1436–1443.
What decision needs to be made? What are medically

appropriate treatments, alternatives, pros/cons, uncertainties?
Involvement of others
Chaplaincy, Social Work, Outpatient Providers,
Palliative Care?
Patient/Surrogate Medical Team
Decides Decides
Eliciting patient’s values/
preferences; deliberation
AUTONOMY SHARED PATERNALISM

DECISION
Figure 25.2 Model of shared decision-making

Adapted from Cai et al. Patient preferences and surrogate decision making in neuroscience intensive care
units. Neurocrit Care. 2015;23:131–141.
Table 25.2 Types of Advanced Care Planning Documents
Name Purpose Applicable Situation Items Commonly Covered Creating Parties
Living will States patient’s preferences for either Patient in coma or minimally Medical machinery and techniques which 1. Patient
potential future desired treatments or conscious state, etc., where it may extend life but not necessary cure 2. Lawyer
treatments that the patient would wish is felt that he or she will not underlying medical condition:
to be withheld or withdrawn. be able to regain capacity for – Cardiopulmonary resuscitation
decision-making.
– Mechanical ventilation
– Artificial nutrition via enteral or
parenteral route
– Dialysis
– Organ and tissue donation
Healthcare Designates a person to give him or her Patient is unable to make or – Decisions regarding resuscitation, 1. Patient
power of broad authority to make health care communicate a choice about treatments, procedures, tests, and 2. Lawyer may assist in drafting,
attorney decisions for a potential patient. a health care decision, in medications but many states have
the opinion of an attending –Access to medical records and information template forms to complete
physician. that only require witnesses
– Authorization for admission or
discharge, including discharge against
medical advice from facilities
– Ability to contract for health care–
related service and “hire or fire” medical
and support personnel
– Organ and tissue donation, including
autopsy
Physician orders Serves as a standing medical order Immediately upon signing by – DNR order 1. Patient or patient’s
for life-sustaining indicating a patient’s wishes regarding physician or other health care – Do not intubate order surrogate decision-maker
treatment specific current treatments (e.g., DNR); provider. Generally intended 2. Physician or other health
– Acceptability of artificial nutrition,
follows patient across health care for patients with a known care provider
including duration
settings. Not otherwise intended to serious medical illness at the
replace other advance directives such end of life.
as a health care power of attorney.
DNR = do not resuscitate.
Adapted from Cai et al. Patient preferences and surrogate decision making in neuroscience intensive care units. Neurocrit Care. 2015;23:131–141.
While advanced care planning documentation can be useful when available,
General Management
limitations exist.
• The vast majority of patients do not have pre-existing advanced care
planning documentation at the time of ICU admission.
• The language contained in documents such as living wills often lacks the
specificity needed to effectively address a particular clinical scenario or
medical decision.
Surrogate Decision-Making
Section 3
Concept of Substitutive Judgment

When an incapacitated patient’s previously stated preferences are not directly
available, the next best option is for a medical team to work with a com-
petent surrogate decision-maker to ensure that the patient’s own values are
incorporated into preference-sensitive decisions. This process of “substitutive
judgment” is not without many practical challenges, not the least of which are
identifying a proper surrogate (a process for which many states have a defined
hierarchy of relatives), ensuring that the surrogate also has proper capacity,
and assessing whether a surrogate truly has the patient’s own values in mind.
During family meetings, it is important to focus on eliciting the patient’s prior
preferences and values as opposed to the surrogate’s own preferences. Open
dialogue with a clear discussion of the pros and cons of the options involved
272
in any shared decision is critical for promoting careful deliberation.

Surrogates’ Preferred Roles in Shared Decision-Making
Making a shared medical decision for an ICU patient with a surrogate falls
along a continuum between complete autonomy and complete paternalism,
and various surrogates have different preferences and personal comfort levels
with their degree of involvement regarding medical decisions. However,
for those decisions that are heavily “value-sensitive” (e.g., continuation or
withdrawal of life support), as opposed to simply technical, the majority
of surrogates do wish to have final control. In a US study of 51 audiotaped
ICU family conferences during which decisions regarding life support were
addressed, four physician approaches to decision-making were empirically
observed: (a) informative, in which the physician provided medical informa-
tion but did not elicit information regarding the patient’s values or provide a
recommendation regarding life support; (b) facilitative, in which a physician
guided a surrogate through the process of identifying a patient’s values but in
which a recommendation was not given; (c) collaborative, in which the physi-
cian provided a recommendation during discussion; and (d) directive, in which
the physician assumed complete responsibility for the decision. Despite these
different approaches, the descriptions of which were validated in a separate
cohort of recorded conferences, no clinician in this derivation cohort asked
about surrogates’ preferred role in decision-making. The authors concluded
that eliciting surrogates’ preferred roles in shared decision-making situations
represents an area for improvement of optimal ICU communication.
Absence of a Surrogate Decision-Maker

In one single center medical ICU study, 16% of admitted patients lacked both
decision-making capacity and a surrogate decision-maker. For patients in this
group for whom decisions about goals of care were made, wide variability
was seen in terms of the medical team’s involvement of a second medical
opinion before decision-making, with only 11% of patients having a court or
hospital ethics committee involved in their situation. Hospitals should have
clear protocols for managing decision-making in such situations that involve
their local ethics committee and avoid unilateral decisions by medical teams
acting alone.
Chapter 25
Requests for Potentially Inappropriate Treatment
A multisociety document published by the American Thoracic Society (ATS)
provides an extensive policy statement regarding the management of requests
for “potentially inappropriate treatment” in ICU medicine.
Definitions
The distinction between an ICU intervention that is strictly “futile” as opposed
to “potentially inappropriate” is important. “Futile” interventions are those
that simply cannot accomplish an intended physiologic goal. In contrast, “po-
tentially inappropriate treatments” have some chance of accomplishing the
effect sought by the requesting party, but ethical considerations may justify
273
not providing them.
Recommendations
This distinction between terms is critical in practice, as there is general
agreement that clinicians should simply not provide strictly futile interventions.
Not providing futile care is important for maintaining the integrity of the
physician and nursing professions and for upholding basic ethical principles
in caring for patients (i.e., do no harm). If, despite best efforts by clinicians
to communicate effectively and empathically with patients and their families,
disagreements regarding truly futile treatments remain, expert consultation
for psychosocial support should be sought, with no obligation during such pro-
cesses for clinical teams to provide the requested interventions.
Conversely, for intractable conflicts involving potentially inappropriate
treatment (despite attempts at intensive communication with the patient and/
or family), the recent ATS guidelines recommend a comprehensive approach,
including hospital review, attempts to transfer the patient to another willing
institution and provider, and opportunity for external review of decisions.
Whether all the steps in this approach are attempted or completed depends
on how time-sensitive a given situation is and how certain clinicians are that a
requested treatment is outside of accepted practice.
Conscientious Objections
The ATS has also independently published a separate, specific policy statement
for management of situations where a clinician may have a moral objection to
a providing a medical service which otherwise does fall within accepted med-
General Management
ical practice or regarding which the boundaries of “accepted medical practice”

are unclear.
Definition
A “conscientious objection,” or CO, is a clinician’s moral objection to providing
or disclosing information about a legal, professionally accepted, and otherwise
available medical service. An example of such a situation in neurocritical care
practice may be one where an elderly patient remains comatose two weeks
following a devastating brainstem hemorrhage. A physician may have a con-
Section 3
scientious objection to the family’s request for placement of a tracheostomy

and permanent feeding tube on the grounds that it violates the physician’s
personal moral beliefs about what is respectful when prognosis may be poor,
even though such placement is an accepted treatment in his country.
Recommendations
COs in ICUs should be managed through institutional mechanisms rather than
ad hoc by clinicians at the bedside, with institutions needing to have clear CO
management policies in place (e.g., obtaining a second medical opinion, con-
sulting the institutional ethics committee, etc.). For those situations in which a
clinician invokes a CO to providing what he or she feels is potentially inappro-
priate care to an ICU patient, the first step of the management process is to
274
determine if the requested medical service is consistent with accepted med-

ical practice. According to the ATS guidelines, the act of invoking a CO itself
“should not be considered sufficient justification [alone] to unilaterally forgo
the treatment against the objections of the patient or surrogate. Clinicians
should instead use a fair [institutional] process-based mechanism to resolve
such disputes.” If the institutional process does eventually deem the care to be
appropriate; however, it may be still reasonable for the clinician to request a
personal exemption from providing what he or she perceives as inappropriate
care, even though the institution should ensure that the patient receives the
requested service.
Further Reading
Bosslet GT, Pope TM, Rubenfeld GD, et al. An official ATS/ AACN/ ACCP/
ESICM/ SCCM policy statement: responding to requests for potentially inap-
propriate treatments in intensive care units. Am J Respir Crit Care Med. 2015 Jun
1;191(11):1318–1330.
Cai X, Robinson J, Muehlschlegel S, et al. Patient preferences and surrogate decision
making in neuroscience intensive care units. Neurocrit Care. 2015;23(1):131–141.
George BP, Kelly AG, Schneider EB, Holloway RG. Current practices in feeding tube
placement for US acute ischemic stroke inpatients. Neurology. 2014;83:874–882.
Hemphill JC 3rd, Farrant M, Neill TA Jr. Prospective validation of the ICH Score for 12-
month functional outcome. Neurology. 2009;73:1088–1094.
Holloway RG, Gramling R, Kelly AG. Estimating and communicating prognosis in ad-

vanced neurologic disease. Neurology. 2013;80:764–772.
Hwang DY, Dell CA, Sparks MJ, et al. Clinicians versus formal scales for predicting intra-
cerebral hemorrhage outcomes. Neurology. 2016;86:126–133.
Johnson SK, Bautista CA, Hong SY, Weissfeld L, White DB. An empirical study of
surrogates’ preferred level of control over value-laden life support decisions in inten-
sive care units. Am J Respir Crit Care Med. 2011;183:915–921.
Kon AA, Davidson JE, Morrison W, Danis M, White DB. Shared decision making in
ICUs: an American College of Critical Care Medicine and American Thoracic Society
policy statement. Crit Care Med. 2016;44(1):188–201.
Lewis-Newby M, Wicclair M, Pope T, et al. An official American Thoracic Society policy
Chapter 25
statement: managing conscientious objections in intensive care medicine. Am J Respir
Crit Care Med. 2015;191:219–227.
Rosen DS, Macdonald RL. Subarachnoid hemorrhage grading scales: a systematic re-
view. Neurocrit Care. 2005;2:110–118.
Saposnik G, Kapral MK, Liu Y, et al. IScore: a risk score to predict death early after hos-
pitalization for an acute ischemic stroke. Circulation. 2011;123:739–749.
Steyerberg EW, Mushkudiani N, Perel P, et al. Predicting outcome after traumatic brain
injury: development and international validation of prognostic scores based on ad-
mission characteristics. PLoS Med. 2008;5:e165; discussion e165.
Turgeon AF, Lauzier F, Simard JF, et al. Mortality associated with withdrawal of life-
sustaining therapy for patients with severe traumatic brain injury: a Canadian
275
multicentre cohort study. Can Med Assoc J. 2011;183:1581–1588.
White DB, Curtis JR, Lo B, Luce JM. Decisions to limit life-sustaining treatment for
critically ill patients who lack both decision-making capacity and surrogate decision-
makers. Crit Care Med. 2006;34:2053–2059.
White DB, Engelberg RA, Wenrich MD, Lo B, Curtis JR. The language of prognostica-
tion in intensive care units. Med Decis Making. 2010;30:76–83.
Chapter 26
Brain Death, Organ Donation,

and Transplantation
Hilary H. Wang and David M. Greer
History of Brain Death

Prior to the mid-twentieth century, death was defined by cardiorespiratory
criteria. Patients with severe brain injury inevitably suffered respiratory arrest
due to their inability to keep an open upper airway and provide oxygenation.
Circulatory arrest would soon follow in the apneic, cyanotic patient. However,
with the advent of endotracheal intubation and mechanical ventilation, coma-
tose patients with devastating brain injuries could be supported. This led to a
276
new neurological state with no brain function, and the need for a non-cardiac
definition of death. In the subsequent half-century, defining “brain death” (or
death by neurologic criteria) and the qualifications for its determination have
been the topics of medical, legal, and ethical debate (Figure 26.1).
Performing the Brain Death Exam

