Critical Care Update 2008

Critical Care Update
2008
Critical Care Update
2008
Editors
Vineet Nayyar MD DNB MPH FRACP FJFICM
Senior Staff Specialist
Intensive Care Unit
Westmead Hospital
Westmead, Sydney NSW 2145 (Australia)
JV Peter MD DNB FRACP FJFICM

Reader, Department of Medicine
Consultant, Medical Intensive Care Unit
Christian Medical College and Hospital
Vellore, Tamil Nadu (India)
Roop Kishen DA MD FRCA

Consultant in Anaesthesia and Intensive Care
Lecturer, Victoria University of Manchester
Intensive Care Unit, Hope Hospital
Salford Royal NHS Foundation Trust
Salford, Manchester M6 8HD (UK)
S Srinivas MD FNB MRCP EDIC

Consultant and Head
Department of Critical Care
Care Hospital, Banjara Hills
Hyderabad, Andhra Pradesh (India)
JAYPEE BROTHERS MEDICAL PUBLISHERS (P) LTD

New Delhi • Ahmedabad • Bengaluru • Chennai • Hyderabad
Kochi • Kolkata • Lucknow • Mumbai • Nagpur • St Louis (USA)
Published by
Jitendar P Vij
Jaypee Brothers Medical Publishers (P) Ltd
Corporate Office
4838/24 Ansari Road, Daryaganj, New Delhi - 110002, India, +91-11-43574357
Registered Office
B-3 EMCA House, 23/23B Ansari Road, Daryaganj, New Delhi 110 002, India
Phones: +91-11-23272143, +91-11-23272703, +91-11-23282021,
+91-11-23245672, Rel: +91-11-32558559 Fax: +91-11-23276490, +91-11-23245683
e-mail: jaypee@jaypeebrothers.com, Website: www.jaypeebrothers.com
Branches
 2/B, Akruti Society, Jodhpur Gam Road Satellite
Ahmedabad 380 015 Phones: +91-79-26926233, Rel: +91-79-32988717
Fax: +91-79-26927094 e-mail: ahmedabad@jaypeebrothers.com
 202 Batavia Chambers, 8 Kumara Krupa Road, Kumara Park East
Bengaluru 560 001 Phones: +91-80-22285971, +91-80-22382956,
+91-80-22372664, Rel: +91-80-32714073
Fax: +91-80-22281761 e-mail: bangalore@jaypeebrothers.com
 282 IIIrd Floor, Khaleel Shirazi Estate, Fountain Plaza, Pantheon Road
Chennai 600 008 Phones: +91-44-28193265, +91-44-28194897,
Rel: +91-44-32972089 Fax: +91-44-28193231 e-mail: chennai@jaypeebrothers.com
 4-2-1067/1-3, 1st Floor, Balaji Building, Ramkote Cross Road
Hyderabad 500 095 Phones: +91-40-66610020,
+91-40-24758498, Rel:+91-40-32940929 Fax:+91-40-24758499, e-mail: hyderabad@jaypeebrothers.com
 No. 41/3098, B & B1, Kuruvi Building, St. Vincent Road
Kochi 682 018, Kerala Phones: +91-484-4036109, +91-484-2395739,
+91-484-2395740 e-mail: kochi@jaypeebrothers.com
 1-A Indian Mirror Street, Wellington Square
Kolkata 700 013 Phones: +91-33-22651926, +91-33-22276404,
+91-33-22276415, Rel: +91-33-32901926 Fax: +91-33-22656075, e-mail: kolkata@jaypeebrothers.com
 Lekhraj Market III, B-2, Sector-4, Faizabad Road, Indira Nagar
Lucknow 226 016 Phones: +91-522-3040553, +91-522-3040554
e-mail: lucknow@jaypeebrothers.com
 106 Amit Industrial Estate, 61 Dr SS Rao Road, Near MGM Hospital, Parel
Mumbai 400012 Phones: +91-22-24124863, +91-22-24104532,
Rel: +91-22-32926896 Fax: +91-22-24160828, e-mail: mumbai@jaypeebrothers.com
 “KAMALPUSHPA” 38, Reshimbag, Opp. Mohota Science College, Umred Road
Nagpur 440 009 (MS) Phone: Rel: +91-712-3245220,
Fax: +91-712-2704275 e-mail: nagpur@jaypeebrothers.com
USA Office
1745, Pheasant Run Drive, Maryland Heights (Missouri), MO 63043, USA,
Ph: 001-636-6279734 e-mail: jaypee@jaypeebrothers.com, anjulav@jaypeebrothers.com
Critical Care Update 2008

© 2009, Jaypee Brothers Medical Publishers
All rights reserved. No part of this publication should be reproduced, stored in a retrieval system, or transmitted in any form
or by any means: electronic, mechanical, photocopying, recording, or otherwise, without the prior written permission of the
editors and the publisher.
This book has been published in good faith that the material provided by contributors is original. Every effort is made to
ensure accuracy of material, but the publisher, printer and editors will not be held responsible for any inadvertent error(s).
In case of any dispute, all legal matters are to be settled under Delhi jurisdiction only.
First Edition: 2009

ISBN 978-81-8448-614-8
Typeset at JPBMP typesetting unit
Printed at Ajanta Press
List of Contributors
Sumesh Arora MD, EDIC, FJFICM Peter T Clark FACEM, FANZCA, FJFICM
Staff Specialist ICU Senior Staff Specialist
Prince of Wales Hospital Westmead Hospital
Randwick, Sydney (Australia) Westmead, Sydney (Australia)
Sean M Bagshaw MD, MSc, FRCPC Mathew Joseph MCh (Neurosurgery)

Division of Critical Care Medicine Professor of Neurosurgery
University of Alberta Hospital Department of Neurological Sciences
3C1.16 Walter C Mackenzie Centre Christian Medical College and Hospital
8440-122 Street, Edmonton (Canada) Vellore, Tamil Nadu (India)
Ian Baguley MD, PhD, FFARM
Associate Professor, Macquarie University Dilip R Karnad MD, FACP, FRCP
Senior Lecturer, University of Sydney Medical Intensive Care Unit
Department of Rehabilitation Medicine Department of Medicine
Westmead Hospital King Edward Memorial Hospital
Westmead, Sydney (Australia) Parel, Mumbai
Maharashtra (India)
HS Bawaskar
Physician Ravin Kishen MBChB
Bawaskar Hospital and Research Centre Specialist Trainee in Emergency Medicine
Mahad, District Raigad Edinburgh Royal Infirmary
Maharashtra (India) 51 Little France Cr, Old Dalkeith Road
Edinburgh (United Kingdom)
PH Bawaskar
Physician
Bawaskar Hospital and Research Centre Roop Kishen MBBS, DA, MD, FRCA
Mahad, District Raigad Consultant in ICU and Anaesthesia
Maharashtra (India) Lecturer
Victoria University of Manchester
Sara Blakeley BM, MRCP, EDIC Intensive Care Unit
Consultant in Intensive Care Medicine Hope Hospital
Department of Critical Care Salford Royal NHS Foundation Trust
Queen Alexandria Hospital Salford, Manchester (United Kingdom)
Portsmouth (United Kingdom)
Ravindra L Mehta MD, DM, FACP, FASN
RM Chakravarthi MD, DNB (Nephro) Professor of Medicine
Consultant and Head Director of Clinical Nephrology
Department of Nephrology UCSD Medical Centre
Care Hospitals, Banjara Hills 200 West Arbor Drive #8342
Hyderabad, Andhra Pradesh (India) San Diego (United States of America)
vi Critical Care Update 2008
Shalini Nair Saxon Ridley MD, FRCA

Associate Professor Consultant in Anaesthesia and Intensive Care
Department of Neurological Sciences Glan Clwyd Hospital, Rhyl
Christan Medical College and Hospital Denighshire (United Kingdom)
Vellore, Tamil Nadu (India)
Shamik H Shah MD, DNB (Nephro)
C Narasimhan MD, DM, AB (Card) Medical Director
Consultant Cardiologist and Electrophysiologist ARCS India Dialysis Centres
Director, Electrophysiology Labs D-92 Sector 63
Department of Cardiology Noida, Uttar Pradesh (India)
Care Hospitals, Banjara Hills
Hyderabad, Andhra Pradesh (India) Karim Somani MD, FRCS(S)
Division of Critical Care Medicine
Vineet Nayyar MD, DNB, MPH, FRACP, FJFICM
University of Alberta Hospital
Senior Staff Specialist ICU
3C1.16 Walter C Mackenzie Centre
Westmead Hospital
8440-122 Street, Edmonton (Canada)
Westmead, Sydney (Australia)
JV Peter MD, DNB, FRACP, FJFICM Ravikiran Sonawane MD

Physician Medical Intensive Care Unit
Medical Intensive Care Unit Department of Medicine
Christian Medical College and Hospital King Edward Memorial Hospital
Vellore, Tamil Nadu (India) Parel, Mumbai, Maharashtra (India)
Kishore Pichamuthu MD, DNB (Med) S Srinivas MD, FNB (Crit Care), MRCP, EDIC
Assistant Professor Consultant and Head
Medical Intensive Care Unit Department of Critical Care
Christian Medical College and Hospital Care Hospitals, Banjara Hills
Vellore, Tamil Nadu (India) Hyderabad, Andhra Pradesh (India)
T Prithviraj MBBS, DNB (Anesth), Dip Crit Care Gareth L Thomas B (Med Sci), MRCP, FRCA, EDIC
Fellow Consultant in Intensive Care Medicine
Department of Critical Care Medicine Salford Royal NHS Foundation Trust
Sundaram Medical Foundation Salford (United Kingdom)
Anna Nagar, Chennai, Tamil Nadu (India)
FE Udwadia MD, FRCP, FCCP, FAMS, FCPS
Ram E Rajagopalan MD, AB (Med) AB (Crit Care)
Emeritus Professor of Medicine
Consultant and Head
Grant Medical College and
Department of Critical Care Medicine
Sir JJ Group of Hospitals
Sundaram Medical Foundation
Consultant Physician and Director ICU
Anna Nagar, Chennai, Tamil Nadu (India)
Breach Candy Hospital
Hygriv Rao MD, DM Mumbai, Maharashtra (India)
Consultant Cardiologist and Electrophysiologist
Department of Cardiology
Care Hospitals, Banjara Hills
Hyderabad, Andhra Pradesh (India)
Preface
As senior editor of the series, I have retained the prerogative of writing the
preface for this year’s Critical Care Update. This year, as in previous years,
the Update contains papers on a range of topics of particular interest to
trainees and clinicians working in the ICU. The Update retains an easy to
read format and familiar themes, which I hope will encourage uptake of key
concepts by readers. Some papers are topical and have their origins in
presentations at recent international meetings, while others revisit old issues.
All have been chosen to provide a balanced mix of topics that will appeal
to most readers.
As in previous years, the first section of the Update carries a series of articles that defines
a theme for the book. The focus area, this year, is Communication. As doctors, we are
taught to cure disease, treat patients, and relieve pain. However, in the course of our
professional duties, we often come across situations where treatment and clinical management
become futile and hope of remission or survival recedes. These situations are frequently
encountered in the ICU. It then becomes the doctor’s duty to inform the patient and/or his/
her relatives of the situation in a caring, empathetic manner. Talking about death and dying
are not easy and doctors find it difficult to confront patients and relatives with bad news.
Although there are no simple answers, the initial three chapters attempt to outline essential
aspects of the art of communication along with some practical tips for senior colleagues as
well as trainees in Intensive Care. Editorial team members have contributed almost exclusively
to this section, for which I am truly grateful. I have enjoyed interacting with the editorial team
towards the production of this book and have increasingly leaned on their support for editorial
work this year.
Individual authors have set aside personal time to make a contribution to the Update this
year. It is no easy feat to sit down and organise a chapter of several thousand words and
meet the deadline for submission of the manuscript. I therefore acknowledge their readiness
to be part of this endeavour. As usual, most papers have been solicited from experts in the
field and have been subjected to minimal peer review. M/s Jaypee Brothers Medical Publishers
(P) Ltd, New Delhi, have once again excelled in the task of bringing out the Update for the
eighth consecutive year. Without their high quality inputs and timely publication, yearly
production of the Update Series would have been impossible.
Sydney (Australia) Vineet Nayyar

31 December 2008
Contents
Section A: Communication
1. Conversations in Medicine 1
Vineet Nayyar
2. Communicating Sad, Bad and Difficult News in the ICU and ER 10
Ravin Kishen, Roop Kishen
3. Inter-professional Communication and Teamwork 21
S Srinivas
Section B: Cardiovascular Problems in ICU

4. What should an Intensivist Know about Pacemakers? 30
Hygriv Rao, C Narasimhan
5. Role of Amiodarone in Recent Onset Atrial Fibrillation 42
Sumesh Arora, Vineet Nayyar
6. Defibrillation during CPR—Is a Delayed, Single, Biphasic Shock Better 56
Ram E Rajagopalan, T Prithviraj
7. Importance of Assessing the Microcirculation in an ICU Patient 63
Saxon Ridley
Section C: Renal Problems in ICU

8. Biomarkers in Acute Kidney Injury 74
RM Chakravarthi
9. Septic Acute Kidney Injury: Epidemiology and Pathophysiology 83
Karim Somani, Sean M Bagshaw
10. Anticoagulation during CRRT—Does Citrate Offer an Advantage? 97
Shamik H Shah, Ravindra L Mehta
11. Can the Choice of Resuscitation Fluid Adversely Affect Renal Function 112
in the Critically Ill?
Roop Kishen, Sara Blakeley
Section D: Trauma/Neuro-critical Care

12. Sodium-Water Disturbances in a Neuro ICU 122
Mathew Joseph, Shalini Nair
x Critical Care Update 2008
13. Delirium in the ICU and its Consequences 133

Gareth L Thomas
14. Cervical Spine Clearance in Obtunded Multi-trauma Patients 146
Peter T Clark
15. Dysautonomia Following Traumatic Brain Injury 155
Ian Baguley
Section E: Tropical Diseases in ICU

16. Artesunate as First Line Therapy for Severe Falciparum Malaria 166
Ravikiran Sonawane, Dilip R Karnad
17. Management of Airway and Autonomic Dysfunction in Tetanus 175
FE Udwadia
18. Management of Haemodynamic Instability after Scorpion Sting 184
HS Bawaskar, PH Bawaskar
19. Heat Stroke 193
Kishore Pichamuthu, JV Peter
Index 211
Conversations in Medicine 1
Section A: Communication
One
Conversations in Medicine
Vineet Nayyar
“The secret of the care of a patient is caring for the patient.”
INTRODUCTION
Skill in communication is a matter of personal ability, which varies widely between individuals
in health care as in any other field. Many doctors handle conversations with their patients
well. Some, on the other hand, do not show enough confidence or empathy in their interactions
with patients or worse still, demonstrate a cold, uncaring attitude. Some are reluctant to engage
in a conversation altogether, fearing the emotional impact of the news they seek to
communicate.
The practice of medicine requires constructive interaction with patients, relatives, colleagues
and managers. A typical career in clinical medicine involves over 100,000 patient interviews1
and innumerable other conversations. Communication, both verbal and non-verbal, is central
to every clinical encounter and may, sometimes be the only component of patient care in
someone terminally ill.
THE NEED FOR GOOD COMMUNICATION

Good communication is important because it forms the basis of all transactions in healthcare.
Evidence from recent research suggests that good communication improves specific medical
outcomes for patients, positively affects patient perception and improves quality and safety
in healthcare. However, for this to be accomplished communication needs to be patient-
centred, informative and interactive.
Better Outcomes
The ultimate objective of healthcare is improvement in outcomes. Not surprisingly, these can
be achieved as much by effectively communicating to the patient the need for lifestyle changes
as with the introduction of a new drug. A wide variety of medical outcomes improve as a
2 Critical Care Update 2008
result of effective communication. Examples include improved postsurgical pain, severity of

headaches, compliance with medications and better glucose control.2 This improvement in
outcomes is related, in part, to adherence (or compliance) with prescribed treatment. Doctors,
who are skilled communicators “enlist” their patients better.
Malpractice
All doctors make mistakes but not all get sued. Research has examined the difference between
clinicians who are named in malpractice suits and those who are not. Poor communication
and relationship is the most important factor to emerge from this investigation. When
researchers reviewed the medical records from a group of doctors with higher than average
malpractice claims, they found no difference in the number of technical or clinical errors
compared to a control group of doctors with no claims. They found that sued doctors were
no more likely to have adverse outcomes compared to others but they did have a significantly
higher rate of dissatisfaction and complaints against them. The “no claim” doctors spent, on
average, more time with patients and engaged and interacted with them more. Not
unexpectedly, they found patient satisfaction to be the highest, if a doctor was a good
communicator.3,4
Physician Satisfaction
Most doctors accept the incredible long hours of work because they expect a very high degree
of satisfaction in their careers as physicians. Higher ratings of job satisfaction among physicians
appear to be closely linked to good relationships and therefore, to good communication with
patients and others health professionals.
Physicians rate their relationship with patients as important in feeling good about their
career. When asked to report on their careers, physicians from a variety of specialties rated
good relationships with patients, relatives and staff as the greatest contributors to job satisfaction.
Job satisfaction, in turn, appears to protect physicians from effects of job stress and prevents
burn out.5 Physicians who described themselves as inadequately trained in communication
skills report a higher rate of burn out compared to those with good skills.6
Better Quality and Safety

Several studies have highlighted that poor communication between healthcare workers
contributes significantly as a latent source of medical error.7 Evidence exists that patients
have better outcomes when doctors and nurses communicate effectively about patient care.8,9
Not only is the likelihood of error minimised, the extent and inclusiveness of communication
across role boundaries impacts positively on patient safety. At the very least, good and open
communication influences the degree to which patient care duties are understood and executed.
BARRIERS TO GOOD COMMUNICATION

There are formidable barriers to successful communication across the divide between patients
and doctors. Communication gaps sometimes limit the real good that medicine has the power
to deliver. Some of these barriers have been identified and they include the following.
The Kuhnian Gap

Kuhn pointed out that as a science becomes more successful, it becomes harder to
communicate the latest advances in simple terms.10 Eventually scientists are able only to
communicate with other scientists working within the same discipline. This theory has, as its
basis, observations made during the course of discoveries in quantum mechanics and the
string theory.
A somewhat similar phenomenon that applies to medicine is related to the fact that as
medical systems get better at delivering high quality health care, there seems to be a paradoxical
increase in the discontent associated with it. Hostile media, increasing litigation, lack of trust
and high-level enquiries are manifestation of this trend that, in turn, adversely affects the way
doctors choose to communicate with their patients.
Explaining Probability
The complexities of probability, as it relates to clinical decision making, makes conversations
between doctors and patients more difficult than they already are. Clinicians armed with
frequential probabilities as their ‘objective data’ are forced to rely upon subjective probabilities
when they deal with one patient. They cannot try their treatment 100 times on each patient.
Doctors must, therefore express their level of confidence in each choice, or offer the patient
an array of data and ask them to choose the treatment. This remains a difficult task even for
the most enlightened. In other words, the process of helping patients arrive at a decision is
complex and challenging.
Facts taken into account depend to some extent on individual clinician’s commitment to
one hypothesis over another. Given the same facts, three different clinicians give different
advise because of what they perceive as patient need. It is inevitable for patients and their
families perplexed by the probabilistic nature of scientific data to become even more so when
provided with multiple opinions. While it is only natural for someone with a serious illness to
seek a second or third opinion, this often leads to increasing confusion. When opinions are
consistently pessimistic or indefinite, it is hardly surprising that many patients turn away from
traditional medicine to alternative forms of therapy to find a cure.
Unfortunately, probability permeates the whole of clinical medicine, as it does most of
nature. The lay public, however, has the impression that all good science is deterministic and
that medicine is a science. The profession does little to correct that view. As a result, it becomes
very difficult for patients to understand that medical advice about outcome of treatment is
nearly always indefinite, because it can only be expressed as a probability without any
guarantee. Decision theory may help in some situations but no amount of numbers ever help
convince a patient’s family that they are receiving the ‘best’ and most empathic advice.
Understanding these difficulties are the first step in overcoming problems that arise in
conversations with patients.
Information Flood
Medical information is generated at a staggering rate of 34,000 articles each month in medical
journals. A new article is added to medical literature every 26 seconds.11 The quality of articles
worthy of attention has increased substantially over the last 10 years with nearly 200,000
RCTs indexed in MEDLINE between 1994 and 2001. For even the most attentive, it is simply
no longer possible to keep up to date by reading the latest, as the volume of published material
far exceeds the human capacity to read and understand it all.
Most doctors are not skilled at separating information from knowledge, and many are
unaware of this distinction. Patients have still less of a chance of being able to appreciate all
the information available on their specific illness. The media often promote disinformation by
hailing some break through in medicine or by highlighting the benefits of one procedure over
another. This distorts the context in which conversations between doctors and families take
place.
Vested Interests
There are several vested interests in and around medicine. Certain groups within and outside
the profession, use health issues to their advantage and are reluctant to give up their position
or privilege. While the motive of most doctors is good, the necessity of curing for profit creates
a set of vested interests in medicine that is hard to ignore. Most doctors want to preserve or
increase their income; on the other hand patients want to minimise expenditure. This creates
a tension in the relationship between doctors and patients and adds to difficulties in
communication.
Social Distance and Linguistic Differences