Prior to initiating a clinical assessment, the American Academy of Neurology
mandates two prerequisites:
1. Establish the irreversible and proximate cause of coma: a combination of
history, clinical, laboratory, and/or neuroimaging evidence is used to elu-
cidate acute central nervous system (CNS) damage compatible with irre-
versible loss of all brain function.
2. Establish the absence of complicating medical conditions:
a. Normothermia, defined as >96.8°F/36°C, to avoid artificial depres-
sion of brainstem function and/or delay in PaCO2 rise during apnea
testing. Many patients require a warming blanket.
b. Normotension, defined as systolic blood pressure (SBP) >100 mm Hg;
hypotension is common from both loss of peripheral vascular tone
and hypovolemia secondary to diabetes insipidus. Vasopressors and/
or volume replacement may be necessary.
c. Absence of severe electrolyte, acid base, or endocrine disturbances.
Evaluate for acidosis, hyperammonemia, or other markedly deviated
laboratory values. Specific values have not been defined, and it is left to
Brain Death and Organ Donation
Mollaret & Goulon describe
persistently comatose
1959 patients, deem the condition
“le coma depassé”
• Unreceptivity and unresponsivity
The Ad Hoc Committee at • No movements or breathing
Harvard Medical School • No reflexes
defines “irreversible coma • Flat electroencephalogram (EEG)
1968 • All tests must be repeated in 24h
as a new criterion for death”
with no change
• Exclusion of hypothermia (<90°F) or
NIH-sponsored multicenter
central nervous system depressants
1977 U.S. Collaborate Study of
Cerebral Death seeks
Chapter 26
prospective neurologic factors
to predict cardiac arrest Led adoption of the national,
Uniform Determination of
The President’s Commission Death Act (UDDA), which
publishes guidelines to define states “An individual who has
death on the basis of expert sustained either (1) irreversible
1981 medical, legal, religious cessation of circulatory and respiratory
testimony functions, or (2) irreversible cessation
of all functions of the entire brain,
277
The American Association including the brain stem, is dead.
of Neurology (AAN) A determination of death must be
1995 publishes evidence-based made in accordance with accepted
practice parameters medical standards”
The AAN updates brain
death guidelines to separate
2010 evidence- and opinion-based
data
Figure 26.1 A history of brain death determination.
the clinician’s judgment as to whether the abnormality could be suffi-

cient to influence the clinical examination.
d. Absence of drug intoxication, poisoning, sedatives, or neuromus-
cular blocking agents. If available, plasma drug levels should be in the
therapeutic range or below; otherwise, use a combination of history,
drug screen, and clearance calculation using 5 times the drug’s half-life,
assuming normal hepatic and renal function (see Table 26.1). Exercise
caution, as up to 4-fold differences in cerebral effects for a given blood
concentration have been reported in critically ill patients. Also note
that therapeutic hypothermia will delay drug metabolism.
General Management
Table 26.1 Confounding Drugs T1/2

Drug T1/2
Phenobarbital 100 hours
Diazepam 40 hours
Amitriptyline 24 hours
Primidone 20 hours
Lorazepam 15 hours
Thiopental 10 hours
Section 3
Midazolam 6 hours
Fentanyl 6 hours
Morphine 3 hours
Codeine 3 hours
Pancuronium 2 hours
Vecuronium 2 hours
Rocuronium 1 hour
Atracurium ½ hour
Adapted from Wijdicks EFM. Brain Death. 2nd ed. Philadelphia: Lippincott
Williams & Wilkins; 2011: Table 2-2, p. 39.
278
If these prerequisite criteria cannot be established, any further clinical eval-

uation should be delayed until corrected, or ancillary testing should be
performed if uncorrectable.
Conditions that can potentially mimic brain death are listed in Box 26.1.
The Clinical Evaluation

The bedside clinical evaluation of brain death is broken into three parts,
consisting of: (a) coma, (b) absence of brainstem reflexes, and (c) apnea.
Although most states only require one neurologic examination to pronounce
brain death in adults, several states require two.
Coma
Brain dead patients must lack all evidence of responsiveness. Loud auditory
and tactile noxious stimuli, both peripherally and on the cranium (e.g., at
Box 26.1 Mimics of Brain Death

• High cervical cord injury
• Fulminant Guillain-Barré syndrome
• Organophosphate intoxication
• Baclofen toxicity, lidocaine toxicity, and delayed vecuronium clearance
Adapted from Busl KM, Greer DM. Pitfalls in the diagnosis of brain death. Neurocrit Care.
2009;11(2):276–287: Table 1, p. 278.
Box 26.2 Reflexive Movements that May Occur in Brain Death
• Deep tendon reflexes
• Triple flexion
• Babinski sign
• Head turning
• Spontaneous, non-purposeful, stereotyped arm or leg movements
• Respiration-like movements
• Lazarus signs
• Plantar flexion, undulating toe sign
Adapted from Busl KM, Greer DM. Pitalls in the diagnosis of brain death. Neurocrit Care.
2009;11(2):276–287: Table 2, p. 279.
Chapter 26
the supraorbital notch and temporo-mandibular joint), should not produce
any eye opening, eye movements, or any motor response other than spinal
reflexes (see Box 26.2).
Brainstem Reflexes
• No blink to visual threat.
• No pupillary response to bright light. A magnifying glass and/or automated
pupillometer should be used. Pupils should be midposition, 4 to 9 mm.
279
Smaller pupils should alert the examiner to possible drug intoxication.
• Absent oculocephalic reflex, both horizontally and vertically. Only should
be tested in patients with known absence of C-spine injury/instability.
• Absent oculovestibular reflex. Ensure patency and lack of obstruction of
auditory canal. Elevate head to 30 degrees. Irrigate one ear at a time with
ice water for at least 60 seconds, observing for any eye movement. Both
ears must be tested, with an interval between tests of at least five minutes.
Prior exposure to ototoxic drugs or severe facial globe trauma/edema can
confound testing.
• Absent corneal reflex, with optimized testing using a Q-tip and slight manual
pressure. Squirting water/saline is not a sufficient stimulus.
• Absent facial movement to noxious stimulation, including pressure on the
supraorbital notch and temporo-mandibular joint, as well as nasal tickle
with a Q-tip. Spontaneous facial myokymias are permissible but should not
appear in relation to external stimulation.
• Absent pharyngeal (gag) reflex by stimulation of the posterior pharynx.
• Absent cough reflex by deep bronchial stimulation.
Severe facial trauma may prevent full examination of brainstem reflexes and
may necessitate the use of ancillary testing.
Apnea Testing
Prerequisites:
• Normotension (SBP ≥100 mm Hg, with or without vasopressors)
• Normothermia (≥36°C)
General Management
• Euvolemia (particular important in patients with diabetes insipidus)

• Normocapnea (PaCO2 40–45)
• Absence of hypoxia
• An arterial line is required for both hemodynamic monitoring and ease of
arterial blood gas sampling (ABGs)
Procedure:
• Preoxygenate to a PaO2 >200 mm Hg.
Section 3
• Establish normocapnea.
• Reduce positive end expiratory pressure to 5 cm H2O (desaturation with
this maneuver may signal difficulty performing the apnea test).
• Uncover the chest and abdomen to aid in observation for respiratory
movements.
• Disconnect the patient from the ventilator—modern ventilators are sensi-
tive and can autocycle, confounding the test.
• Preserve oxygenation by providing oxygen (flow rate of 4–6 liters/min) to
the level of the carina via a flow catheter through the endotracheal tube.
Higher flow rates may wash out CO2, making it more difficult to attain the
requisite values, or may cause barotrauma. In patients who cannot tolerate
being off of the ventilator, a trial using a T piece with 100% oxygen flow, or
280
the continuous positive airway pressure setting may be used.

• Observe for respiratory movements and hemodynamic stability for at least
10 minutes.
• Abort if the SBP falls to <90 mm Hg, or if the O2 saturation falls to <85%
for >30 seconds.
• Send ABG after 10 minutes (or longer). If no respiratory movements were
observed, and the PaCO2 is ≥60 mm Hg (or ≥20 points higher than the
known baseline value when it is elevated secondary to pre-existing pulmo-
nary disease causing CO2 retention), the patient can be declared dead by
brain criteria.
• If the test is inconclusive (the PaCO2 values are not met), but the patient
was hemodynamically stable during testing, the apnea test can be repeated
for a longer duration, after once again establishing normocapnea and ade-
quate preoxygenation.
• If the patient develops hypoxia and/or hypotension during apnea testing,
temporary hyperventilation can be used to correct both the hypoxia and
the acidosis that is likely causing the hypotension.
Ancillary Testing
Ancillary tests can be divided into those that test electrical activity (e.g., elec-
troencephalography [EEG]) versus cerebral blood flow (e.g., angiography).
In the United States, ancillary testing is not mandatory in adults and is most
often used in patients who are not able to undergo or complete adequate clin-

ical testing, such as those with severe facial trauma or instability during apnea
testing. Clinicians should be cautious in using ancillary tests in lieu of clinical
testing, as all ancillary tests have potential false-positives, false-negatives, and
inter-test discrepancies.
• Catheter cerebral angiography—contrast will fill the external cerebral
circulation, including the meningeal system. Intracranial flow, both for
the internal carotids and vertebral arteries, should arrest at the point of
entry at the dura. Four-vessel angiography under hand injection pres-
sure must be used (i.e., CTA and MRA are not validated, see below).
• Single-photon emission computed tomography (SPECT)—99mTc-HMPAO
is injected 15 to 30 minutes before scanning. There should be an absence of
intracranial uptake of tracer, representing lack of cerebral metabolism and
Chapter 26
absent blood flow. The persistent extracranial circulation allows uptake of
tracer in the scalp, meninges, and face/nose. Lateral views must be used to
evaluate for brainstem uptake.
• EEG—specific criteria have been established, including sensitivity of at
least 2 microvolts. There should be a complete absence of cerebral elec-
trical activity, although electrocardiography (ECG) artifact may be seen.
EEG is often confounded by electromagnetic interference in the intensive
281
care unit (ICU) setting, making the test indeterminate. Furthermore, EEG
does not evaluate brainstem activity, and thus the authors recommend
against its use.
• Transcranial Doppler—bilateral anterior and posterior examinations must
be performed twice, at least 30 minutes apart. Sharp systolic spikes and os-
cillating flow must be observed, with zero or negative values during diastole.
Absence of flow signal where it was once detected is not felt to be reliable
and could be due to technical limitations. TCD has not been validated in
patients with extracranial drains or craniotomies.
• CT angiography (CTA) and magnetic resonance angiogram (MRA) have not
been validated prospectively, and false positives have been reported. They
are not recommended for use. Evoked potentials have also fallen out of
favor due to relatively poor predictive value.
Documentation
The time of death is the time the arterial PaCO2 that reached the target value
is officially reported, or when the ancillary test, if used, is reported by the
attending physician interpreting the test. Documentation should include the
cause of the neurological state, prerequisites established, details of clinical
and apnea testing (including ABG values), and results of ancillary testing, if
used. Organ procurement organizations should be contacted for all cases of
potential brain death testing, as mandated in federal and state law (Box 26.3).
General Management
Box 26.3 Religious Objections to Brain Death

Note that despite acceptance of the UDDA, in New York and California, statutes require
that physicians/hospitals accommodate religious and cultural practices and concerns. New
Jersey law further states that in the event of religious objections, death shall be declared
“solely on the basis of cardiorespiratory criteria.”
Adapted from Ehrle R. Timely referral of potential organ donors. Crit Care Nurse. 2006;26.2:88–93.
Section 3
Organ Donation and Brain Death

As of August 2018, 114,419 patients in the United States are in need of a
lifesaving organ transplant; of those, 74,646 are on an active waiting list. Yet
from January to July of 2018, only 21,042 transplants from 10,120 donors had
been performed, resulting in an average of 20 patients per day who die while
waiting for a transplant.
Role of Organ Procurement

Organizations and Notification Timing
282
Organ procurement organizations (OPOs) are a group of 58 federally regu-

lated, independent, private, non-profit organizations with specific geographic
territories. OPOs provide 24- hour, immediate response to organ donor
referrals, organ donor case management, public and professional education,
and donor family aftercare. Timing of OPO notification of a potential donor
is not federally mandated, but a list of potential clinical triggers for initiating
contact is presented in Box 26.4.
Although OPOs are responsible for determining final medical suitability for
organ donation, this decision is often made in conjunction with local transplant
surgeons. In general, there are very few absolute contraindications for organ
donation, and all suitability decisions take into account comorbidities and
the age of the potential donor. For reference, Box 26.5 lists some common
contraindications to donation.
Box 26.4 Triggers for OPO Notification

• Unresponsive patient receiving mechanical ventilation and clinical findings consistent with
Glasgow Coma Scale score <5
• As soon as a formal “brain death” examination is contemplated
• Before initiating a discussion that may lead to withdrawal of life-sustaining therapy
Box 26.5 Potential Contraindications to Donation
Infections
• Tuberculosis
• Gangrenous bowel/perforated bowel
• Multisystem organ failure due to sepsis
• HIV infection
• Rabies
• Reactive Hepatitis B
• Active HSV, VZA, CMV viremia/pneumonia
• West Nile virus
• Active infection with Cryptococcus, Aspergillus, Histoplasma, Coccidioides
• Active infection with T. Crizi, Leishmania, strongyloides, or malaria
• Creutzfeldt-Jakob disease
Chapter 26
• Viral encephalitis or meningitis
Hematologic conditions and malignancies

• Aplastic anemia
• Agranulocytosis
• Current malignant neoplasms except nonmelanoma skin cancers and nonmetastatic pri-
mary central nervous system tumors
• Previous malignant neoplasms with current metastatic disease
283
• Prior melanoma
• Hematologic malignancies (e.g., leukemia, lymphoma)
Adapted from Wijdicks EFM. Brain Death, 2nd ed. Philadelphia: Lippincott Williams & Wilkins;
2011: Table 5-4, p. 108.
Physiologic Changes in the Brain Dead

Patient
After an acute intracranial insult, a rapid rise in intracranial pressure (ICP)
causes mass effect, which leads to cerebral ischemia and venous engorgement.
This forces the brainstem through the foramen magnum and causes arterial
compression, leading to further brain swelling and greater elevations in ICP
until all cerebral circulation ceases. The resulting ischemia causes a number of
physiologic derangements, including
• Almost invariable hypothermia, due to hypothalamic damage, reduced
metabolism, vasodilation, and heat loss
• Hypotension (81%–97%) due to hypovolemia, reduced coronary blood flow,
pituitary insufficiency, and myocardial dysfunction
• Diabetes insipidus (46%–78%) due to damage to the posterior pituitary
• Disseminated intravascular coagulation (DIC) (29%–55%) due to coagulopathy
and tissue factor release
• Arrhythmias (25%–32%) due to a sympathetic surge and myocardial damage
General Management
• Pulmonary edema (13%–18%) due to acute blood volume diversion and

capillary damage
ICU Management of Brain Dead Donors

Due to these complex physiologic derangements, brain dead donors pre-
sent a unique challenge for ICU management in the pretransplant window.
Section 3
Consensus recommendations, including those from the Canadian Forum on

Medical Management to Optimize Donor Organ Potential (summarized in
Figure 26.2), focus on a multisystem approach to optimize organ physiology
for multiorgan donors.
Pediatric Brain Death Determination

The specifics of brain death testing from a clinical standpoint are similar for
children compared to adults, with specific requirements based on the age of
the patient and what are the appropriate hemodynamic parameters, as well as
284
Figure 26.2 ICU management of brain dead donors.

the timing and number of clinical evaluations to be performed and the need

for ancillary testing. A full discussion is beyond the scope of this chapter, and
the reader is referred to the 2011 Pediatric Brain Death Guidelines.
Further Reading
American Electroencephalographic Society. J Clin Neurophysiol. 1994;11:10–13.
Boucher BA, Wood GC, Swanson JM. Pharmacokinetic changes in critical illness. Crit
Care Clin. 2006;22:255–271.
Busl KM, Greer DM. Pitfalls in the diagnosis of brain death. Neurocrit Care.
2009;11.2:276–287.
Cantu RC. Brain death as determined by cerebral arteriography. Lancet.
1973;301:7816:1391–1392.
Chapter 26
Ducrocq X, Hassler W, Moritake K, et al. Consensus opinion on diagnosis of cerebral
circulatory arrest using Doppler-sonography: Task Force Group on cerebral death
of the Neurosonology Research Group of the World Federation of Neurology. J
Neurolog Sci. 1998;159(2):145–150.
Ehrle R. Timely referral of potential organ donors. Crit Care Nurse. 2006;26.2:88–93.
Facco E, Zucchetta P, Munari M, et al. 99mTc-HMPAO SPECT in the diagnosis of brain
death. Intensive care medicine 1998;24.9:911–917.
285
Greer DM, Strozyk D, Schwamm LH. False positive CT angiography in brain death.
Neurocrit Care. 2009;11(2):272–275.
Health Services and Services Administration, U.S. Department of Health and Human
Services. Organ Procurement and Transplantation Network. Available at http://
optn.transplant.hrsa.gov. Accessed October 20, 2015.
Luce JM. The uncommon case of Jahi McMath. Chest. 2015;147:1144.
McKeown DW, Bonser RS, Kellum JA. Management of the heartbeating brain-dead
organ donor. Br J Anaesthes. 2011;108:i96–i107.
Nakagawa TA, Ashwal S, Mathur M, Mysore M. Clinical report—guidelines for the de-
termination of brain death in infants and children: an update of the 1987 task force
recommendations. Pediatrics 2011;128(3):e720–e740.
Shemie SD, Ross H, Pagliarello J, et al. Organ donor management in Canada:
recommendations of the forum on Medical Management to Optimize Donor Organ
Potential. Can Med Assoc J. 2006;174(6):S13–S30.
Wijdicks EFM. Brain Death, 2nd ed. Philadelphia: Lippincott Williams & Wilkins, 2011.
Wijdicks EFM, Varelas PN, Gronseth GS, Greer DM. Evidence-based guideline up-
date: determining brain death in adults. Report of the Quality Standards Subcommittee
of the American Academy of Neurology. Neurology 2010;74.23:1911–1918.
Chapter 27
Rehabilitation
in Neurocritical Care
Michael E. Reznik and Amy K. Wagner
Introduction
Rehabilitation and recovery in critically ill neurologic patients can be a pro-
longed, often lifelong process, and its foundations must begin in the intensive
care unit (ICU). As mortality rates fall in patients with severe neurologic injury,
the intensivist will be expected to assume more responsibility in the initiation
and implementation of this process, many of the components of which are
outlined in this chapter. This overview is intended as an introduction to reha-
286
bilitation issues in multiple forms of brain injury; however, it must be noted

that most of the clinical evidence currently available in the rehabilitation litera-
ture relates to traumatic brain injury (TBI), followed by stroke, and then other
forms of acquired brain injury. Though these recommendations are certainly
not interchangeable, the concepts introduced in one patient population may
potentially be applicable to others.
Prognosis for Recovery

Unlike in degenerative and other progressive diseases, robust evidence to aid
with prognostication is not available for most forms of acute brain injury; even
for the most severe of injuries, a prognosis as to which patients will have a
meaningful recovery often cannot be made with absolute certainty. Because
advances in critical care have allowed the possibility of prolonging life even
when a patient is neurologically devastated, when patients do die, it is often
because of withdrawal of care after extensive goals of care discussions. As
such, outcomes are often driven by self-fulfilling prophecy, much of which is
determined by clinician framing and bias. Therefore, it is important to gauge
the desires of patients and their families for potentially prolonged rehabilitation
and supportive care on a case-by-case basis, while preparing for the possibility
that there may not be any meaningful functional improvement. Rehabilitation
physicians can and should be incorporated into these discussions given their
experience with outcomes beyond the ICU. Clinicians should give patients
and family members reasonable expectations, while also conveying that there
Rehabilitation in Neurocritical Care

may be a good deal of variability with respect to recovery from one patient
to another.
Demographic characteristics play a major role in the potential for recovery
and overall outcomes: older age, female sex, poor previous functional status,
and medical comorbidities are all associated with worse outcomes in various
forms of brain injury. Mechanism and location of injury also influence prog-
nosis. Extensive injuries such as diffuse axonal injury (DAI), as seen in TBI, are
more likely to result in worse functional outcomes and disorders of conscious-
ness (DOC) than focal injuries; in the latter, functional neuroanatomy can be
used to predict deficits, though these can be mitigated to a varying degree
through rehabilitation.
Indeed, DOC presents a particular prognostic challenge. The continuum
from coma to vegetative and minimally conscious states may be separated
Chapter 27
only by subtle clinical features indicating patient arousal and awareness, but
the distinctions between these states may have an impact on a patient’s ca-
pacity for recovery. Still, according to longitudinal studies, a significant pro-
portion of patients discharged with DOC may not remain that way, though
relatively few will return to functional independence. Most data exists in the
severe TBI population, however, which by virtue of being overwhelmingly
young and male likely represents a more optimistic prognosis than in most
287
other forms of acquired brain injury. In one study, Nakase-Richardson et al.
examined the outcomes of patients discharged to rehabilitation after severe
TBI with a persistent DOC and found that 68% regained consciousness during
inpatient rehabilitation (almost all of whom were discharged to a community
setting). Of those who did not, 59% regained consciousness at one year, 66%
at two years, and 74% at five years; meanwhile, nearly 20% had regained func-
tional independence at five years postinjury.
In the most severe injuries, ancillary studies such as brain imaging and
neurophysiologic testing (like electroencephalography [EEG] and electrophys-
iologic studies) are often used to aid with prognostication, but their use in a
protocolized fashion is controversial due to a relative lack of evidence. Still,
there are some situations where they can be helpful. For example, nonreac-
tive patterns on EEG are negative prognostic indicators, while the presence of
sleep architecture is considered positive. Meanwhile, the bilateral absence of
cortical responses on somatosensory evoked potentials (SSEPs) almost uni-
formly portends a poor prognosis; positive SSEPs, however, are not specific
to a good functional outcome.
Early Rehabilitation
Multiple guidelines have recommended early mobilization after brain injury,
especially stroke, ostensibly to prevent immobility-
related complications.
There is also hypothesized to be a narrow window of peak brain plasticity
that may be facilitated by early rehabilitation. In the stroke population, mul-
General Management
tiple studies by Wang et al. have shown that earlier admission to an inpa-
tient rehabilitation facility leads to improved functional outcomes. However,
the recent AVERT trial concluded that ultra-early high-dose mobilization after
stroke (i.e., at less than 24 hours) may actually be detrimental, suggesting that
guidelines for mobilizing patients with acquired brain injury may benefit from
further refinement.
Given the time-sensitive nature of recovery, and the medical complexity
associated with patients with brain injury, enlisting the aid of physical medicine
Section 3
and rehabilitation (PM&R) consultants to assist in management of the various

issues outlined here can be advantageous. A study by Wagner et al. suggested
that early PM&R involvement may facilitate shorter lengths of stay and favor-
able outcomes, as they can help coordinate a rehabilitation and disposition
plan and medically optimize patients to make for a more seamless postacute
care transition.
Multidisciplinary Care
Patient participation is central to the rehabilitation process, and the overall
goal is a return to functional independence. This goal is not only important in
288
attaining a good quality of life but also in preventing the accumulation of com-
plicating conditions that may limit recovery. To that end, primary targets often
include improving motor strength, mobility, coordination, balance, stamina,
communication skills, and overall cognition. The rehabilitation plan can be
initiated to some degree in an acute care setting and requires the collabo-
ration of a multidisciplinary team consisting of PM&R, physical and occupa-
tional therapists, speech and language pathologists, neuropsychologists, social
workers, and nurses, among others.
Early on in the acute care setting, the focus is most often on assessing and
improving task readiness, which varies depending on the underlying condition
and associated impairment. Symptoms like spasticity, dysautonomia, cognitive
impairment, sleep disturbance, and agitation may affect range of motion, tol-
erance, and endurance while sitting or standing, interactions with the environ-
ment, and engaging in functional activities like swallowing and communicating.
After the transition to postacute rehabilitation, limitations due to secondary
conditions and complications are more comprehensively addressed; mean-
while, remediation with complex tasks and skill-building may enable progress
with motor programming via repetition and stepwise progression through the
functional components of these tasks.
Because of the degree of expertise required, as well as the familiarity
with disease- specific practices, dedicated neurorehabilitation centers
geared toward patients with brain and spinal cord injury may offer ad-
ditional benefits over general rehabilitation services. Likewise, having a
contemporary neurorehabilitation program as part of a comprehensive
stroke center or level 1 trauma center may help maximize care quality

and early recovery.
Management of Complications in the

Acute Setting
Disruptions in neurotransmitter pathways occur frequently after acquired brain
injury, with the specific pathways affected depending on the mechanism of injury,
its location, and whether it is focal or diffuse. As a result, clinical consequences
are diverse and range from impaired arousal and motivation to mood
disturbances and agitation, sleep/wake cycle dysfunction, and dysautonomia,
all of which hinder overall functional recovery. Polypharmacy may further ag-
gravate these issues, and though complex drug regimens are often unavoid-
Chapter 27
able in critically ill patients, care must be taken to streamline medications when
possible. Patients with brain injury, especially the elderly, are more sensitive to
the effects of anticholinergic, antihistamine, and antidopaminergic medications,
as well as benzodiazepines and other GABA agonists. Many antiepileptics are
also an underrecognized cause of sedation, and phenytoin in particular has also
been shown to potentially hinder functional recovery.
289
Arousal and Stimulation
Early animal models showed enhanced recovery of function from various
forms of experimental brain injury with the administration of D-amphetamine
and other neurostimulants. These benefits have been attributed to enhanced
brain plasticity, an increase in neural sprouting, synaptogenesis, long-term po-
tentiation, and the improved participation that comes with increased arousal.
Neurostimulant medications of potential utility in the ICU are summarized in
Table 27.1 with their mechanisms of action and side effects.
Clinical studies have been mixed with regard to treatment effects, however,
largely due to inconsistent design, and there have thus far been only two large
prospective trials that have investigated the use of such medications:
• In a randomized controlled trial (RCT) in patients with ischemic stroke,
Chollet et al. showed that treatment with fluoxetine early in the course
of rehabilitation significantly improved motor recovery.
• It is unclear how much of this improvement was due to treatment of
concomitant mood disturbances.
• The sample sizes were also somewhat small (59 patients in each arm).
• In a RCT in patients with severe TBI, Giacino et al. showed that treatment
with amantadine in the subacute phase postinjury significantly hastened re-
covery in patients in vegetative or minimally conscious states.
• However, the clinical improvement was modest given the relatively
short time period examined and the severity of initial injury.
• The benefits also appeared to dissipate once amantadine was discontinued.
General Management
Table 27.1 Neurostimulant Medications with their Mechanisms