Social class, age, race, culture and ease of communication influence a patient’s trust and co-
operation with the doctor. The possibility of misunderstanding arising out of communication
difficulties is substantial. Quite apart from obvious problems posed by patients who do not
speak the same language as their doctor, conversations in a shared language that is understood
differently by the patient can also seriously compromise communication. Common medical
terms such as cancer, shock, stroke and heart attack are poorly understood by the general
public. Using them in explanations tends to inadvertently create the impression of subsequent
disability and suffering, which the doctor never intended to convey.
A wider social distance between doctor and patient makes conversations difficult. A patient
who has marked feeling of social inferiority is less likely to ask questions or initiate discussion.
This in turn, aggravates suspicions and reinforces pre-existing prejudice and distrust.
COMMUNICATION SKILLS CAN BE TAUGHT

To some extent communication skills can be taught even though the context of each
conversation is different. Most doctors do not get the same type of clear, systematic guidance
in learning communication skills as they do with other procedures during their training years.
As with clinical examination techniques, communication can be mastered over time.
The gold standard teaching tool is individual feedback on videos of real meetings. A
classification of communication skills has been proposed by the Calgary Cambridge programme
to analyse content and provide feedback. Three domains make up this framework: content
skills, process skills and perceptual skills. Evidence suggests that there are major differences
in competency levels among doctors in individual domains.12
Content Skills
This component is easy to understand and assess. It consists of what is actually said, including
the language used and information given. Most of the skills are intuitive such as avoiding
technical language and using a level of explanation appropriate to the relative’s understanding
of events.
Process Skills
These skills consist of how a meeting with the family is structured. Although some doctors
develop process skills spontaneously, most do not. The doctor, who has trouble getting across
all he/she wanted to say or who finds the start or finish of a conversaion awkward, may
benefit from learning process skills.
Perceptual Skills
These relate to recognising and dealing with feelings and emotions that arise in the course of
a conversation. Experience alone is a poor teacher of process and perceptual skills and there
is some evidence that these skills deteriorate over a medical career13. Examining perceptual
skills can reveal underlying problems of attitude and relationships, but these can be improved
with training.
MANAGING DIFFICULT COMMUNICATION TASKS

There are a number of difficult communication tasks, which doctors have to carry out including
breaking bad news of a serious illness, coping with denial or managing patient distress or
anger. Doctors who work in critical care face, in addition, special circumstances during which,
well thought out levels of communication are Table 1.1: Clinical situation in ICU
requiring of effective communication
particularly important14 (Table 1.1). An approach to
some of these tasks is detailed in the section below. • Consent for intervention
• Explaining medical risks
Conveying Prognosis • Uncertain prognosis
• Breaking bad news
There is justifiable major interest in defining medical
• Unexpected death
outcomes and communicating these to family
• Procedure related complications
members of critically ill patients. When prognosis is
• Disclosure of error
poor, it is important to avoid instilling false hope.
• Advance directives
Hedging and focussing on irrelevant positives such a
• Discussion regarding CPR
pulse or blood pressure serve to distract the family
• Withdrawal of life support
from the work of acceptance that they must begin.
• Limitation of active therapy
Dealing with Difficult Questions • Brain death and organ donation
The world of medicine has become increasingly

foreign to lay people, so patient families have a number of questions, some unspoken, that
need clarification. Simple concerns about the diagnosis and interpretation of test results are
easy to address. Concerns about available options, especially those that require a procedural
intervention or a need for further tests are more difficult to answer. Even more difficult are
questions related to the future course of events or questions about why the illness occurred.
Handling Denial
When faced with denial, it is tempting to force family members to confront reality. This may
turn out to be counter-productive. Denial is generally a temporary defence mechanism and
lasts until family is ready to face the reality. Patience and repeating the same message over
and over again is the key to managing denial. Occasionally, certain inconsistencies in the
arguments used by families need to be confronted. Challenging inconsistencies works frequently,
but on occasions when they fail to dent denial, the treating doctor must work on a different
approach to develop awareness among family members.
Overcoming Unrealistic Expectations

The antipathy that develops between family and doctors as a result of unrealistic expectations
can be prevented by early, honest and consistent communication. Public expectations are
generated by socio-cultural norms and are sustained by the media and access to medical
information. Once formed, expectations are difficult to change, particularly when it is necessary
to lower them. Medical profession unwittingly encourages these expectations by emphasising
the power of science to solve any problem. The expectation that medical science should cure
every illness is thus, taken for granted.
Unrealistic expectations can be avoided by assuring consistency in communication, early

focus on the patient as a person and his / her personal preferences, involving other specialists
for support, offering realistic time limited trials of treatment with clear end points and re-
emphasising the available alternatives.15
Breaking Bad News

No doctor wants to give bad news, but most accept it as part of their duty. There are special
skills that can help effectively communicate bad news. In addition, qualities such as empathy
are crucial in engaging with those receiving bad news. This is covered in greater detail in the
next chapter.
Managing Distress
Family members can become tearful or upset while receiving bad news. Sometimes, it is best
to say nothing and allow for tears to flow. At other times, the assurance that it is perfectly
acceptable to feel sad can be helpful. Crying patients or relatives tend to make doctors anxious
and do something silly like leave the room or send in someone else to comfort the grieving.
Empathic lapses such as these can leave a lasting impression.
Distressed and angry relatives pose a special challenge because of safety issues. Family
members may become angry for many reasons, but most often this is due to an underlying
fear or sadness. Regardless of the source of anger, doctors must address safety concerns first
and ask for additional members of staff or other family members to help.
Avoiding Collusion
Although honesty and transparency in matters between the doctor and patients is the expected
norm, there exist family members who seek to protect their sick relatives from information
that is likely to cause undue distress. These families often seek the doctor’s support in misleading
the patient, for example by not discussing the diagnosis or its seriousness. This must be gently
but firmly avoided.
Resisting collusion with relative to hide facts from a patient, however, requires tact and
great patience. Part of the motivation to prevent all information from being disclosed stems
from the family members own dread of witnessing the distress of a loved one. This is
understandable and yet, the fear can be countered by upholding the principle of honesty.
Facing the truth together is associated with short-term pain but is usually followed by increased
closeness and comfort. Hiding from reality ends up hurting rather than enhancing acceptance.
Working within a Team

Good inter-professional communication influences the quality of care and patient safety. Regular
rounds, documentation of important treatment goals and review of progress by the ICU lead
clinician has been shown to reduce ICU length of stay.9 Failure of inter-professional
communication, on the other hand, is associated with avoidable medical mishaps.16
Handing over clinical information at change of shifts in the ICU and at the time of a
patient’s discharge is important. Both verbal and written communication is needed. Skills for
communicating in a crisis situation are necessary while working as a team in the ICU. This
is covered in a separate chapter in this book.
Negotiating with Colleagues

The way in which clinical management of an ICU patient is negotiated with other specialists
has received little attention.12 Clear communication is also central to clarifying and resolving
differences of opinion that arise in the course of management of an ICU patient. However,
little is known about how best this can be achieved. At the very least, ongoing contact with
the primary team needs to be actively encouraged in order to build a joint understanding of
the patient’s progress through the ICU.
CONCLUSION
It is now widely recognised that good communication skills are a necessary feature of good
medical practice. Conversations with patients or their near ones must be effective and ethical.
A measure of whether this is the case or not, is reflected in the extent to which patients
express their satisfaction with the care and support they receive in hospital. Healthcare
professionals have traditionally learnt communication skills by modelling the practice of their
seniors. Although a time-honoured method, this results in a series of anecdotal lessons that
rarely equip the doctor to handle difficult communications tasks in the ICU. With increasing
emphasis on teaching communication tasks and structured feedback, doctors can expect to
develop a number of skills that, in turn, can improve teamwork and help patients and relatives
through difficult times.
REFERENCES
1. Bayer Institute for Healthcare Communication. Clinician-patient communication to enhance health
outcomes: a workshop manual. West Haven, CT Bayer Institute, 1998.
2. Stewart MS. Effective physician-patient communication and health outcomes: a review. Can Med Assoc
J 1995;152:1423-33.
3. Hickson GB, Clyton EW, Entman SS, et al. Obstetricians’ prior malpractice experience and patients’
satisfaction with care. JAMA 1994;272:1583-8.
4. Levinson W, Roter DL, Mullooly JP, et al. Physician-patient communication: the relationship with malpractice
claims among primary care physicians and surgeons. JAMA 1997;277:553-9.
5. Ramirez AJ, Graham J, Richards MA, et al. Mental Health of hospital consultants: the effect of stress and
satisfaction at work. Lancet 1996;347:724-8.
6. Graham J, Ramirez AJ. Mental health of hospital consultants. J Psychosom Res 1997;43:227-31.
7. Sutcliffe KM, Lewton E, Rosenthal MM. Communication failures: an insidious contributor to medical mishaps.
Acad Med 2004;79:186-94.
8. Dodek PM, Raboud J. Explicit approach to rounds in an ICU improves communication and satisfaction
of providers. Intensive Care Med 2003;29:1584-8
9. Pronovost PJ, Berenholtz SM, Dorman T, Lipsett PA, Simmonds T, Haraden C. Improving communications
in the ICU using daily goals. J Crit Care 2003;18:71-5.
10. Little JM. Humane Medicine. Cambridge UK, Cambridge University Press, 1995
11. Nayyar V. Even information needs intensive care. In Nayyar V (Ed). Critical Care Update, Jaypee Brothers,
New Delhi, 2005.
12. Gauntlett R, Laws D. Communication skills in critical care. Cont Ed Anaesth Crit Care Pain 2008;8(4):121-
4.
13. Aspergen K, Lonberg-Madsen P. Which basic communication skills in medicine are learnt spontaneously
and which need to be taught and trained? Med Teach 2005;27:539-43.
14. Macdonald E. The doctor’s perspective. In Macdonald E (Ed). Difficult conversations in Medicine. Oxford,
Oxford University Press, 2004.
15. Meier DE. Communication failure in the ICU. Virtual Mentor 2006;8(9):564-70.
16. Reader T, Flin R, Lauche L, Cuthbertson BH. Non-technical skills in the intensive care unit. Br J Anaesth
2006;96:551-9.
Two
Communicating Sad, Bad and Difficult
News in the ICU and ER
Ravin Kishen, Roop Kishen
“Practice of medicine involves communicating lots of little bad news everyday”
INTRODUCTION
In 2007, the National Confidential Enquiry into Patient Outcome and Death (NCEPOD), UK,
published a report on emergency admissions into the UK hospitals. In his introduction to the
report,1 the chairman of NCEPOD, Prof. Tom Treasure recalls a funeral he attended where
the husband of the deceased, a lung cancer victim, said ‘Thank God it was cancer’. The
bereaved husband was thankful for the time he and his family had (18 months, in this case)
to prepare and accept the inevitable.
In the event of major trauma, critical illness or sudden unexpected deterioration among
patients admitted to emergency and intensive care units, such time is usually not available to
the patient or the family. The very suddenness and the nature of critical illness makes it
impossible, or at the very least, difficult for them to understand or come to terms with acute
life threatening illness. Whereas, in situations like cancer, health care workers (HCWs) and
attending doctors have time to get to know the patient and the family; in critical care and
emergency medicine, there is very little time for establishing relationships. Palliative care
specialists, hospice workers and physicians treating chronically ill patients have opportunities
for detailed discussions, repeated interviews and unhurried conversations, often not possible
in critical care or emergency medicine setting. Nonetheless, bad and difficult news has to be
conveyed more often in critical and emergency care than in most other areas of the hospital.
Effective communication with sensitivity and clarity is vital for breaking bad and difficult news.
Failure to do so results in dissatisfaction and formal complaints against treating doctors.2
SCOPE OF THE PROBLEM