of Action and Side Effects
Drug Mechanism of Action Side Effects
D-amphetamine Directly releases DA and Anxiety, insomnia, GI upset,
NE and inhibits reuptake headache, tachycardia, HTN,
euphoria, mania
Methylphenidate DA-NE reuptake inhibitor Anxiety, insomnia, GI upset,
headache, tachycardia, HTN,
euphoria, mania
Section 3
Modafinil ? inhibit reuptake of DA Headache, nausea, anxiety

transporter, ↑ histamine in
hypothalamus, inhibit GABA
release in basal ganglia
Amantadine Multiple: dopaminergic, Orthostasis, seizures,
noradrenergic, serotonergic; hallucinations, anxiety, agitation
NMDA receptor antagonist
Bromocriptine DA (especially D2) agonist Orthostasis, nausea, headache,
hallucinations
Levodopa Catecholamine precursor Orthostasis, nausea,
(DA, NE, epi) hallucinations, dyskinesias
DA = dopamine; NE = norepinephrine; GI = gastrointestinal; HTN = hypertension.
290
Nevertheless, despite a lack of robust evidence, some practitioners have

extrapolated these results to other patient populations. Though such use
remains controversial until more studies are done, the following represents
some suggestions for use:
• The side effect profile of amphetamines likely prohibits their use in the
inpatient setting, especially in the ICU. However, methylphenidate has
many of the same effects with relatively milder side effects, so its use
can be considered in patients for whom a strong dopaminergic and nor-
adrenergic effect is desired (i.e., in severe DOC)—especially younger
patients who can better tolerate the side effects and for whom the risk
of dysrhythmia is low.
• For patients with damage to their reticular activating system (brainstem,
thalamus) and/or bilateral hemispheres, methylphenidate and modafinil
may be most effective, though bromocriptine can also be considered.
• In patients with lobar injuries—whether focal, as in stroke, or diffuse, as in
DAI—a primarily dopaminergic agent such as amantadine or bromocriptine
may be sufficient (via its effects on the mesocortical pathway).
• A dopaminergic agents, including levodopa itself, should be considered
when there is involvement of the basal ganglia, or when clinical signs of
parkinsonism are evident.
• Bromocriptine can be a preferred agent in patients with both paroxysmal
sympathetic hyperactivity (see later discussion) and hypoarousal states.
• Note that for all of these medications, a neuromodulatory effect can be

seen with doses much smaller than those used for other clinical indications.
• Doses should be given early in the day (preferably in the morning) so as
not to cause insomnia at night; morning stimulation may also facilitate sleep
cycle regulation.
Sleep Disturbances
Disturbances in sleep may be a result of a patient’s brain injury but may also be
exacerbated by the abundance of external stimuli in a hospital setting, especially
the ICU. Multiple studies suggest that critical illness may be associated with a
decrease in circulating plasma melatonin levels, thereby disrupting natural circa-
dian rhythms. Possible manifestations include fatigue, insomnia, impaired sleep
quality, or a combination of these issues. Evaluation and management should
begin with an assessment of possible causes, followed by nonpharmacologic
interventions that attempt to minimize any additional external disruptions.
Chapter 27
(Given the heterogeneity of patients in the neurointensive care unit, including
ventilated patients with severe injuries and those who are intact but admitted
for neuromonitoring, the applicability of these strategies may vary.)
• Existing medications should be assessed for their sedating or stimulating
potential and for possible interactions.
• For those medications that are sedating, their dosing should be retimed
291
so that they are given before bedtime rather than during the day.
• Stimulating medications such as those given to improve hypoarousal
states should be administered early in the day, preferably early in the
morning.
• Sleep hygiene should be maintained as circumstances allow, including
• Maximizing natural light during the day and maintaining darkness at night.
• Reducing stimulation before bedtime (e.g., turning televisions off ).
• Reducing naps during the day as tolerated.
• If possible, minimize overnight examinations, measurements of vital signs,
lab draws, medication administration, and noise.
• Ventilator adjustments may be considered for ventilated patients (e.g.,
“resting” overnight on assisted modes).
• Aim for earlier transfer out of the ICU when possible.
Often, though, such strategies are insufficient or impossible, making the ini-
tiation of a pharmacologic treatment an attractive next step. Unfortunately,
there are limited studies examining sleep-promoting medications in critically
ill patients (let alone those with acquired brain injuries), rendering formal
guidelines a challenge. However, some practical pearls have emerged from
clinical practice:
• Melatonin (3–10 mg at bedtime): used in the general population to treat
insomnia and sleep phase disorders, generally not associated with any side
effects and therefore likely low risk.
• A small trial by Bourne et al. showed an increase in sleep duration and
General Management
efficiency in ventilated patients who received melatonin.

• A study by Hatta et al. suggested that ramelteon, a melatonin receptor
agonist, may help prevent delirium in elderly patients admitted to ICU
or ward settings.
• Trazodone (25–100 mg at bedtime): serotonin antagonist and reuptake
inhibitor antidepressant frequently used in hospitalized patients for its se-
dating effects; no evidence available delineating risks or efficacy in ICU
patients. In healthy patients, increases amount of total sleep and percentage
Section 3
of deep sleep.
• Antipsychotics: given their negative impact on motor and cognitive
recovery, should not be used for nocturnal sedation unless there is con-
current agitation. In this case quetiapine (25–100 mg at bedtime) may be
favored given its mild antidopaminergic and strong antihistaminic effect.
• Benzodiazepines and nonbenzodiazepine GABA agonists: prob-
ably best avoided, as both the former and the latter (a class which includes
zolpidem and zaleplon) may negatively impact cognitive recovery, while also
potentially exacerbating delirium.
Agitation
Agitation is common in the acute phase after acquired brain injury, particularly
292
in older patients, those with underlying cognitive impairment, and after TBI.
However, it can also occur after subarachnoid hemorrhage, stroke, and other
forms of brain injury and critical illness. It arises in the context of the injury
itself and as a product of the hospital environment, especially the ICU. The
resulting disorientation may make it difficult for patients to communicate their
needs to caregivers and may also make patients more sensitive to environ-
mental stimuli. When agitation arises, a disciplined approach should be utilized
as outlined in Figure 27.1: first investigating possible sources, then attempting
nonpharmacologic interventions prior to considering pharmacotherapy.
Judicious use of sedating medications is recommended because many of
those historically used to control agitation may actually interfere with overall
recovery and even prolong the period of agitation itself. For example, Feeney
et al. and Wilson et al. showed that, after experimental brain injury, haloper-
idol had detrimental effects on the recovery of motor and cognitive function,
respectively. Though newer atypical antipsychotics have a more favorable
side effect profile, a study by Kline et al. showed that they too have the po-
tential to hinder cognitive recovery. Of these, quetiapine has the lowest
antidopaminergic activity (instead acting as a potent antihistamine) and is
therefore perhaps most worthy of consideration. Beta-blockers, which have
even fewer side effects, may also be useful—in fact, a Cochrane review by
Fleminger et al. showed that they have the best evidence for efficacy with
minimal side effects. In contrast, benzodiazepines, which are often used for
sedation in the ICU, should be used with caution in agitated patients as they
may exacerbate states of delirium and worsen cognitive outcomes when given
• Pain • Address possible • Propranolol
Identify possible sources
Pharmacologic treatment
Nonpharmacologic interventions
• Infection sources • Consider low doses
• New intracranial • Remove noxious of quetiapine or
process stimuli olanzapine; avoid
• Sleep/wake • Minimize lines and typical antipsychotics
disturbance restraints if possible if possible
• Adverse medication • If restraints are needed, • If continuous infusion
effect/interaction a net bed may be required, consider
• Urinary retention preferable to other dexmedetomidine
or constipation forms of restraints, • Avoid benzodiazepines
• Medication or with mitts to reduce if possible (unless in
alcohol withdrawal or prevent inadvertent withdrawal state)
line removal • Consider amantadine
• Abdominal binder to if concurrent
protect PEG tube hypoarousal state
Chapter 27
Figure 27.1 Evaluation and management of agitation after acquired brain injury.
over prolonged periods of time. Finally, the American College of Critical Care
Medicine’s 2013 clinical practice guidelines suggest that dexmedetomidine can
293
also be considered if a continuous infusion is desired.
Despite treatment, agitation may persist until there has been substantial
cognitive improvement. It can undermine functional gains and overall partic-
ipation in therapy while becoming a source of frustration for providers and
family alike and may also endanger a patient’s safety or that of others.
Dysautonomia
Dysautonomia has been described after multiple forms of acquired brain
injury, though most typically after TBI. Other associated etiologies include
anoxic brain injury and stroke (especially hemorrhagic), while more recent
cases have also described an association with some forms of autoimmune
encephalitis. Evidence is limited to case reports and series, and while there is
no definite consensus on diagnostic criteria, dysautonomia is conceptually de-
fined by episodic hyperadrenergic states which consist of hypertension, tachy-
cardia, tachypnea, hyperthermia, diaphoresis, and tonic posturing. As such,
it is perhaps more accurately termed paroxysmal sympathetic hyperactivity
(PSH). The exact mechanism for this phenomenon is unclear, but speculation
implicates a disruption of the sympathetic and autoregulatory centers in the
brain, including the brainstem and diencephalon. Because it is also commonly
described as being temporally related to stimulation, afferent oversensitivity
has also been considered as a possible mechanism.
While the incidence has not been well established, dysautonomia seems to
be associated with more severe injuries and worse outcomes and is itself a
potential cause of further morbidity. Sequelae may include hypermetabolism
leading to unintended weight loss, cardiomyopathy from prolonged tachy-
General Management
cardia and hypertension, and rhabdomyolysis from either fever or prolonged

posturing. Treatment should therefore be initiated quickly, after other eti-
ologies like infection, increased intracranial pressure, and serotonin syn-
drome or other medication effects are ruled out. Initial interventions include
repositioning the patient and removing stimuli when possible; pain control is
also paramount.
Though evidence is limited for specific pharmacotherapy, interim pharmaco-
logical management strategies are often tailored toward symptoms as follows:
Section 3
• Beta-blockers for hypertension and tachycardia, preferably a centrally

acting one like propranolol if there is concurrent agitation
• Centrally acting alpha-agonists like clonidine, also for hypertension, tachy-
cardia, and agitation
• Opiates for pain control, with long- acting forms for maintenance and
immediate-acting forms for breakthrough pain
• Gabapentin may potentially mitigate afferent oversensitivity and reduce the
need for opiates
• Benzodiazepines can be helpful for posturing movements along with other
symptoms, though they may cause oversedation
• Bromocriptine can be a preferred agent if there is a concurrent
294
hypoarousal state
Note also that these medications may be effective at lower doses for PSH
than for their primary indication—one should consider starting at a low dose
first before gradually titrating upwards.
Bowel and Bladder Dysfunction
Because of a push toward minimizing catheters to prevent catheter-associated
infections, issues with bowel and bladder dysfunction often arise as part of pa-
tient care in the ICU. Various injuries along the neuraxis can result in bladder
and bowel disinhibition, a loss of perception of fullness, and poor sphincter
control. It is generally not recommended that either urinary or bowel incon-
tinence be treated pharmacologically (i.e., with anticholinergic or antimotility
agents, respectively) in an ICU setting due to possible untoward side effects
that may cloud an otherwise complex clinical picture. Instead, focus should be
placed on increasing awareness and frequent cleaning in order to prevent skin
breakdown and an increased risk of systemic infection.
Urinary retention, on the other hand, may benefit from a trial of an alpha-
blocker like tamsulosin, which can relax muscles in the bladder neck but may
require intermittent catheterization. Constipation can be treated medically as
well, with a combination of stool softeners and laxatives, but maintaining ade-
quate hydration and increasing activity out of bed are also key. Since both con-
stipation and urinary retention can be exacerbated by multiple medications
(especially opiates and sedatives), the use of such medications should be
limited to minimum required amounts and accompanied by a bowel reg-

imen. Complications associated with urinary retention include urinary tract
infections, delirium, agitation, and sympathetic hyperactivity; meanwhile, se-
vere constipation may lead to bowel obstruction and respiratory compromise
from pressure on the diaphragm.
Spasticity
Spasticity is common after central nervous system injury, typically taking
days to weeks to develop, and is defined as a velocity-dependent increase
in muscle tone. It can limit mobility and be a source of pain for patients.
Treatment can often be difficult, so there should be an early focus on pre-
ventive strategies while in the ICU and hospital setting, including splinting
and passive range of motion exercises (i.e., stretching). Often, though,
pharmacotherapy will be required, which includes both systemic and local-
ized treatments (the latter of which includes intrathecal pumps and local
Chapter 27
injections). The medications that are likely most familiar to intensivists in-
clude baclofen (an inhibitor at pre-and postsynaptic GABAB receptors),
tizanidine (an alpha-2 agonist), and benzodiazepines. However, all of these
are centrally acting and may cause oversedation and cognitive side effects,
especially in the setting of acquired brain injury. For this reason many reha-
bilitation specialists prefer using dantrolene, a postsynaptic muscle relaxant
295
that acts via calcium inhibition, as it tends to cause less sedation. Note,
however, that dantrolene has occasionally been associated with hepatotox-
icity, so its use necessitates frequent (typically weekly) monitoring of liver
function tests.
Further Reading
AVERT Trial Collaboration Group. Efficacy and safety of very early mobiliza-
tion within 24 h of stroke onset (AVERT): a randomized controlled trial. Lancet.
2015;386(9988):46–55. doi:10.1016/S0140-6736(15)60690-0.
Barr J, Fraser G L, Puntillo K, et al. Clinical practice guidelines for the management of
pain, agitation, and delirium in adult patients in the intensive care unit. Crit Care Med.
2013;41(1):263–306.
Bourne RS, Mills GH, Minelli C. Melatonin therapy to improve nocturnal sleep in crit-
ically ill patients: encouraging results from a small randomised controlled trial. Crit
Care. 2008;12(2):R52.
Chollet F, Tardy J, Albucher JF, et al. Fluoxetine for motor recovery after acute is-
chemic stroke (FLAME): a randomized placebo-controlled trial. Lancet Neurol. 2011
Feb;10(2):123–130.
Duncan PW, Zorowitz R, Bates B, et al. Management of adult stroke rehabilitation
care: a clinical practice guideline. Stroke. 2005 Sept;36(9):e100–e143.
Feeney DM, Gonzalez A, Law WA. Amphetamine, haloperidol, and experience interact
to affect rate of recovery after motor cortex injury. Science. 1982;217:855–857.
Fleminger S, Greenwood RJ, Oliver DL. Pharmacological management for agitation and
General Management
aggression in people with acquired brain injury. Cochrane Database Syst Rev. 2006
Oct:18(4):CD003299.
Ganguly K, Byl NN, Abrams GM. Neurorehabilitation: motor recovery after stroke as
an example. Ann Neurol. 2013;74:373–381.
Giacino JT, Whyte J, Bagiella E, et al. Placebo-controlled trial of amantadine for severe
traumatic brain injury. N Engl J Med. 2012 Mar 1;366(9):819–826
Hatta K, Kishi Y, Wada K, et al. Preventive effects of ramelteon on delirium: a random-
ized placebo-controlled trial. JAMA Psychiatry. 2014 Apr:71(4)397–403.
Section 3
Kline AE, Hoffman AN, Cheng JP, et al. Chronic administration of antipsychotics im-
pede behavioral recovery after experimental traumatic brain injury. Neurosci Lett.
2008;448:263–267.
Langhorne P, Duncan P. Does the organization of postacute stroke care really matter?
Stroke. 2001 Jan;32(1)268–274.
Nakase-Richardson R, Whyte J, Giacino JT, et al. Longitudinal outcome of patients with
disordered consciousness in the NIDRR TBI Model Systems programs. J Neurotrauma.
2012;29(1):59–65.
Perkes I, Baguley IJ, Nott MT, Menon DK. A review of paroxysmal sympathetic hyperac-
tivity after acquired brain injury. Ann Neurol. 2010;68:126–135.
Wagner AK, Fabio T, Zafonte RD, Goldberg G, Marion DW, Peitzman AB. Physical
medicine and rehabilitation consultation: relationships with acute functional out-
come, length of stay, and discharge planning after traumatic brain injury. Am J Phys
296
Med Rehabil. 2003;82:526–536.

Wang H, Camicia M, Terdiman J, et al. Time to inpatient rehabilitation hospital admis-
sion and functional outcomes of stroke patients. PM R. 2011;3:296–304.
Wang H, Camicia M, DiVita M, et al. Early inpatient rehabilitation admission and stroke
patient outcomes. Am J Phys Med Rehabil. 2015 Feb;94(2):85–100.
Wilson MS, Gibson CJ, Hamm RJ. Haloperidol, but not olanzapine, impairs cognitive per-
formance after traumatic brain injury in rats. Am J Phys Med Rehabil. 2003;82:871–879.
Index
Tables, figures, and boxes are indicated by an italic t, f, and b following the page number.
A amyotrophic lateral apnea testing, brain death,

sclerosis, 137 279–80
abscess. See also specific types
anemia arbovirus meningitis, 106t–
CNS, 179
acute ischemic stroke, 107t, 108–109
intracranial, 117
avoiding, 31 arousal
abulia, 7
subarachnoid coma, 3
abuse head trauma, pediatric,
hemorrhage, 52 on rehabilitation,
241–242
traumatic brain injury, 92 289–91, 290t
acute bacterial meningitis,
anoxic brain injury, 58–66 arteriovenous malformations,
105–108, 106t
cardiac catheterization, pediatric, 243
acute coronary syndrome,
emergent, 59 ASPECT score, 176
neurological patient, 216
critical care management, Aspergillus fumigatus meningitis,
acute disseminated
59–64, 62f 107t, 109–111
encephalomyelitis,
dysautonomia, 293–4 aspirin, pregnancy, 249
pediatric, 244
etiology, 58–59, 60t ataxia, 141
acute ischemic stroke. See
hemodynamic atrial fibrillation, neurological
stroke, acute ischemic
management, 64 patient, 216–217
acute kidney injury,
incidence and autoimmune orobuccolingual
neurological patient,
297
epidemiology, 58 dyskinesia, 147–148
218–221
initial patient evaluation, autonomic dysreflexia, 102
acute lung injury, subarachnoid
58–59, 60t, 61f
hemorrhage, 51
neuroimaging, 178
acute respiratory distress
syndrome, subarachnoid
neurological B
prognostication,
hemorrhage, 51 barbiturates
64–66, 66t
adrenal insufficiency, coma from, 90, 262–3
outcomes, 58, 58t
neurological patient, 222 for fulminant hepatic failure,
post-arrest seizures, 64
advanced care planning, 262–3
severity, 59, 61f
269–72, 271t for intracranial pressure, 90
targeted temperature
agitation, 292–3, 293f pediatrics, 234t
management, 61–63, 63t
airway, breathing, circulation, on pupils, 158t
ventilator management,
coma, 9–11 for sedation, 18
63–64, 63t
airway management anterior cord syndrome, 99 B-cell mediated autoimmune
coma, 9–11 anticoagulants. See also CNS diseases, 124t.
neurological patient, specific types See also inflammation
213–214 direct oral, intracerebral and demyelination,
postoperative, hemorrhage from, autoimmune
neurosurgery, 39–40 benzodiazepines. See also
226–228, 227b pregnancy, 249–250 specific types
akinesia, 141 anticoagulants, reversal, 217, for encephalopathy and
akinetic mutism, 7 219t–220t delirium, 19–20
Alberta Stroke Program Early target-specific oral for sleep disturbances, 292
CT Score (ASPECT), anticoagulants, 202, 204t, for status epilepticus,
29, 30f 217, 219t 72, 75t
alcohol withdrawal, 79 warfarin, 202, 203t, best gaze, acute ischemic
alteplase, for intracerebral 217, 219t stroke, 24t
hemorrhage, 41 antipsychotics, for sleep bladder dysfunction, 294–5
amantadine, 290–291, 291t disturbances, 292 blood pressure management
D-amphetamine, antishivering measures and acute, antihypertensives for,
290–291, 291t drugs, 191–192, 191t 202–203, 206t–208t
blood pressure management pediatric care, cerebral metabolism
Index
(Cont.) comatose, 234t monitoring, 162, 163t

intracerebral sudden, 58–66 cerebral microdialysis, 162, 163t
hemorrhage, 40 temperature control, 188 cerebral oxygenation
postoperative, cardiac catheterization, monitoring, 158–160
neurosurgery, 230–231 emergent, anoxic brain brain tissue, 159, 160t
botulism, 135–136 injury, 59 invasive, 158–159
pediatric, 245 cardiovascular care, jugular bulb oximetry,
bowel dysfunction, 294–5 neurological patient, 158–159, 259
bradykinesia, 141 215–217 near-infrared spectroscopy,
brain death, 276–85 carotid duplex ultrasound, 159–160
ancillary testing, 280–1 acute ischemic non-invasive, 159–160
apnea testing, 279–80 stroke, 178 cerebral perfusion pressure
brainstem reflexes, 279 catatonia, 7 (CPP), 85, 85f, 258
clinical evaluation, 278–80 catheter-based cerebral definition, 155
coma, 278–9 angiography, 180 intracranial pressure, 85
conditions mimicking, 278b brain death, 281 traumatic brain injury, 88
documentation, 281, 282b catheter-based cervical cerebral perfusion pressure
drugs, confounding, angiography, 180 (CPP) monitoring,
277, 278t central cord syndrome, 99 155–158
exam, performing, 276–8 central fever, 186 invasive, 155–156, 155t
historical perspectives, cerebellar degeneration, noninvasive, 155t, 157–158
276, 277f 120t. See also parenchymal
ICU management, 284, 284f inflammation and (microtransducers), 156
organ donation, 282 demyelination, traumatic brain injury, 87
organ donation, autoimmune cerebral salt wasting,
contraindications, cerebral angiography subarachnoid
298
282, 283b catheter-based, 180 hemorrhage, 51

organ procurement catheter-based, brain cerebral sinus thrombosis,
organizations and death, 281 pregnancy, 247–249
notification timing, pregnancy, 248 cerebral vasospasm, delayed
282, 282b cerebral autoregulation, 84, cerebral ischemia, 48–49
pediatric, determination, 84f, 85f cerebral venous sinus
284–5 traumatic brain injury, thrombosis, pregnancy,
physiologic changes, 85, 85f 249–251
283–4 cerebral blood flow, 84, 84f, cerebrospinal fluid (CSF)
pregnant mother, 255 160, 161 drainage, historical
brainstem reflexes, brain cerebral blood flow overview, 165
death, 279 monitoring, 160–162 Monro-Kellie doctrine,
brain tissue oxygenation fulminant hepatic failure, 165–166
(PbtO2) 258–9 ventriculostomy and
cerebral, 159, 160t invasive, 161 external ventricular
monitoring, 159, 160t Laser-Doppler drainage, 165–171
monitoring, flowmetry, 161 (see also external
multimodal, 155t non-invasive, 161–162 ventricular
sedation interruption on, 19 thermal diffusion probe, drainage (EVD))
traumatic brain 161–162 cervical angiography,
injury, 87, 91 cerebral edema catheter-based, 180
bromocriptine, 290, 290t fulminant hepatic failure, cervical spine precautions,
Brown-Sequard syndrome, 99 257, 260, 260b coma, 10–11
bundles of care, 217, 218t intracerebral chorea, 146–148
hemorrhage, 38 clinical nihilism, 265–6
cerebral ischemia. See also clobazam, for status
C anoxic brain injury;
specific disorders
epilepticus, 78t
Clostridium botulinum,
cardiac arrest. See also anoxic delayed, subarachnoid 135–136
brain injury hemorrhage, 48–49 coagulopathy, neurological
pediatric care, 240–241 pathophysiology, 84, 84f patient, 217–218
Coccidioides immitis meningitis, constipation, 294–5 pregnancy, thrombotic
Index
107t, 109–111 constriction velocity, 158t events, 247
coma, 3–14 cooling methods, 192–193 spinal cord injury, traumatic,
arousal, 3 craniectomy. See also 101–102
brain death, 278–9 hemicraniectomy, deep venous thrombosis
content, 3 decompressive (DVT) prophylaxis
definitions, 3–4 decompressive, malignant acute ischemic stroke, 31
etiology, structural, 4–6, 5t cerebral edema, 35 malignant edema, 34
etiology, toxic/metabolic, craniotomy, for intracerebral traumatic brain
5t–6t, 6–7 hemorrhage, 40–41 injury, 94–95
evaluation, initial, 9–12, creatine kinase, deferoxamine mesylate,
10f, 11t neuroleptic malignant for intracerebral
Full Outline of syndrome, 142 hemorrhage, 41
Unresponsiveness critical illness myopathy, delayed cerebral ischemia,
(FOUR), 8, 9t 137–138, 138f subarachnoid
fundamentals, 3 critical illness hemorrhage, 48–49
Glasgow Coma Scale, 8, 9t neuromyopathy, 137 delirium, 15–21
mimics, 7–8 critical illness benzodiazepines for, 19–20
physiology of polyneuropathy, 137 definition, 15
consciousness, 3–4 Cryptococcus neoformans dexmedetomidine
prognosis and treatments, meningitis, 107t, for, 19, 20
13–14, 13t 109–111 morbidity and
computed tomography (CT) CT angiography (CTA) mortality, 15, 21
coma, 12 acute ischemic stroke, opioids for, 20–21
fulminant hepatic multiphase, 177 prevalence and
failure, 259 acute ischemic stroke, relevance, 15
computed tomography pregnancy, 248 prevention, 17
299
head (CTH) brain death, 281 propofol for, 20
acute ischemic stroke, 176 intracerebral risk factors and causes, 16
acute ischemic stroke, dual hemorrhage, 174 screening, 16–17
energy, 178 subarachnoid sedation and, 17–19, 18f
anoxic brain injury, 178 hemorrhage, 175 treatment, 17
encephalitis, 179 traumatic brain injury, 173 demyelination, 119–128.
encephalopathy, 180 CT perfusion (CTP), acute See also inflammation and
epidural hemorrhage, 176 ischemic stroke, 177 demyelination
intracerebral Cushing’s reflex, 230 dexmedetomidine, for
hemorrhage, 174 encephalopathy and
meningitis, 179 delirium, 19, 20
pediatric, 239, 240–241
pregnancy, 248
D diabetes insipidus, neurological
patient, 221
subarachnoid decision-making diazepam, for status
hemorrhage, 175 shared, 269, 270f epilepticus, 74t
subdural hemorrhage, 176 surrogate, 272–3 diffusion weighted
traumatic brain injury, 173 decompressive craniectomy, imaging (DWI)
Confusion Assessment malignant cerebral anoxic brain injury, 178
Method for the ICU edema, 35 encephalitis, 179
(CAM-IU), 16–17 decompressive encephalopathy, 180
congestive heart failure, hemicraniectomy traumatic brain injury, 174
neurological intracerebral digital subtraction
patient, 216 hemorrhage, 41 angiography (DSA)
conscientious objections, pregnancy, 249 acute ischemic stroke,
273–4 traumatic brain injury, 91 177–178
consciousness deep brain stimulation, on intracerebral
definition, 3 coma, 14 hemorrhage, 174
disorders, recovery deep venous traumatic brain injury, 173
prognosis, 287 thrombosis (DVT) direct oral anticoagulants,
level, 24t postoperative, intracerebral hemorrhage
physiology, 3–4 neurosurgery, 229 from, 39–40
direct thrombin inhibitors, Rickettsia rickettsia, 115t, faciobrachial dystonic seizures,
Index
reversal, 204t, 217, 219t 116–117 120t. See also inflammation