Bad news is any information that is likely to drastically change a patient’s (and/or relatives’)
view of their future life in a negative way, either at the time of diagnosis, or when faced with
futility of further support or curative treatment as in Intensive Care Units (ICU) or Emergency
Communicating Sad, Bad and Difficult News in the ICU and ER 11
Rooms (ER).3 Bad news is also the news or information that changes the cognitive behaviour
of the receiver.4
It has long been recognised that even though doctors are highly trained professionals,
they often find it difficult to communicate bad news.3-5 Doctors and other HCWs often deal
with death and dying in their professional lives; however, they avoid talking about death that
may be imminent or already taking place. Until recently, doctors not wanting to discuss death
or futility of treatment, withheld truth and sometimes even falsely reassured patient and/or
relatives by predicting a better prognosis and longevity than the situation warranted.6 Many
justified this on the basis that patients and relatives needed hope and withholding truth, at
least initially, is good practice.7 With changing times and patients’ expectations and more
readily accessible medical information, withholding of information except at the patient’s
request8 is generally unacceptable. In the West, 98% of patients prefer a realistic and
individualised approach when receiving bad news.9
Medical literature is full of articles that deal with ways and means of giving bad news to
oncology and paediatrics patients. Unfortunately, much less has been written about breaking
bad news to ICU or ER patients; in fact, extensive search of literature reveals very few articles
on the subject. As has been pointed out above, giving bad news in emergency situations has
to be done with little or no previous contact with either the patient or their relatives—a situation
quite different from oncology. Oncologists have developed protocols and guidelines for giving
bad news to their patients (e.g., SPIKES6 protocol—see below). However, what oncologists
may achieve in several sittings, intensivists and emergency physicians have to do so in a
single, often hurried conversation.
Doctors’ Dilemmas and Difficulties

It has already been stated that doctors find breaking of sad and difficult news is not easy.3,5
There are many reasons for this. Doctors fear that they will be blamed for patient’s poor
prognosis, incurable disease and untimely death3,6 (not unlike the execution of bearers of
bad news in ancient times). Many doctors are also fearful of the reaction they may evoke in
patients or relatives when they deliver bad news and may be reluctant to handle the
consequences.6,10 They are also apprehensive that they may not be able to answer difficult
questions or help with future plans.6 These fears and doubts are in most part, related to
inadequate or no training in communication skills in the formative years of medical training
at undergraduate and postgraduate level.11
Death and dying are an integral part of life and are encountered frequently by HCWs in
the course of their professional duties. Yet, many find it difficult to speak of death. Socially,
most doctors are brought up in a society where talking about death and dying is a taboo.
Superimposed on this, is the dominant medical culture that invariably considers death as
treatment failure. Not surprisingly, junior doctors have great difficulty in breaking bad news
and handling the emotional reaction that follows.
Studies in last two decades have shown that communicating bad news needs improving,
not only because the impact it has on patients and relatives but also for the psychological well
being of doctors themselves. While doctors use different ways of communicating with the
patients and their relatives, when it comes to giving bad and difficult news, each one develops
his own way of coping with patient’s or relatives’ reaction. In general, doctors tend to regard
death with negative feelings of sadness and shame. They report personal stress when
communicating bad news; emotions like sorrow, guilt and inadequacy, carrying blame for
what has happened and feeling of failure are prominent among them.5 This situation is not
helped by inadequate training or being left alone by their seniors to communicate bad news
early in their career.5
Traditionally, undergraduate medical curricula have paid little attention to communication
skills. Although training in conversational skills has formed part of many medical school
programmes in Europe and North America for the last decade or so, the overall emphasis has
been on communication skills in general. Many young doctors, however, still feel under
prepared for communicating effectively with dying patients and their relatives because this is
not emphasised during training.12 Consequently, young physicians continue to remain deficient
in important competencies of communication.
What kind of difficult news is communicated in ICU and ER

In oncology and general wards, diagnosis, confirmation of diagnosis following tests and/or
planned treatment and long-term prognosis are the usual kinds of information given to patients.
In these conversations, patients are usually present and take full part in exchange of information,
unless they desire not to take part and have indicated so. In the emergency room, it is the
resuscitation status, failed resuscitation, immediate surgical or medical interventions to save
the patient’s life and transport to other facility or operating rooms or admission to intensive
care unit (ICU) that are the usual subjects for discussion with the relatives. Most of the patients
are incapable of either understanding their clinical condition or of giving consent for further
management. Consider a typical scenario in the emergency room; an out-of-hospital cardiac
arrest has been brought in and resuscitation has failed and the distraught relatives are waiting
outside. This difficult situation demands an immediate interview with relatives who may not
even know that their relative has died. In ICUs there is a little more time but again, the patient
may not be able to take part in the interview. Here, the scenario may be slightly different, in
that the patient improved initially but later developed multiple organ failure. Relatives think
that the patient is getting better but the doctor has to convey a different message. These and
other situations are somewhat different from the calm atmosphere of an outpatient consultation,
and although the principles of giving bad news are the same, essential differences exist as
they apply to ER and ICU.
What should the patients and the relatives be told

It is generally accepted that doctors should be honest and forthright about what they
communicate with patients and/or their relatives.8 When death has already occurred (as
happens sometimes in the emergency room) or is imminent (critical care or emergency), truth
cannot and must not be withheld and relatives must be told without undue delay. How this
news is broken is a matter for the clinicians and nurses to decide. Studies have shown that
most, if not all, patients (and their relatives) appreciate being told the truth about their (or
their relatives’) condition.9 However, there are situations and times, when doctors may not
want to disclose the truth or where the patients (and generally their relatives) prefer not to
know it. It is argued that such circumstances arise when the physician is requested to disclose
the diagnosis or prognosis in a staged manner. This helps to preserve hope in relatives, provides
them with coping mechanisms and avoids swings in relatives’ expectations when the patients
seem to rally early in the course of their illness (ICU) or resuscitation (emergency room).7
Fallibility of the physicians’ own judgement of poor prognosis in the early phase of illness is
another reason for staged disclosure of truth.7 It thus becomes important for disclosing doctors
to ask the patients and/or their relatives as to how much they want to know and who such
information should be disclosed to (Invitation of SPIKE protocol—see below). Although this
approach may well work in oncology, it does not always work in ER or ICU as little is gained
by ‘staged disclosure’ when death has occurred or is imminent!
HOW SHOULD BAD NEWS BE BROKEN

The issue of breaking bad news has been dominated by anecdotal opinion for many years.
Senior clinicians’ experience and the way they have dealt with this issue has been the ‘de
facto’ standard that has lead the way for many doctors until recently.13
Furthermore, there has been very little appreciation of how patients and relatives receive
this news, what they think of the HCWs who give this news and the effect, delivering such
news has on doctors and HCWs. In one study of communicating cancer diagnosis, patients
(and relatives) describe feelings of a sense of shock, fright, sadness, desperation and
acceptance.14 Other accounts corroborate these findings;5 however, few accounts or studies
are available for situations in ICU and ER in this regard. Nonetheless, patients associate poor
delivery of news with bluntness on part of deliverer of the news or initiation of such
conversations at inappropriate times and places.15
Bad news is not easy to absorb—the human mind tends to block out the unpleasant,
dreadful reality. Most humans block out bad news by ‘it is not happening to me’ compensating
mechanism. Thus, a lot of information that is given may be forgotten. In appropriate situations
(e.g., a dying patient in ICU), it is important to repeat the same information later on in the
day (or night). It is also useful to remind relatives that they will forget much of what has been
said and they should not hesitate to ask for information again; this tells them that they are
being cared for and have not been forgotten.
SPIKES METHOD6
The SPIKES protocol is intended to help doctors improve communication in an empathetic
way. Devised after research into how bad news is given and how relatives and patients want
to receive such news, the protocol offers a 6 step approach to preparation for and delivering
of bad and difficult news. These 6 steps are described below (the authors have also included
their experiences here).
S–Setting
It is recommended that a secluded and private location is chosen. The doctors are advised
to become familiar with the patient’s history, especially diagnosis and plan for further treatment
as well as have some idea about the prognosis. The doctor should rehearse the questions that
may be asked and think of answers ahead of time. Some thought should be given to how the
patient’s (relatives’) reactions may be handled. It is highly advisable to have a nurse, especially
the one who has been looking after the patient. When going into the interview room, introduce
yourself, extend your hand and shake hands with the relatives present (according to local
custom). Ask who you are talking to. Identify a lead person in the family (may be the spouse,
eldest child or other closely related person). In an Indian setting, everyone, including neighbours
want to be ‘in on the conversation’. Discourage this as much as you can (can be very difficult)
and only speak to close relatives.
P–Perception
Ask what the patient (relative) knows? They may have been expecting bad news. ‘We know
it is bad news and he is bleeding seriously’. This makes the conversation easy to start. On the
other hand, they may not know anything about their condition and then it becomes important
for the doctor to introduce the situation about to be discussed. The doctor should then say
simply and plainly “I am afraid, I have some bad news. Your relative (name) is not doing
well. I know that you all are aware that he was admitted with breathlessness. We now know
that it is due to advanced pneumonia and this is serious illness.” It is enough at the first
interview to stop here and go to next phase (I of SPIKES). Clinicians should note that it is not
helpful to ‘lessen the impact’ of the bad news by saying ‘chest infection’ instead of ‘pneumonia’.
Most lay people interpret ‘chest infection’ as a minor illness against ‘pneumonia’, which they
see as a serious illness.
I–Invitation
Find out what they (the patient and/or the relatives) want to know. Most studies have suggested
that most will want to know everything about the disease and prognosis.9 However, some
patients (and relatives) will opt for minimal details about their disease and prognosis. It is also
important to find out at this stage if anyone else should be involved in the conversation
(in an Indian context, the family may wish an elderly relative, not necessarily the next of kin,
or another relative to be present and involved in these conversations). Stage ‘I’ is more
important in oncology where the patient may not want to ‘know the whole story in one go’
and the clinician can then opt for staged disclosure. However, in ICU and much more so in
ER, there may not be room for stage ‘I’ or it may be totally inappropriate. Nevertheless, stage
‘I’ should be remembered as often it is a useful ‘lead-on’ for conveying the actual bad news.
K–Imparting Knowledge
This phase deals with the actual giving of information that the doctor (and the nurse) has
come to give. It is essential to be precise, non-technical and clear. Doctors deal with technical
terms everyday and assume that their audience (the relatives and/or the patients) understand
medical terms. Terms like ‘tumour’ and ‘malignant’ mean very little to the lay public. It is
tempting to hide behind ‘platitudes’ and ‘aphorisms’ as it makes these conversations easy;
however, in communicating difficult news they are ultimately of little help. If death is imminent
or has already taken place, this must be stated in simple but sympathetic terms. It may be
useful to pause after using words like ‘death’ and ‘dying’ to allow the relatives to absorb this
profound and devastating news. If treatment is to be withheld or withdrawn, it is important
to be specific and stick to the point. Explain that withdrawal of support/treatment does not
mean withdrawal of care. Relatives are often anxious about the pain the patient may be
feeling; assure them that you are taking care of it (or will take care of it, if not already doing
so). If the patient is brain dead and brain stem function tests are to be carried out to confirm
this, it is the authors’ practice to say ‘we are going to assess brain function’ rather than to say
‘we are going to do brain stem tests’ as in most lay peoples’ view, a ‘test’ as a prelude to
further treatment. It is also important to approach the subject of withdrawal of active support
with tact and care. Doctors will do well to remember that asking relatives to ‘give permission
to switch off the ventilator’ is inappropriate as the relatives usually do not feel capable of
agreeing to that and they will feel guilty of ‘I killed him.’ A better way is to suggest that
treatment is of no further value, it is not doing anything beneficial, is inappropriate and doctor
believes it should be withdrawn. Be empathetic (see below). One way of putting it is ‘continuing
support is not going to make him live but prolong death’. This approach (which is true under
the circumstances) usually has the desired effect and relatives come round to accepting
withdrawal of support.
Be prepared for their reactions (anger, grief, tears etc). Do not leave the relatives or the
patient at this time, they need your support as they will have questions to ask and doubts to
voice. Answer the questions as truthfully as you can. If the relatives (patient) become too
emotional, offer to come back in a short while, if possible. Look at the relatives you are
talking to, engage their gaze and do not stare at empty space or into the distance through a
window.
E–Empathy
Relatives (patients) can display a variety of emotions when faced with bad news. They may
remain silent or may speak together at the same time or cry and become angry or even
demonstrate hostile and threatening behaviour. If crying, offer a tissue paper; it says that it
is OK to cry. In some cultures, especially in the West, it is OK to touch the patient and/or the
relatives as a way of comforting them; however, in India, this may not be viewed as appropriate.
The nurse, if female, may well do so to a female relative, and that will almost always be
acceptable. If the relatives say, ‘I am frightened’, don’t say ‘I know’. You, the doctor, can
hardly know what or how the relatives are feeling. Be empathetic rather than sympathetic.
Doctors should try and settle the patient’s (relatives’) emotions as it is difficult to go on with
the conversation until the emotion is cleared. At times this may prove difficult and the interview
may have to be completed later (if there is time).
S–Summary and Forward Planning

Summarise what you have said in a few sentences. Relatives should be offered the plan of
further management, even if it is just to keep the patient comfortable. Detail what is going to
happen next, over next few hours or days. If there has been an emotional outburst, there
may be lingering doubts, questions and still more emotions to deal with.
ABCDE PROTOCOL16
This is rather less well known but the protocol intends to help in the same way as SPIKES;
it is also easy for the intensivists and emergency physicians to remember. The points listed
here are a repetition (albeit a summarised version) of what has been described, in detail,
above under SPIKES.
• Advanced preparation. Arrange a private room for interview; prepare yourself with facts,
read the case notes, especially the diagnosis and progress; mentally rehearse what you are
going to say.
• Build therapeutic relationship: Introduce yourself (often a hand shake); warn that you are
about to give bad or sad news; find out what and how much the patient and/or the relatives
want to know? Do the family want other people present?
• Communicate well.
• Deal with patient’s/family’s reactions.
• Encourage/validate emotions.
Having detailed the available protocols, it is important to remember that not all
conversations in ICU and ER will follow this ‘structured’ pattern. Peoples emotions vary as do
their reactions to sad and bad news. Thus, although these structured approaches are a good
guide as to how these difficult interviews should be conducted, it may not always be possible
to keep to these protocols and follow the guidelines. This is due to the fact that the doctor
or HCWs may not always be able to plan such interviews because of little or no time to
rehearse. Moreover, interview may not focus on one single piece of information (patient’s
deteriorating condition or death) or involvement of many relatives (especially in Indian context)
may mean different emotions, opinions and questions are tackled at the same time.17 There
is thus ongoing debate about the strategies that should be developed to address these kinds
of situations.12,17
DO’s AND DON’Ts: SOME PRACTICAL POINTS6,13,18,19

The following points are the result of literature review as well as incorporating authors’ own
experience. Many of these points that have been gleaned from years of experience of delivering
bad and difficult news and have proven of great value over many years of clinical practice.
They are not listed in any order of importance:
• Choose a private room, if at all possible or speak to the relatives by the bedside (in certain
circumstances) but make sure that no one else is listening.
• Do take time in familiarising with the patient (or deceased) before you go into the interview
room.
• Switch off your pager, mobile phone, blackberry (or give them to some one else).
• Identify yourself and your position and check who you are speaking to (wrong sets of
relatives have been interviewed inadvertently).
• Do take a nurse with you (preferably the nurse taking care of the patient).
• Look the patient and/or relatives in the eye when talking to them.
• Be precise, non-technical but factual.
• Don’t conduct an interview in the corridor, if you can avoid it.
• If speaking to relatives in a corridor, stand with one shoulder leaning against the wall,
unbutton your white coat (if wearing one) and let your arms hang loose. This conveys an
impression of unhurriedness and attentiveness to the relatives.
• Give the news, as you know it. Don’t speculate and if you do not know the answer to a
difficult question, say so. Offer to find someone else who does know.
• Listen to the patient/relatives; they are trying to absorb the news, analyse it, think of all
the inevitable consequences and find answers to their sudden dilemma.
• Do break your giving of information into ‘chunks’ with quiet periods in-between. This
allows patients/relatives time to digest the information and formulate questions to ask and
overcome their emotions.
• Be prepared for reactions: anger, emotional upset, tears and the feeling of helplessness
and distress. A nurse to support the relatives (and indirectly you) is invaluable here.
• Be empathetic rather than sympathetic. Say ‘I can only imagine how you are feeling’
rather than ‘I know how you are feeling’. You do not know how the patients (or the relatives)
are feeling.
• Do not give bad news over the telephone. Such news should be given in person and in
a ‘face to face’ interview.
• If a corridor interview is inevitable (a common occurrence in India), do not stand straight
upright with arms folded like a general about to order his troops to advance
• Do not point your stethoscope, if carrying one, at the relatives nor shake it to emphasise
a point.
• Do not abandon an upset patient or a distressed relative. Stay with them; ask if you can
do anything for them at that moment e.g., ‘Can I call someone (relative, a friend or a
priest) for you?’
• Whatever the circumstances, never start a conversation with sentences like ‘This won’t
take long’.
• Use the ‘D’ (death) word where appropriate. Doctors usually hesitate to use this word.
• Offer to repeat the information as human mind tends to block out ‘bad news’ and relatives
will forget the things they have been told.
• Be prepared to talk to a relative who may have absolutely no idea what is going on with
the patient in ER. Quite often, there will have to be a ‘piecemeal’ communication of bad
news (remember P for Perception in the SPIKES protocol). A simple sentence may be said
to impart a ‘warning shot’—implying that bad news is coming (for the relative). This helps
to gauge reaction and plan the next step in the interview.
SPECIAL SITUATIONS
Failed or futile resuscitation in the emergency room or ICU is a sudden and catastrophic
event for which the relatives are never prepared. Special skills and empathy is required to
handle relatives, give difficult news and support the grieving in this situation.
Brain stem death in ICU is yet another special and unique situation. The lay public have
little or no concept of what brain stem death means or why a patient who is declared dead
medically, has a beating heart and feels normal. Patience, empathy, skill and supportive
colleagues (appropriate nurses) make communicating news of brain death easier. Brain stem
death is a prelude to a request for organ donation. This subject must be approached with tact
and sensitivity. The very circumstances surrounding such a request make it extremely difficult
for doctors to initiate discussion about organ donation. Social, religious and racial barriers
also apply and must be kept in mind at the time of the request. In the senior author’s experience,
right, timely and sensitive approach results in 90% of such requests being accepted by the
relatives.
Special Considerations in Indian Context