direct thrombolysis, viral, 115t, 113–116 and demyelination,
pregnancy, 250 encephalomyelitis, acute autoimmune
disorders of consciousness, disseminated, factor Xa inhibitors
recovery prognosis, 287 pediatric, 244 intracerebral
dopaminergic drugs, encephalopathy, 15–21. hemorrhage, 39–40
formulations and See also delirium reversal, 204t, 217, 219t
administration, 144t definition, 15 false localizing sign, 87
drug intoxication, 277, 278t hepatic, 257–8, 257t feeding tube placement,
drug reactions, acute limbic, 120t ischemic stroke,
dystonic, 151 neuroimaging, 180 265, 266f
dual energy computed Wernicke’s, 180 fentanyl, for encephalopathy
tomography head, acute endocrine care, neurological and delirium, 20–21
ischemic stroke, 178 patient, 222 fever, 186
dysarthria, acute ischemic endovascular therapy acute brain injury, 185–186
stroke, 27t acute ischemic stroke, 29 central, 186
dysautonomia, 293–4 pregnancy, 250 control, 185–193 (see
dystonia, 150–151 enterobacterial intracranial also temperature
dystonic storm, 151 abscess, 117 control, body)
enterovirus ICU, 186
encephalitis, 115t on neurologic
E meningitis, 106t, 108–109
pediatric CNS
outcomes, 186
postoperative,
eclampsia, 252–254 infection, 244 neurosurgery, 228–229
edema epidural hemorrhage, subarachnoid
malignant, 33–35 neuroimaging, 176 hemorrhage, 50
300
vasogenic, 179 ethics, 269–74 traumatic brain

electrical activity monitoring, advanced care planning, injury, 92–93
162–163 269–72, 271t fluid balance, traumatic brain
electrocorticography, 162–163 conscientious objections, injury, 93
electroencephalography 273–4 fluid therapy,
(EEG), 162–163 potentially inappropriate postoperative, 231
brain death, 281 treatment requests, 273 fluoxetine, 289
coma, 12 shared decision-making, fontanelle, bulging, pediatric,
fulminant hepatic 269, 270f 237–238
failure, 259 surrogate decision-making, fosphenytoin, for status
inflammation and 272–3 epilepticus, 75t
demyelination, external ventricular drainage Full Outline of
autoimmune, 121, 122t (EVD), 165–171 Unresponsiveness
pediatric, 240 complications, 169–171 (FOUR), 8, 9t
seizures, 69 historical overview, 165 fulminant hepatic failure,
status epilepticus indications, 165–166 256–9
monitoring, 80–81 management strategy, cerebral blood flow
electrolytes 168–169 monitoring, 258–9
abnormalities, neurological monitoring, 155–156 clinical manifestations,
patient, 221 pediatric, 241 257–8
traumatic brain injury, 93 surgical technique, definition, 256
empyema, subdural, 117 166–168, 167f diagnostic testing, 258–9
encephalitis, 115t, 113–117 weaning, etiology, 256
definition, 113 ventriculostomy, 169 pathophysiology, 256–7
immune-mediated, 117 extinction and inattention, fulminant hepatic failure,
Listeria monocytogenes, 115t, acute ischemic stroke, 27t management, 259–63
116–117 barbiturate coma, 262–3
Mycoplasma pneumoniae, cerebral edema, 260, 260b
115t, 116–117
Naegleria fowleri, 115t,
F general measures, 259–60,
261f
116–117 facial palsy, acute ischemic hyperosmolar therapy, 262
neuroimaging, 179–180 stroke, 25t hyperventilation, 261
hypothermia, 262 hemodynamic agents, 202– fulminant hepatic
Index
prognosis, 263 203, 206t–209t failure, 262
propofol sedation, 261 vasopressors, for hypertonic saline, 198–199
transplantation, 263 hemodynamic support, mannitol, 195–198
functional magnetic resonance 202–203, 208t–209t hyperpyrexia
imaging (fMRI), anoxic hemodynamic support, control, 185–193
brain injury, 178 vasopressors for, 202– (see also temperature
fungal meningitis, 107t, 203, 208t–209t control, body)
109–111 hemorrhage. See also traumatic brain injury
“futile” interventions, 273 specific types management, 92–93
epidural, neuroimaging, 176 hypertension
intracerebral, 37–42 acute, antihypertensives for,
G subarachnoid, 44–52
subdural, neuroimaging, 176
202–203, 206t–208t
intracranial, fulminant
GABA agonists. See also hemostasis resuscitation, 218 hepatic failure, 256–63
specific types hemostatic agents, neurological patient, 215
for sleep disturbances, 292 anticoagulant reversal, postoperative,
gaze, acute ischemic 202, 203t–204t, 217, neurosurgery, 231
stroke, 24t 219t–220t hyperthermia. See
Glasgow Coma Scale hepatic encephalopathy, fever; temperature
(GCS), 8, 9t 257–8, 257t control, body
pediatric neurocritical care, herpes simplex virus 1 (HSV- hypertonic saline, intracranial
237–238 1) encephalitis, 115t, pressure management,
traumatic brain injury, 113–116 198–199
86–8 7, 86t herpes simplex virus 2 fulminant hepatic
traumatic brain injury (HSV-2) failure, 262
severity, 83 encephalitis, 115t, 113–116 hyperventilation
301
glycemic control, neurological meningitis, 106t, 108–109 spontaneous, systemic
patient, 222 Histoplasma capsulatum inflammatory response
granulomatous disorders, meningitis, 107t, 109–111 syndrome, 50
124t. See also HIV meningitis, 107t, 108–109 for traumatic brain
inflammation and hydrocephalus injury, 88
demyelination, coma, 5t hyperventilation, for
autoimmune fontanelle, enlarged, intracranial pressure
Guillain-Barré syndrome, 131, 237–238 fulminant hepatic failure,
136–137 intraventricular 260t, 261, 261f
pediatric, 245 hemorrhage, 38, 174 infarction and malignant
meningitis, 110, 112, edema, 34–35
113, 179 neurological patient, 215
H obstructive, cerebellar
infarcts, 35
pregnancy, 250
hypokinetic movement
Haemophilus influenza Parkinsonism, acute, 145t disorders, 141–146
meningitis, 105–108, 106t stroke, pediatric, 235t, 243 acute Parkinsonism,
healthcare proxy of subarachnoid 144–145, 145t
attorney, 271t hemorrhage, 46–47 neuroleptic malignant
heat conservation, 189–191, ventriculitis, pediatric, 244 syndrome, 141, 141b
191t, 192t ventriculostomy for Parkinson drug
heat generation, 189–192, (see external ventricular management, ICU,
191t, 192t drainage (EVD)) 142–144, 144t
hematologic care, neurological hyperkinetic movement Parkinsonism-
patient, 217–218 disorders, 146–151. hyperpyrexia syndrome,
hemiballismus-hemichorea, See also movement 141–142
146–147 disorders, hyperkinetic; psychosis in Parkinson
hemicraniectomy, specific types disease, 145–146
decompressive hypernatremia, neurological hyponatremia
intracerebral patient, 221 neurological patient, 221
hemorrhage, 41 hyperosmolar therapy, subarachnoid
pregnancy, 249 intracranial pressure hemorrhage, 51
traumatic brain injury, 91 management, 195–198 hypotension, 215–216
hypothermia natural history and clinical intubation
Index
control, 185–193 course, 37–38 neurological patient,

(see also temperature neuroimaging, 174–175 213–214
control, body) outcome prediction tools, postoperative,
mild, 63, 63t, 189, 190t 267, 268t neurosurgery, 93,
(see also temperature prevalence, 37 226–227, 227b
control) “spot sign,” 39, 39f, 174 ischemia. See also anoxic brain
physiology and intracerebral hemorrhage, injury; specific disorders
pathophysiology, clinical trials, 38–41 cerebral, delayed, 48–49
185–187 anticoagulant-associated cerebral, pathophysiology,
therapeutic (see ICH, 39–40 84–85, 84f, 85f
therapeutic blood pressure myocardial, 215–216
hypothermia (TH)) management, 40 stroke, 23–35, 176–178,
traumatic brain injury, 189 hemostasis, 38–39, 39f 247–249 (see also stroke,
neuroprotection, 41 acute ischemic)
platelet transfusion, 40
I surgical management, 40–41
intracranial abscess, 117 J
ICU-acquired weakness, intracranial hypertension
137–138, 138f management, fulminant JC virus, 126
immune-mediated hepatic failure, 256–63. jugular bulb (venous) oximetry,
encephalitis, 117 See also fulminant hepatic 158–159
infarction. See also specific types failure fulminant hepatic
cerebellar, obstructive intracranial pressure (ICP) failure, 259
hydrocephalus, 35 brain death, 283
hyperventilation for, 34–35 cerebral perfusion pressure
large territorial, 33–35 and, 85 K
302
myocardial, 216 definition, 155 Kernohan notch