Until recently, a paternalistic attitude by treating physician has been the norm in India. However,
with improving standards of living and global spread of information (sometimes inadequate
or wrong information). Indian patients and their relatives have almost same expectations as
in the West. The public are also becoming aware of the fact that they have a right to information
and Indian newspapers are exploring these issues openly, as they should.20 However, unlike
in the West, Indian patient (and the relative) can often be demanding and expect a lot more
than medicine can presently offer, probably because in most cases they have to pay for the
treatment from their savings and expect to be able to buy a ‘cure’ with money. Many other
social, cultural and religious issues are intermixed with language, class and regional factors.
In India, illiteracy is a big barrier in understanding the diagnosis and prognosis of a disease.
The patient and relatives may not understand what is being said or the implication of the
information given. For the majority in India, ‘the doctor diagnoses and cures’; fallibility of
modern medicine is something a lot of people find difficult to comprehend.21 There is also
focus on collective decision-making and therefore, collective reception of bad news.21 The
Indian socio-cultural atmosphere does not put much emphasis on patient confidentiality and
autonomy, which is paramount in the West. Thus, information about a patient is freely shared
among many relatives and friends. The physician may also be requested ‘not to tell the patient
the truth’.22 However, the Indian clinical community is realising the importance of giving bad
news in a structured way23 and also the need for training in gaining these skills.24
CONCLUSION
Doctors and HCWs find it difficult to give bad, sad and difficult news. Poor training, social
upbringing, self-preservation instincts are some of the factors responsible for these deficiencies.
Doctors also feel stressed while breaking difficult news due to a lack of formal training in
communication skills. Programmes aimed at improving standards and communication of bad
news have now become established in the medical school curricula in the West, although
much remains to be done. Training of junior doctors and nurses in communicating bad news
is a priority. It has to be emphasised that attending courses for communicating bad and difficult
news (and there are many available in the West) alone does not impart these skills. Listening
to and learning from experts and senior colleagues and observing them at interviews is probably
the best way to enhance communication skills.
In ER and ICU, often there is little time to get to know the relatives and have time for
repeated conversations to get the difficult news across. Medical staff need to be empathetic
and sensitive when approaching relatives with bad news. SPIKES method and ABCDE
approach can make this task easier. However, robust research is needed to evaluate the impact
of guidelines and protocols adapted from oncology on patients’ or relatives’ perception of
receiving bad news in ICU and ER.
REFERENCES
1. Treasure T. Foreword to Emergency admissions: a journey in the right direction? Report of the National
Confidential Enquiry into Patient Outcome and Death (NCEPOD). 2007, pp 10. Available at http://
www.ncepod.org.uk/2007ea.htm
2. Matthews H. Ombudsman calls for better communication. Br Med J 1998;317:1472.

3. Buckman R. Breaking bad news: why is it still so difficult? Br Med J 1984;288:1597-9.
4. Ptacek JT, Eberhardt TL. Breaking bad news: a review of the literature. JAMA 1996;276:498-501.
5. Fallowfield L, Jenkins V. Communicating sad, bad and difficult news in medicine. Lancet 2004; 363:312-
9.
6. Baile WF, Buckman R, Lenzi R, et al. SPIKES–a six-step protocol for delivering bad news: application to
the patient with cancer. Oncologist 2000;5:302-11.
7. Berry PA. The withholding of truth when counselling relatives of the critically ill: a rational defence. Clin
Ethics 2008;3:42-5.
8. Tuckett AG. Truth-telling in clinical practice and the arguments for and against: a review of the literature.
Nurs Ethics 2004;11:500-13.
9. Hegarty RG, Butow PN, Ellis PM, et al. Communicating with realism and hope: incurable cancer patients’
views on the disclosure of prognosis. J Clin Oncol 2005;23:1278-88.
10. Mueller PS. Breaking bad news to patients: the SPIKES approach can make this difficult task easier. Postgrad
Med 2002;112:15-6.
11. Barnett MM, Fisher JD, Cooke H, et al. Breaking bad news: consultants’ experience, previous education
and views on educational format and timing. Med Edu 2007;41:947-56.
12. Wittenberg-Lyles EM, Goldsmith J, Sanchez-Reilly S, et al. Communicating a terminal prognosis in a
palliative care setting: deficiencies in current communication training protocols. Soc Science Med
2008;66:2356-65.
13. Mir NU. Breaking bad news: practical advice for busy doctor. Hosp Med 2004;65:613-5.
14. Back AL, Curtis JR. Communicating bad news. West J Med 2002;176:177-80.
15. Minichiello TA, Ling D, Ucci DK. Breaking bad news: a practical approach for the hospitalist. J Hosp Med
2007;2:415-21.
16. Rabow MW, McPhee SJ. Beyond breaking bad news: how to help patients who suffer. West J Med
1999;171:260-63.
17. Eggly S, Penner L, Albrecht AL, et al. Discussing bad news in the outpatient oncology clinic: rethinking
current communication guidelines. J Clin Oncol 2006;24:716-19.
18. Harrahill M. Giving bad news gracefully. J Emerg Nurs 2005;31:312-4.
19. Mehra MR, Uber PA, Ventura HO. Death messengers. Br Med J 2007;335:1296-7.
20. Chaudhuri S. Communicating with the terminally ill. Times of India 1999.
21. Shubha R. End-of-life care in Indian context: the need for cultural sensitivity. Indian J Pal Care 2007;13:50-
64.
22. Mohandas KM. Ethical dilemmas in breaking bad news. Indian J Med Ethics 1995;3:4.
23. De U, Bhattacharya P. Communicating bad news: an approach. Indian J Surg 2003;65:286-9.
24. Govindan VK. Enhancing communication skills using OSCE and peer review. Med Edu 2008; 42:535-6.
Inter-professional Communication and Teamwork 21
Three
Inter-professional
Communication and Teamwork
S Srinivas
INTRODUCTION
Communication essentially consists of transmitting information from one person to another.
Many scholars and experts use the Lasswell’s maxim “who says what to whom, in what way
and to what effect” to describe communication. Communication can also be viewed as an
exchange of understanding.
In the field of modern medicine, development in drugs and technology receive close
attention while non-technical skills are relatively neglected. Communication plays a crucial
role in every aspect of care, right from asking the patient about his/her history to following
up an abnormal result from the laboratory. Within the ICU, communication between various
professionals—medical, paramedical, nursing and support, forms the bedrock upon which
good teamwork is built. Good inter-professional communications between doctors and nurses
is essential not only for efficient functioning of the ICU, but also enhanced safety and quality
of care as well. Communication between consultants of various specialties involved in the
care of an ICU patient is, likewise, very valuable. Sadly, skills necessary to improve teamwork
are rarely emphasised or formally targeted in the training curricula of intensive care programmes
all over the world. This is in sharp contrast to high-reliability industrial organizations where
the importance of teamwork and good communication skills is recognised as important and
methods to assess competency within these domains have been developed.1
COMMUNICATION RESEARCH IN ACUTE CARE AREAS

Research has examined the relationship between communication gaps and error in acute
care environments in a hospital. Williams et al2 examined 328 separate incident reports where
poor communication was deemed responsible for errors and found numerous underlying
problems. Most of these related to ineffective delegation, poor role clarity, shift changes, loss
of information on handover, staff not attending rounds, hierarchical team structures and
inaccurate assumptions about skills of team members. In an observational study on surgical
cases,3 poor communication and information flow was found to have a negative effect on
team performance as a whole. In total, 88 distinct events were identified where crucial
information was either lost or degraded, where as a result, >85% of events lead to adverse
consequences for patient care. The researchers identified patient handover and the movement
of patients from one phase of care to the next as the most vulnerable for information loss.
Inadequate discussion of clinical information was cited as a commonly occurring failure. More
recently, a review of surgical malpractice claims identified insufficient verbal communication
between attending surgeons and other team members as a problem.4 Once again, patient
transfers and handover times were found to be susceptible to information loss. Thus, studies
in operating theatres have consistently highlighted the need for better transfer of information
in reducing the probability of error. Although relevant to the operating theatre environment,
there are important lessons in these studies for those working in the ICU.
There is unfortunately a paucity of good quality data from the ICU. Although, Cook et al5
assert that gaps and discontinuities are responsible for adverse events as a result of information
loss in the ICU, evidence available to support the claim is scant. Findings from the sentinel
study by Donchin et al6 are, in fact the only ones to lend support to this view. In this large
study done in an ICU, communication problems were found to be at the core of a large
number of errors, but the nature or impact of these errors was not clearly articulated.
INEFFECTIVE INFORMATION TRANSFER

ICU patients have complex life threatening illnesses, some of which take weeks to resolve.
Almost all patients are monitored on a 24-hour basis necessitating shift work and with it, the
need to pass on patient care responsibilities during a change of shift. Information gaps do
develop during handovers from doctor to doctor and from nurse to nurse.7 In addition, for
patients with evolving pathology for those who stay in the ICU for a long time, care plans
made during an earlier stage are often forgotten or simply not incorporated into ongoing
plans. Factors that contribute to poor information flow in the ICU are summarised in Table
3.1. Among these, ambient noise levels in the ICU, faulty design of human-computer interface,
frequency of interruptions and other environmental influences in the workplace are also
included. While these factors come into play
throughout the patient’s stay in the ICU, they Table 3.1: Factors precluding effective transfer
of information
become critical during handovers or transfer of
care to another physician or following transfer • Prolonged admission
to another facility. • Complex life threatening admission
• Information overload
• Rapidly changing clinical problems
Handover • Ambient noise
• Poor human-computer interface
Published data from both the UK and Australia • Shift system requiring handover of information
reveals that handover of patient care is • Cross coverage
suboptimal and the process is not formalised.8,9 • Language and cultural factors
Moreover, there is lack of advice and guidance • Hierarchical and social factors
Table 3.2: Comparison of bedside and office handover
Bedside Office
Bedside visual cues Used and improve handover quality Not available
Ability to check charts Information can be cross-checked Not possible
Interaction with nurses Available and useful Possible
Process Less structured More formal
Confidentiality Maybe compromised Preserved
Staff debriefing Minimal Significant
Interruptions Likely Possible
Preference Preferred by nurses Suitable for doctors
on how best to achieve an effective handover. Only one published study so far, has explored
the effectiveness of bedside handover compared to office handover by nurses in five acute
care settings. While no method was shown to be clearly superior, the study identified potential
strengths and weaknesses of both types of handover (Table 3.2).
Handovers can also be viewed according to the modality used to facilitate transfer of
information. Verbal handovers are popular with ICU doctors as they are time efficient and
expedient. However, there is potential for disintegration and distortion of information with
verbal handovers. Although published data are not available to evaluate the effectiveness of
one technique over another, there appears to be a consensus that verbal handovers are
probably the least effective of all processes.10
Norgaard et al11 used a checklist to assess trainee performance during handover of informa-
tion and concluded that the more junior the trainee, the less effective was the information
transfer. Lee et al12 introduced a computerised sign out card for doctors to handover
information and demonstrated a reduction in adverse event rate from 14.9 to 5.8%. A similar
pattern emerged from studies done among ICU nurses.13
Cross Coverage
Petersen14 studied the relationship between house-staff coverage schedules and the occurrence
of preventable adverse events over a 4-month period. On multi-variate analysis, they found
cross-coverage by a doctor from another team was strongly associated with a higher incidence
of adverse events. This study identified the role of cross coverage and, as a corollary inadequate
information transfer as important in the causation of adverse incidents.
Interruptions
Interruptions are common during work in the ICU. In an observational pilot study conducted
in a single ICU over four weeks, clinicians were observed for a total of nearly 30 hours during
rounds.15 The average number of interruptions during morning handover and during the
main ward rounds were as high as 32 per hour. Only 22% of these were essential interruptions
but many lead to a break in task or diversion of attention of the clinician. Similar data has
been also reported from an observational study conducted in an emergency department;
leading Chisholm et al16 to conclude that critical care physicians work in an “interrupt-driven”
environment which can potentially drive clinicians to error. Workplace interruptions may be
an important ergonomic factor contributing to inefficiency and stress. A high burden is placed
on short-term memory, as interruptions tend to disrupt items stored in working memory. It is
conceivable that a workplace peppered with interruptions could cause individuals to forget
important tasks.17 In the care of critically ill patients, these omissions could have dire
consequences.
INTER-PROFESSIONAL COMMUNICATION
Communication between different groups of professionals seems to have a positive, synergistic
effect on patient care. Lingard et al18 attempted to lay down the rules of this game of inter-
professional collaboration. Their analysis explored how team members in the ICU interact to
achieve daily clinical goals, delineate professional boundaries, and negotiate complex system
issues. Participants in their study consisted of a sample of four nursing focus groups, two
resident groups, and one intensivist group. Results reported by them demonstrate that forces
of ownership and trade have a central role in the daily negotiations that constitute teamwork
in the ICU. Ignoring these issues leads to accumulation of tension and sluggishness of
collaboration.
Friedman et al19 viewed improving team structure and communication as a key to improving
hospital efficiency. Their study population comprised of patients admitted to their hospital’s
general surgery wards. At inception of the restructuring programme, responsibility of each
team member was well defined. Emphasis was laid on improving regular communication and
collaboration between all team members. A formalized schedule of meetings was designed to
facilitate this collaboration. In the end, this study showed that such a programme reduced
length of stay in hospital, fewer inpatient days, improved surgical volumes and better training
for residents.
Reader et al20 reported the findings of a cross-sectional survey of ICU nurses and doctors
using a previously established measure of ICU inter-disciplinary collaboration. This study
examined whether individuals who report higher levels of open communication within the
ICU also report having a better understanding of their patient care goals. The importance of
leadership of senior ICU staff in fostering a perception of communication openness was also
reported. This study confirmed a disparity in the perception of inter-disciplinary communication
between doctors and nurses. This difference extended to communication between senior
doctors and trainees. The authors identified several factors as being responsible for such a
difference in perceptions. These included hierarchal factors, gender, differing patient care
responsibilities, differing perceptions of requisite communication standards and differences in
the training methods of nurses and doctors. Open communication facilitated the understanding
of core goals by the junior team members. This study also found that unit leadership is an
important determinant of open communication.
TEAMWORK
Highly educated, experienced and motivated clinicians work and interact everyday in the
complex environment of modern medicine. Despite their best intentions, the hospital
environment creates a situation where errors arise, not as a result of technical incompetence
but as a result of unclear communication, clashing priorities and poor inter-personal skills.
Teamwork and communication deficiencies have been demonstrated in trauma care, ICU
and operating theatres. In fact, the landmark report of the Institute of Medicine21 identified
areas such as critical care, emergency room, and wards caring for patients at extremes of age,
as being vulnerable to errors.
The study of teamwork in medicine is a relatively new field, where efforts are being made
to understand the role of skills such as co-ordination, communication, leadership, situational
awareness, conflict resolution and vigilance in preventing adverse events. Pronovost and
colleagues report22 on web-based patient safety reporting system identified team factors as
being responsible for nearly a third of critical incidents. Overall, 57% of these errors were
related to problems with verbal and written communication during routine care while 37%
were related to communication issues during handover. Thus, communication issues during
routine care are as important, if not more important than handover.
In recent years, high fidelity simulators have been used to investigate communication skills
that are most likely to result in effective team performance. For example, Lighthall et al 23
found during analysis of critical events in the ICU that team members did not communicate
their care priorities to one another, that doctors overloaded nurses with requests and more
importantly, that ineffective leadership resulted in inefficient use of time and personnel. Studies
have also looked at communication abilities of ICU residents during resuscitation of critically
ill patients.24 Video recordings of simulated scenarios were analysed by experts. Overall, the
study found residents with three years or more of experience scored higher on communication
ability compared to those into their first year of postgraduate training. Ottestad et al 25 also
developed a scoring system for assessing communication and teamwork of critical care teams.
In particular, teams were assessed during management of septic shock in a simulated
environment and the relationship between teamwork and technical performance was studied.
Teams were rated high, if they made clear and direct requests, employed close loop
communication, delegated tasks effectively, prioritised aspects of care and shared information
on treatment plans. In this study, a close correlation between these ratings and clinical
competence (diagnosis, antibiotic use, etc.) as well as behavioural aspects of performance
was noted. In other words, these early studies have all indicated an improvement in
performance in teamwork with better interaction and communication between individual
members.
This is not unexpected, as the same effect has been demonstrated in complex fields of
operation like aviation and defence. However, in contradistinction to standards in the industry,
medical teams often perform a number of critical tasks under conditions of high workload
and with a constant change in the membership of the team.26 Within these teams, separate
hierarchies of power, competence and position add an additional layer of complexity. Moreover,
medical teams face a rapidly evolving situation to which they must adapt in order to achieve
their goal. Teamwork skills are therefore even more critical in these dynamic situations.
Pronovost et al22 have therefore stressed the importance of implementing team training
programmes and team based activity that encourages inter-disciplinary communication during
patient decision making. Central to effective teamwork is communicating in a manner that
allows for the development of a shared mental model about the situation at hand and the
ability for this to evolve, as clinical circumstances change.26
INTERACTION BETWEEN DOCTORS AND NURSES

Interaction between nurses and doctors during a typical ward round determine the way an
ICU patient is managed for the rest of the day. However, such interactions are often one-
sided with seniority and hierarchy playing a crucial and sometimes, a deterrent role.
Manias et al27 studied the pattern of participation of critical care nurses in ICU rounds.
Marginalisation of inputs from nurses during these rounds was a crucial finding identified by
Manias et al.27 The general feeling was that consultants/ senior medical staff tends to take
opinion of nursing staff only on housekeeping issues rather than on patient care issues.
Considering the fact that bed-side nurses spend more time with the ICU patient than the
residents or registrars, their inputs are as valuable as those of the junior medical staff, if not
more. Manias et al27 also identified that, failure to recognize the bed space as a place where
nursing care issues take precedence, also led to poor communication between the doctors
and nurses.
The issue of healthy work environments, nurse-physician communication and patient
outcomes were analysed by Manojlovich et al.28 Several studies have reported verbal
miscommunication between nurses and physicians as being responsible for a number of errors
in the ICU. This study gathered inputs solely from nurses and examined the relationship
between nurse-physician communication and selected outcomes sensitive to nursing care.
The study looked at nosocomial infections – Ventilator Associated Pneumonia and Catheter
related sepsis as well as nursing report on medication errors. The authors found that
empowerment at the workplace significantly and positively correlated with communication
scale. Nurse associated rates of VAP, catheter associated sepsis and medication errors were
influenced more by communication factors rather than by environment factors. The authors
suggested that providing nurses with more information, support, resources and opportunities
at work could improve communication with physicians.
Narasimhan et al29 evaluated the effect of a standardised worksheet on physicians and

nurses perceptions of their goals of care and on patient’s length of stay in an ICU. The authors
designed a simple worksheet as a template with spaces for the team to fill in the plan during
morning rounds. This worksheet was reviewed during evening rounds, a step that made it
easy to review the day’s progress. This goal directed worksheet was perceived to improve
communications between physicians and nurses.
Parbury et al30 attempted to identify causes of breakdown in the collaboration between
nurses and physicians using a model of types of knowledge. The type of knowledge used in
clinical work, was classified as case knowledge, patient knowledge and person knowledge.
Case knowledge was the domain of scientifically established knowledge of anatomy, physiology,
pathophysiology, genetics, disease processes and therapeutics. Such knowledge was largely
the province of physicians. However, nurses observe the changes in response to disease and
act as the eyes and ears of physicians. Patient knowledge involved understanding a particular
patient’s experience of disease and response to treatment. Nursing work normally focuses on
this knowledge, which in reality requires proximity to patients over extended periods of time.
Nurses, being present at the bedside more often than doctors, are able to make comparisons
and interpret responses. They are, therefore able to alert physicians about atypical response
to a particular intervention. Breakdown of such collaboration occurs when physicians disregard
nurses’ clinical assessment and concerns about a patient, as these contributions do not figure
in case knowledge. A better understanding and acknowledgement of each others’ areas of
knowledge helps in better collaboration between the two sets of professionals involved directly
in patient care.
Rafferty et al31 studied whether team work and professional autonomy are compatible
and whether such compatibility results in improved patient care. They evaluated the response
of more than ten thousand staff nurses in 32 hospitals in England, to a questionnaire about
professional autonomy and teamwork. Results of this study re-emphasised the value of
teamwork and its association with a range of positive occupational and organizational attributes.
The study also found a strong positive association between autonomy and teamwork by
emphasising that the interaction between autonomy and teamwork was of synergy rather
than that of conflict.
CALLING FOR HELP