infection. See also specific types forces, 91, 92f phenomenon, 87
neuroimaging, 179–180 malignant edema, 34–35 ketamine, for status
pediatric care, 243–244 postoperative, epilepticus, 78t
inflammation neurosurgery, 93 Kocher’s point, 167, 167f
beneficial effects, 186–187 ventriculostomy for,
pediatric care, 243–244 165–171 (see also
inflammation and
demyelination,
external ventricular
drainage (EVD))
L
autoimmune, 119–128 intracranial pressure (ICP) lacosamide, for status
classification, 119, 120t management epilepticus, 77t
clinical and laboratory hypertonic saline, 198–199 language, acute ischemic
studies, 120, 121t mannitol, 195–198 stroke, 27t
CNS biopsy, 122–123 pregnancy, 250 large territorial
CSF studies, 121, 123t traumatic brain infarction, 33–35
diagnostic strategy, 120– injury, 88–91 laryngoscopy, neurological
123, 121t–123t intracranial pressure (ICP) patient, 214
EEG studies, 121 monitoring, 155–158 Laser-Doppler flowmetry, 161
imaging and diagnostic external ventricular drain, last known well (LKW)
studies, 120–121, 122t 155–156 time, 23
treatment, options, 123– fulminant hepatic level of consciousness, 24t
126, 124t–125t failure, 258 levetiracetam, for status
treatment, risks, 126–128, optic nerve sheath diameter, epilepticus, 75t
126t, 127t 157–158 levodopa, 290, 290t
inotropes, 209t parenchymal life-sustaining therapies,
Intensive Care Delirium (microtransducers), 156 variable outcomes, 265
Screening Checklist pediatric, 240 limb ataxia, acute ischemic
(ICDSC), 16–17 pupillometry, stroke, 26t
intracerebral 157–158, 158t limbic encephalopathy, 120t.
hemorrhage, 37–42 traumatic brain injury, 87 See also inflammation and
future directions, 41–42 ultrasound, 158 demyelination, autoimmune
Listeria monocytogenes neuroimaging, 179 monoamine oxidase B (MAO-
Index
encephalitis, 115t, 116–117 pediatric care, 243–244 B) inhibitors, drug
meningitis, 105–108, 106t postoperative, interactions, 143–144
liver transplantation, fulminant neurosurgery, 228 Monro-Kellie doctrine, 85, 91,
hepatic failure, 263 viral, 106t–107t, 108–109 165–166
living will, 271t methylphenidate, 290, 290t motor examination, traumatic
locked-in syndrome, 7 microtransducers, 156 brain injury, 87
lorazepam, for status midazolam motor function, acute
epilepticus, 74t for encephalopathy and ischemic stroke
low arousal states, 3–14. delirium, 19 arm, 25t
See also coma for status epilepticus, leg, 26t
definitions, 3–4 74t, 76t motor neuron disease, 137
fundamentals, 3 middle cerebral artery (MCA) movement disorders, 141
physiology of acute ischemic stroke, 176 movement disorders,
consciousness, 3–4 ultrasound monitoring, 158 hyperkinetic, 146–151
prognosis and treatments, Miller-Fisher syndrome, 120t. autoimmune
13–14, 13t See also inflammation orobuccolingual
lumbar drains, 166 and demyelination, dyskinesia, 147–148
autoimmune chorea, 146–148
minimally conscious state, 4 drug reactions, acute
M modafinil, 290, 290t dystonic, 151
monitoring, multimodality, dystonia, 150–151
magnetic resonance 155–163. See also dystonic storm (status
angiography (MRA), specific types dystonicus), 151
acute ischemic cerebral blood flow, hemiballismus-hemichorea,
stroke, 177 160–162 146–147
magnetic resonance cerebral metabolism, myoclonus, 148–150,
imaging (MRI)
303
162, 163t 149b, 149f
anoxic brain injury, 178 cerebral microdialysis, other conditions, 150
encephalitis, 179–180 162, 163t posthypoxic myoclonus, 150
infections, brain, 179 cerebral oxygenation, serotonin syndrome, 148–
intracerebral hemorrhage, 158–160 150, 149b, 149f
174–175, 175t cerebral perfusion pressure, movement disorders,
meningitis, 179 155–158 hypokinetic, 141–146
pregnancy, 248 definitions, 155 acute Parkinsonism,
traumatic brain injury, 174 EEG, 162–163 144–145, 145t
malignant edema electrical activity, 162–163 neuroleptic malignant
decompressive electrocorticography, syndrome, 141, 141b
craniectomy, 35 162–163 Parkinson drug
ICU care, 33–35 external ventricular drain, management, ICU,
mannitol, intracranial pressure 155–156 142–144, 144t
management, 195–198 intracranial pressure, Parkinsonism-hyperpyrexia
fulminant hepatic 155–158 syndrome, 141–142
failure, 262 jugular bulb oximetry, psychosis in Parkinson
mean arterial pressure (MAP), 158–159, 259 disease, 145–146
155, 258 Laser-Doppler multimodality monitoring,
mechanical ventilation, flowmetry, 161 155–163. See also
neurological patient, mean arterial pressure, 155 monitoring,
214–215 near-infrared spectroscopy, multimodality;
melatonin, 291–2 159–160 specific types
meningismus, 243 parenchymal myasthenia gravis, 132–135,
meningitis, 105–113 (microtransducers), 156 133f, 133t–134t
bacterial, 105–108, 106t pediatric care, 239–240 pediatric, 245
evaluation and treatment, pupillometry, myasthenic crisis,
105, 106t 157–158, 158t 132–135, 134t
fungal, 107t, 109–111 thermal diffusion probe, Mycoplasma pneumoniae
Lyme, 107t, 111 161–162 encephalitis, 115t,
mycobacteria, 107t, 111 ultrasound, 158 116–117
myocardial infarction, 216 postanoxic injury, 178 neuromuscular respiratory
Index
myocardial ischemia, 215–216 traumatic brain injury, failure, 138–139, 139b

myoclonic status epilepticus, 173–174 neuromyotonia, 120t.
69–71, 79, 150 neuroinflammation. See also See also inflammation
myoclonus, 120t, 148–150. inflammation and demyelination,
See also inflammation beneficial effects, 186–187 autoimmune
and demyelination, neurointerventional neuropharmacotherapy, 195–
autoimmune procedures, 180–184 209. See also specific agents
definition, 148 agents, 181t–183t anticoagulant reversal,
posthypoxic, 150 complications, 184 hemostatic agents,
serotonin syndrome, 148– indications, 180 202, 203t–204t, 217,
150, 149b, 149f intraprocedural 219t–220t
myoclonus status considerations, 184 hemodynamic agents,
epilepticus, 69–71 postprocedural care, 184 202–203, 206t–209t
myopathies, 131–132 preprocedural hyperosmolar
preparation, 184 pharmacotherapy,
treatments, 180–184, intracranial pressure
N 181t–183t management, 195–198
neuroleptic malignant pharmacodynamics, 195
Naegleria fowleri encephalitis, syndrome, 141, 141b pharmacokinetics, critically
115t, 116–117 neurological patient, general ill and elderly, 195,
natriuresis, subarachnoid ICU care, 213–223 196t–197t
hemorrhage, 51 cardiovascular, 215–217 shiver control, therapeutic
nausea and vomiting, electrolytes, 221 temperature
postoperative, 224, 225t endocrine, 222 management, 202,
near-infrared spectroscopy, hematologic, 217–218 204t–206t
159–160 nutrition, 222–223 status epilepticus and
neglect, 27t
304
pulmonary, 213–215 refractory status

Neisseria meningitides renal, 218–221 epilepticus, 199–202,
meningitis, 105–108, 106t sepsis, 217, 218t 199t–202t
neonatal asphyxia, stress ulcer neuroprotection
temperature control, 188 prophylaxis, 223 hypothermia, 186
nerve agents, 136 venous thromboembolism hypothermia, anoxic brain
neuroangiography, prophylaxis, 222 injury, 61
catheter-based Neurological Pupil Index hypothermia, pediatric, 233
indications, 180 (NPi), 157, 158t intracerebral
treatments, 180–184, neuromuscular conditions, hemorrhage, 41, 42
181t–183t 131–139 intracranial pressure,
neurogenic pulmonary botulism, 135–136 propofol and
edema, 51 Guillain-Barré syndrome, dexmedetomidine, 89
neurogenic stress 131, 136–137 spinal cord injury, traumatic,
cardiomyopathy, ICU-acquired weakness, 102–103
subarachnoid 137–138, 138f subarachnoid
hemorrhage, 50–51 motor neuron disease hemorrhage, 49
neurogenic stunned (ALS), 137 neurostimulants
myocardium, myasthenia gravis, 132–135, mechanisms of action and
subarachnoid 133f, 133t–134t side effects, 290t
hemorrhage, 50–51 myopathies, 131–132 on recovery, 289–91, 290t
neuroimaging, 173–180. neuromuscular respiratory neurosurgical patient,
See also specific types and failure, 138–139, 139b postoperative
disorders organophosphate management, 224–231
acute ischemic stroke, toxicity, 136 airway, 226–228, 227b
176–178 neuromuscular disorders. blood pressure and fluid
coma, 12 See also specific types therapy, 230–231
encephalopathy, 180 pediatric, 245 fever, 228–229
hemorrhagic stroke, neuromuscular junction pain control, 229
174–176, 175t transmission postoperative nausea and
infection, 179–180 abnormalities, vomiting, 224, 225t
pediatric care, 238–239 medication, 132, 133t sodium balance, 230
nihilism, clinical, 265–6 P phenytoin, for status
Index
NIH Stroke Scale score, 24t–27t epilepticus, 75t
normotension, 276 pain control, physician orders for
normothermia, 276 postoperative, 229 life-sustaining
nutrition, neurological patient, parenchymal monitors, ICP treatment, 271t
222–223 and CPP, 156 Pittsburgh Cardiac Arrest
paresis, 141 Category scale, 59, 61f
Parkinsonism, acute, platelet transfusion,
O 144–145, 145t
Parkinsonism-hyperpyrexia
for intracerebral
hemorrhage, 40
opioids, for encephalopathy syndrome, 141–142 positron emission
and delirium, 20–21 Parkinson’s disease, 141 tomography (PET)
opsoclonus, 120t. See also drug management, ICU, anoxic brain injury, 178
inflammation and 142–144, 144t autoimmune CNS disease,
demyelination, psychosis, 145–146 120, 122t
autoimmune paroxysmal sympathetic posterior cord syndrome, 99
optic nerve sheath diameter, hyperactivity, 293 posthypoxic myoclonus, 150
157, 158 pediatric neurocritical care, postoperative nausea and
optic neuritis plus, 120t. 233–245 vomiting, 224, 225t
See also inflammation arteriovenous potentially inappropriate
and demyelination, malformations, 243 treatment requests, 273
autoimmune cardiac arrest, 240–241 preeclampsia, 252–254
organ donation cardiac arrest, pregnancy, 247–255
brain death and, 282 comatose, 234t brain death, maternal, 255
contraindications, 282, 283b CNS infection and cerebral venous sinus
organ procurement inflammation, 243–244 thrombosis, 249–251
organizations and epidemiology and ischemic stroke and cerebral
notification timing,
305
outcomes, 233 sinus thrombosis,
282, 282b general approach, 247–249
organophosphate 237, 238f physiologic changes, 247
toxicity, 136 imaging, 238–239 preeclampsia and eclampsia,
organ procurement management guidelines, 252–254
organizations, 282, 282b 233–237, 234t–236t reversible cerebral
osmolality, calculated, 197 monitoring, 239–240 vasoconstriction
osmotic gap, 197 neuromuscular syndrome, 251–252
outcome prediction tools, disorders, 245 seizure, epileptics, 254
267, 268t physical exam, 237–238 subarachnoid hemorrhage,
oximetry, jugular bulb, sinus venous 254–255
158–159, 259 thrombosis, 243 prognostication, 265–9
oxygenation, 215 spinal cord injury, 242 challenges, 265–7
jugular venous oximetry, status epilepticus, 242–243 communication, optimal,
158–159, 259 stroke, 243 267–9, 270f
mechanical ventilation for, therapeutic fundamental
214, 215 interventions, 239 uncertainty, 267
mechanical ventilation for, traumatic brain injury, 234t, importance, 265
pediatric, 239 240, 241–242 outcome prediction tools,
oxygenation, brain tissue. See pentobarbital, for status 267, 268t
brain tissue oxygenation epilepticus, 76t overconfidence, 266
(PbtO2) pharmacodynamics, 195 overestimation, 267
oxygenation monitoring, pharmacokinetics recovery, rehabilitation,
cerebral, 158–160 critically ill and elderly, 195, 286–7
brain tissue oxygen, 196t–197t propofol
159, 160t definition, 195 for encephalopathy and
invasive, 158–159 pharmacotherapy, delirium, 20
jugular bulb oximetry, 195–209. See also for status epilepticus, 76t
158–159, 259 neuropharmacotherapy; proxy of attorney,
near-infrared spectroscopy, specific agents healthcare, 271t
159–160 phenobarbital, for status psychogenic
non-invasive, 159–160 epilepticus, 77t unresponsiveness, 7–8
psychosis, Parkinson disease, clinical presentation, 54 therapeutic hypothermia,
Index
145–146 definition and antishivering medications,

pulmonary care, neurological pathophysiology, 54 202, 205t–206t
patient, 213–215 diagnosis, 55, 55f shock, 215
pulmonary edema, pregnancy, 251–252 single-photon computed
neurogenic, 51 recurrence risk, 56 tomography (SPECT)
pulmonary embolism (PE) signs and brain death, 281
hypotension, 230 symptoms, 54–55 status epilepticus, 81
neuromuscular respiratory treatment, 55–56 sinus venous thrombosis,
failure, 138 rhabdomyolysis, 132 pediatric, 243
pregnancy, 247 Rickettsia rickettsia encephalitis, sleep disturbances, 291–2
spinal cord injury, 101 115t, 116–117 sodium balance,
traumatic brain injury, postoperative,
prophylaxis, 94–95 neurosurgery, 230
pulsatility index, 157, 158
pupillary examination,
S spasticity, 295
spinal cord injury,
traumatic brain injury, 87 saline, hypertonic, intracranial traumatic, 97–103
pupillometry, 157–158, 158t pressure management, cervical spine
198–199 clearance, 100
fulminant hepatic clinical assessment and
R failure, 262
sedation
classification, 97–99,
98f
rapid sequence intubation, for agitation, 292–3, 293f diagnosis, 97–100
neurological patient, delirium and, 17–19, 18f epidemiology, 97
213–214 for fulminant hepatic pathophysiology, 97
recombinant factor failure, 261 pediatric, 242
VIIa (rFVIIa), for seizure. See also status radiographic
306
intracerebral epilepticus assessment, 99–100

hemorrhage, 38–39, 39f alcohol withdrawal, 79 treatment, 100–102
refractory status epilepticus anoxic brain injury, 64 Staphylococcus aureus
treatment, 199, 201t, 202 causes and initial workup, intracranial abscess, 117
rehabilitation, 286–95 69, 70b meningitis, 105–108, 106t
agitation, 292–3, 293f diagnosis and status dystonicus, 151
arousal and stimulation, classification, 68–69 status epilepticus (SE),
289–91, 290t EEG, 69 12, 68–81
bowel and bladder faciobrachial classification, 69–71
dysfunction, 294–5 dystonic, 120t complications,
clinical evidence, 286 pregnancy, epileptics, 254 immediate, 71
dysautonomia, 293–4 prevalence, 68 definition, 69, 242
early, 287–8 subarachnoid EEG monitoring,
multidisciplinary care, hemorrhage, 49–50 ICU, 80–81
288–9 seizure prophylaxis generalized convulsive, 69
neurotransmitter pathway subarachnoid myoclonic, 69–71, 79, 150
complications, 289 hemorrhage, 49–50 non-convulsive, 69
polypharmacy traumatic brain injury, 94 pediatric, 242–243
complications, 289 “self-fulfilling prophesies,” prevalence, 68
recovery prognosis, 286–7 265–6 refractory, 71
sleep disturbances, 291–2 sensory loss, acute ischemic special considerations, 79
spasticity, 295 stroke, 26t super-refractory, 71
renal care, 218–221 sepsis care, neurological status epilepticus (SE)
renal replacement therapy, patient, 217, 218t management, 71–72,
neurological patient, 221 serotonin syndrome, 148–150, 73f, 74t–78t, 199–202
repetitive nerve stimulation, 149b, 149f emergent control, 199t
132, 133f shared decision-making, outcomes, 79–80
respiratory muscle weakness, 269, 270f urgent control,
131–132 shivering, 189–192, anticonvulsants,
reversible cerebral 191t, 192t 200t–201t
vasoconstriction bedside assessment, stent placement,
syndrome, 54–56 202, 204t pregnancy, 249
stiff-person syndrome, 120t neuroimaging, systemic inflammatory
Index
stimulants, on rehabilitation, 174–176, 175t response, subarachnoid
289–91, 290t pediatric care, 236t, 243 hemorrhage, 50
Streptococcus intracranial stroke, ischemic
abscess, 117 neuroimaging, 176–178
Streptococcus pneumoniae
meningitis,
pregnancy, 247–249
Stroke Scale score, NIH,
T
105–108, 106t 24t–27t Takotsubo
stress cardiomyopathy stunned myocardium, cardiomyopathy, 216
neurogenic, subarachnoid neurogenic, 50–51 targeted temperature
hemorrhage, 50–51 subarachnoid management, anoxic
neurological patient, 216 hemorrhage, 44–52 brain injury, 61–63, 63t
stress ulceration, GI agitation, 292–3, 293f T-cell mediated autoimmune
prophylaxis, neurological anemia and transfusion, 52 CNS diseases, 124t.
patient, 223 clinical presentation and See also inflammation
trauma, 102 diagnosis, 44–45 and demyelination,
stroke, acute ischemic, 23–35 early brain injury autoimmune
after IA and cerebral temperature, body. See also
thrombectomy, 32–33 vasospasm, 47–49 fever; hypothermia;
after IV-tPA, 31–32, 32t epidemiology, 44 therapeutic
Alberta Stroke Program fever and systemic hypothermia (TH)
Early CT Score inflammatory on acute brain injury, 185
(ASPECT), 29, 30f response, 50 brain, fever on, 185–186
consciousness level, 24t grading scales, 45, 46t temperature control, body,
dysarthria, 27t hyponatremia and 185–193
extinction and intravascular volume, 51 acute ischemic stroke,
inattention, 27t management, acute, 46–47 188–189
307
facial palsy, 25t neurogenic pulmonary cardiac arrest, 188
feeding tube placement, edema, 51 clinical evidence and
265, 266f neurogenic stress indications, 187–189
gaze, best, 24t cardiomyopathy, 50–51 cooling methods, 192–193
general care, 30–31 neuroimaging, 175 heat conservation and
language, best, 27t outcome prediction tools, generation, 189–192,
large territorial infarction 267, 268t 191t, 192t
and malignant pregnancy, 254–255 laboratory measurements,
edema, 33–35 risk factors, 44 192, 192t
last known well time, 23 seizures and seizure measures and drugs,
limb ataxia, 26t prevention, 49–50 191–192, 191t
management, acute, systemic complications neonatal asphyxia, 188
23, 28–30 and critical care physiology and
motor function, management, 49–52 pathophysiology,
arm, 25t subdural empyema, 117 185–187
motor function, subdural hemorrhage, traumatic brain injury, 189
leg, 26t neuroimaging, 176 therapeutic hypothermia
neuroimaging, 176–178 substitutive judgment, 272 (TH), 187
NIH Stroke Scale score, sudden cardiac arrest, acute ischemic stroke,
24t–27t 58–66. See also anoxic 188–189
outcome prediction tools, brain injury antishivering medications,
267, 268t surrogate decision-making, 202, 205t–206t
pediatric care, 272–3 bedside shivering
235t–236t, 243 Surviving Sepsis Campaign, assessment, 202, 204t
recovery, prognosis, 286–7 217, 218t, 222 cardiac arrest, 188
rehabilitation, early, 287–8 syndrome of inappropriate effects, 189, 190t
sensory loss, 26t secretion of antidiuretic fulminant hepatic
temperature control, hormones (SIADH) failure, 262
188–189 sodium balance, 230 laboratory measurements,
visual fields, 25t subarachnoid 192, 192t
stroke, hemorrhagic hemorrhage, 51 measures and drugs,
dysautonomia, 293–4 traumatic brain injury, 93 191–192, 191t
therapeutic hypothermia (TH) pediatric care, 234t, 240, encephalitis, 115t,
Index
(Cont.) 241–242 113–116

neonatal asphyxia, 188 primary injury, 83 meningitis, 107t, 108–109
other indications, 189 recovery, prognosis, 286–7 vasodilators, 206t–207t
traumatic brain injury, 189 rehabilitation, early, 287–8 vasogenic edema,
thermal diffusion probe, secondary injury, 83–84 neuroimaging, 179
161–162 severity, 83 vasopressors, for
thiamine, for status temperature control, 189 hemodynamic support,
epilepticus, 72 traumatic brain injury (TBI) 202–203, 208t–209t
thrombectomy management, 88–95 vasospasm, cerebral,
acute ischemic brain tissue oxygen, 91 subarachnoid
stroke, 32–33 care setting, 88 hemorrhage, 47–49
pregnancy, 249, 250 cerebral perfusion pressure vegetative state, 4
tissue plasminogen activator goal, 88 venous thromboembolism
(tPA), IV fluid balance and prophylaxis
acute ischemic stroke, 23, electrolytes, 93 neurological patient, 222
28–29, 31–32, 32t GI prophylaxis, 94 traumatic brain
pregnancy, 248–249 glycemic control, 93 injury, 94–95
topiramate, for status hyperpyrexia, 92–93 ventilator mode, neurological
epilepticus, 77t hypoxia and oxygen patient, 214–215
transcranial Doppler (TCD), saturation, 91–92 ventriculostomy, 165–171.
brain death, 281 intracranial See also external
transcranial Doppler (TCD) pressure, 88–91 ventricular drainage (EVD)
monitoring mortality, modern care, 95 video laryngoscopy,
acute ischemic stroke, 178 nutrition, 93–94 neurological patient,
cerebral blood flow, seizure prophylaxis, 94 214
fulminant hepatic steroid avoidance, 95 viral encephalitis, 115t,
308
failure, 258 surgical intervention, 91, 92f 113–116

subarachnoid venous thromboembolism viral meningitis, 106t–107t,
hemorrhage, 175 prophylaxis, 94–95 108–109
transfusion, neurological traumatic spinal cord injury, visual fields, acute ischemic
patient, 217–218 97–103. See also spinal stroke, 25t
traumatic brain injury cord injury, traumatic
(TBI), 83–88 trazodone, 292
agitation, 292–3, 293f
cerebral ischemia
triple-H therapy, 49
W
pathophysiology, 84–85, warfarin
84f, 85f U intracerebral hemorrhage
clinical assessment and from, 39–40
ultrasound, intracranial
classification, 86–87, 86t reversal agents, 202, 203t,
pressure monitoring, 158
definitions, 83–84 217, 219t
uncertainty, fundamental, 267
diffusion weighted weaning,
urinary retention, 294–5
imaging, 174 ventriculostomy, 169
dysautonomia, 293–4 Wernicke’s encephalopathy
epidemiology, 83
monitoring, 87
V avoiding, 72
neuroimaging, 180
neuroimaging, 173–174 valproic acid, for status West Nile virus (WNV)
outcome prediction tools, epilepticus, 75t encephalitis, 115t, 113–116
267, 268t varicella Zoster virus (VZV) meningitis, 107t, 108–109

Neurocritical Care.2018 - Lori A. Shutter

Uploaded by

Copyright:

Available Formats

Neurocritical Care.2018 - Lori A. Shutter

Uploaded by

Document Information

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 - Lori A. Shutter

Uploaded by

Copyright:

Available Formats

EOITEO tY

SERIES EOITOk: JOHN A. KELLUM

Published and Forthcoming Titles

Lori A. Shutter, MD, FNCS, FCCM

Bradley J. Molyneaux, MD, PhD

Published in the United States of America by Oxford University Press

© Oxford University Press 2018

All rights reserved. No part of this publication may be reproduced, stored in

You must not circulate this work in any other form

Library of Congress Cataloging-​in-​Publication Data

We also must acknowledge the advance practice providers, nurses, pharmacists,

Section 1: Neurological Conditions

Deepa Malaiyandi and Saša A. Živković 131

Section 2: Interventions and Monitoring

Section 3: General Management and Special Populations

Opeolu Adeoye, MD, MS, Gretchen M. Brophy,

Assistant Professor MSc, FNCS, FCCM

School of Medicine University of Pittsburgh School

Division of Neurocritical Care and

Deepa Malaiyandi, MD David M. Panczykowski, MD

Professor of Critical Care Medicine Chief Fellow, Adult Critical Care

Jon C. Rittenberger, MD, MS Jeremy G. Stone, MD

Hilary H. Wang, MD, MBA

Coma and Low Arousal States

Coma is a failure of neuronal systems responsible for arousal and awareness

otherwise remaining unresponsive, which defines a vegetative state. A patient

sciousness with variable levels of arousal and inconsistent responsiveness but

Coma and Low Arousal States

Etiology Associated Findings

cyanide acidosis, high cardiac output, almond odor

Coma and Low Arousal States

efforts to avoid injury while suddenly becoming comatose. Neurologic exam

other low arousal state.

Initial Evaluation of Coma

Structural or Unclear Nonstructural

Figure 1.1 Suggested algorithm for initial diagnosis of coma etiology.

Normalization of vital signs is crucial to ensure adequate perfusion of the

Coma and Low Arousal States

Table 1.3 Laboratory Values Compatible with Coma

may alert examiners to the possibility of an endocrine derangement. Checking

sciousness. To avoid additional injury, this should be obtained after stabiliza-

Prognosis and Treatments for Coma and

Coma and Low Arousal States

Table 1.4 Commonly used Pharmacologic Interventions

Encephalopathy and Delirium

Prevalence and Relevance

Risk Factors and Causes

of underlying risk factors, delirium may be caused or worsened by multiple

o Benzodiazepines (several studies report strong effects while others

Encephalopathy and Delirium

Sedation and Delirium

pain, urinary retention, hypoxia, ventilator dyssynchrony), correcting the un-

commonly used for sedation in the ICU include propofol, benzodiazepines

Yes Tylenol 650 mg q4 PO/PR/IV PRN

Fentanyl 25–50 mcg IV PRN

Figure 2.1 Sample neurologic ICU sedation protocol.

Encephalopathy and Delirium

rate. Risk of propofol infusion syndrome is elevated in younger patients on

Encephalopathy and Delirium

multicenter, prospective cohort study. Crit Care Med. 2014;42:1899–​1909.

Library of Congress Cataloging-in-Publication Data

multicenter, prospective cohort study. Crit Care Med. 2014;42:1899–1909.

dation in mechanically ventilated patients. Nurs Crit Care. 2007;12:93–104.

ICU Care after IV-tPA Administration

patients with acute ischaemic stroke. Lancet. 1997;349:1569–1581.

coagulation studies (see also c hapter 20, Table 20.2).

ongoing, recently completed, and future clinical trials. STOP-IT, SPOTLIGHT,

III Drowsiness, confusion, mild focal deficit 14–13 Present

farction in SAH, the pathogenesis of SAH-associated brain injury is multifacto-

Pulmonary hypertension/right ventricular failure

Pulmonary embolism Chest x-ray

(OHCA). Post-arrest brain injury can be conceptualized as a primary ischemia-