The National Institute for Health and Clinical Excellence (NICE) in the UK issued detailed
guidelines for recognition of and response to acute illness among adults admitted to hospital.32
This seems to be the most comprehensive set of guidelines to date which deal with the issue
of “calling for help”. The NICE guidelines attempt to stratify patients needing help into different
groups. This stratification has been drawn to clarify several issues. First, the subset of patients
who need to be attended to immediately compared to less emergent situations is sought to
be identified. Next, the seniority of the person who attends to the sickest patient is also identified.
Lastly, the destination of the patients needing any form of help is described in the NICE
guidelines. The issue of parameters to be monitored and recorded also figures prominently
in the guidelines. Most of the criteria mentioned in the guidelines include vital signs monitored
in several other early warning systems discussed in an article in an earlier edition of Critical
Care Update.33 The NICE guidelines recommend the use of Track and Trigger systems to
identify patients at risk. The frequency with which the observations need to be recorded in
patients with varying degrees of morbidity is also addressed in the NICE guidelines.
CONCLUSION
Successful teamwork, including good communication lays essential foundations for effective
care in dynamic and high stakes environments of Emergency and Critical Care Medicine.
There is an urgent need to carefully identify and delineate team behaviours that are necessary
for efficient functioning in the modern-day healthcare system, particularly those that are essential
during a crisis situation. Only then, activities associated with teamwork in the ICU will achieve
a desirable standard of consistency and safety.
REFERENCES
1. Baker DP, Day R, Salas E. Teamwork as an essential component of high reliability organizations. Health
Serv Res 2006;41:1576-98.
2. Williams R, Silverman R, Schwind C, et al. Surgeon information transfer and communication: factors
affecting quality and efficiency of inpatient care. Ann Surg 2007;245:159-71.
3. Christensen C, Gustafson S, Roth EM, et al. A prospective study of patient safety in the operating room.
Surgery 2006;139:159-73.
4. Greenberg C, Regenbogen S, Studdert D, et al. Patterns of communication breakdown resulting in injury
to surgical patients. J Am Coll Surg 2007;204:533-40.
5. Cook RJ, Render M, Woods DD. Gaps in continuity of care and progress on patient safety BMJ
2000;320:791-4.
6. Donchin Y, Gopher D, Olin M, et al. A look into the nature and causes of human error in the intensive
care unit. Crit Care med 1995;23:294-300.
7. Venkatesh B, Miller A, Karnik A. Information Exchange in intensive care: how can we improve. Chapter
in Vincent JL (ed) Yearbook of Intensive Care and Emergency Medicine. Springer, Berlin 2006.
8. Roughton VJ, Severs MP. The junior doctors handover: current practices and future expectations. J R Coll
Physicians Lond 1996;30:213-4.
9. Bomba DT, Prakash R. A description of handover processes in an Australian public hospital. Aust Health
Rev 2005;29:68-79.
10. O’Connell B, Penney W. Challenging the handover ritual. Recommendations for research and practice.
Collegian 2001;8:14-8.
11. Norgaard k, Ringsted C, Dolmans D. Validation of a check list to assess ward round performance in
internal medicine. Med Edu 2004;38:700-7.
12. Lee LH, Levine JA, Schultz HJ. Utility of a standardised sign-out card for new medical interns. J Gen
Intern Med 1996;11:753-5.
13. Balas MC, Scott LD, Rogers AE. The prevalence and nature of errors and near errors reported by hospital
staff nurses. Appl Nurs Res 2004;17:224-30.
14. Petersen LA, Brennan TA, O’Neill AC, Cook EF, Lee TH. Does house staff discontinuity of care increase
the risk of preventable adverse events? Ann Intern Med 1994;121:866-72.
15. Nimmo G, Mitchell C. A preliminary audit of interruptions in intensive care: implications for patient safety.
J Intensive Care Soc 2008;9:240-2.
16. Chisholm C, Collison E, Nelson D, et al. Emergency department workplace interruptions: are emergency
physicians “interrupt-driven” and “multi-tasking” Acad Emerg Med 2001;8:686-8.
17. Alvarez G, Coiera E. Interruptive communication patterns in the intensive care unit ward rounds. Int J
Med Inform 2005;74:791-6.
18. Lingard L, Espin S, Evans C, et al. The rules of the game: Inter-professional collaboration on the Intensive
Care Unit team. Crit Care 2004;8:403-8.
19. Friedman DM, Berger DL. Improving team structure and communication – a key to hospital efficiency.
Arch Surg 2004, 139:1194-8.
20. Reader T W, Flin R, Meams K, et al. Interdisciplinary communication in the Intensive Care Unit. Br J
Anaesth 2007;98:347-52.
21. Kohn LT, Corrigan J, Donaldson MS. To err is human: building a safer health system. Washington, DC:
National Academy Press, 2000.
22. Pronovost P, Thompson D, Holzmueller C, et al. Toward learning from patient safety reporting systems.
J Crit Care 2006;21:305-15.
23. Lighthall GK, Barr J, Howard SK, et al. Use of a fully simulated intensive care unit environment for critical
event management training for internal medicine residents. Crit Care Med 2003;31:2437-43.
24. Kim J, Neilipovitz D, Cardinal P, et al. A pilot study using high-fidelity simulation to formally evaluate
performance in the resuscitation of critically ill patients: the Crisis Resource Management I Study. Crit
Care Med 2006;34:2167-74.
25. Ottestad E, Boulet J, Lighthall G. Evaluating the management of septic shock using patient simulation.
Crit Care Med 2007;35:769-75.
26. Eppich WJ, Brannen M, Hunt EA. Team training for emergency and critical care pediatrics. Curr Opin
Pediatr 2008;20:255-60.
27. Manias E. Nurse-Doctor interactions during critical care ward rounds. J Clin Nurs 2001;10(4):442-50.
28. Manojlovich M, De Cicio B. Healthy work environments, Nurse Physician communication and patient
outcomes. Am J Crit Care 2007;16:536-43.
29. Narasimhan M, Eisen LA, Mahoney CD, et al. Improving Nurse Physician Communication and satisfaction
in the Intensive Care Unit with a daily goals worksheet. Am J Crit Care 2006;15:217-22.
30. Stein-Parbury J, Liaschenko J. Understanding collaboration between nurses and physicians as knowledge
at work. Am J Crit Care 2007;16:470-8.
31. Rafferty AM, Ball J, Aiken LH. Are teamwork and professional autonomy compatible and do they result
in improved hospital care? Qual Health Care 2001;10 (suppl II) ii32-ii 37.
32. NICE clinical guideline 50-www.nice.org.uk/CG050.
33. Williams O. Early warning signs of impending critical illness-implications for the quality of care. In Nayyar
V (ed) Critical Care Update. Jaypee Brothers Medical Publishers, New Delhi 2003.
Section B: Cardiovascular Problems in ICU
Four
What should an Intensivist Know

about Pacemakers?
Bhim Shankar PR, Ajit Thachil, Hygriv Rao, C Narasimhan
INTRODUCTION
In recent times, the number of patients being admitted to the hospital with a pacemaker
implanted has increased significantly. As many of these patients have other medical problems,
Critical Care specialists are increasingly involved in their management. Although implantation
and management of pacemakers falls under the purview of the Electrophysiologist, it is essential
for others involved in the care of the patient to be conversant with issues that impact on the
overall management of the patient.
This review addresses specific issues that may arise while dealing with a patient who has
a pacemaker in situ. The attempt is to provide the reader with an understanding that will
enable recognition of the capability and problems associated with an implanted pacemakers.
For a more fundamental understanding of issues related to pacing in general, pacing modalities,
nomenclature and indications for use, the reader is referred to a standard textbook of
Cardiology. This chapter, is organised around a series of clinical questions listed below.
a. Does my patient need temporary pacing?
b. Is pacemaker the reason for patient’s admission to the ICU?
c. Is a normally functioning pacemaker the reason for admission?
d. Can the pacemaker contribute to the diagnosis or treatment?
e. Will the pacemaker interfere with ICU procedures?
DOES MY PATIENT NEED TEMPORARY PACING

The insertion and management of a temporary transvenous pacemaker can be a life-saving
procedure in an emergency. Any patient presenting with symptomatic bradycardia either due
to sinus node dysfunction or AV nodal disease will require urgent attention. Advanced AV
nodal disease and other medical conditions (Table 4.1) that present with severe bradycardia
/shock that will need emergent insertion of a temporary cardiac pacing lead to ascertaining
the cause and determining further course of management can be done later with inputs from
a specialist. Apart from primary abnormalities of cardiac automaticity and/ or conduction
What should an Intensivist Know about Pacemakers? 31
disturbances, other causes for bradycardia/ pause requiring emergency pacing include
conditions summarized in Table 4.1.
Table 4.1: Indications for temporary pacing in the absence of primary

conduction disturbances or abnormality of cardiac automaticity
Conditions Examples
Side effect/overdose of cardiac drugs Beta blockers

Ca channel antagonists
Digoxin or digoxin like substances
Acute Myocardial Infarction ** Mobitz II second-degree AV block
Bilateral bundle branch block
Poisoning with cardio-active agents Poisons i.e. yellow oleander
Electrolyte imbalance Severe Hyperkalemia (K >6 mmol/l)
Cardiac surgery Ostium Primun repair
** See www.americanheart.org
For temporary pacing, either the superficial femoral vein or internal jugular vein is accessed.
In the ICU, where fluoroscopy is not readily available balloon-tipped pacing wires (inserted
via internal jugular approach) are preferred with placement confirmed by ECG or echo-
cardiographic guidance. However, if time permits, and if there is access to fluoroscopy, it is
recommended that the pacing wire be placed under fluoroscopic guidance. The ideal location
is in the apex of the right ventricle (RV), but it may be necessary to accept another location
if the lead position is stable there and the RV apex is not readily reachable. An X-ray is
needed to document and confirm the position of the lead following placement.
Is Temporary Pacing Programmed Appropriately

Besides placing the lead at a suitable location, appropriate programming of the temporary
pulse generator is essential. It is important that the intensivist is aware of basic parameters.
The pacing threshold is defined as the minimal current that is necessary to capture the ventricle.
It can be determined by finding the amplitude at which capture no longer occurs. Generally,
a threshold of 1v or less is ideal, but in temporary pacemakers higher thresholds of up to
2-2.5 v may be acceptable, as battery longevity is not so much an issue like in permanent
pacemakers. In most emergent situations, a temporary pacemaker is used in the ventricular
demand mode (the pacemaker detects the intrinsic heart rate and dose not pace unless the
intrinsic rate is less than the lower rate of the pacemaker). To determine the sensing threshold,
the rate of the pacemaker is set lower than the intrinsic rate of the patient. This step is not
necessary in patients who are dependent on pacing. The sensitivity of the pacemaker is set
in millivolts and is the voltage that must be sensed to inhibit the firing of the pacemaker. The
threshold must be checked twice a day by the ICU staff to detect any possible lead
dislodgement.
IS PACEMAKER THE REASON FOR THE PATIENT’S ADMISSION TO ICU

A patient with a permanent pacemaker must have a heart rate at the lower programmed rate
of the pacemaker. When the heart rates are less than this value it may indicate malfunctioning
of the device and these patients can present with syncope or dizziness. Malfunction of an
implanted device may contribute to a patient’s acute decompensation. This may take several
forms and can be grouped into four categories:
a. Failure to output/battery depletion,
b. Failure to capture,
c. Undersensing/Oversensing
d. Inappropriate pacemaker rate.
It may be necessary in every patient with a pacemaker admitted with any other medical
problems to have the pacemaker interrogated by the EP services of the hospital or by industry
representatives. It is necessary to know the company manufacturing the pacemaker as the
programmer differs from one manufacturer to another.
Battery Depletion
Pacemakers reprogramme automatically to slower rates, single-chamber pacing, and/or non-
physiologic (fixed rate) pacing modes on detecting that its battery is depleted beyond a
particular limit. All of these changes may diminish cardiac output. Once this is recognized, it
may be useful to either pace at a higher rate with a temporary pacing unit or to attempt AV
sequential pacing while arrangements are made to change the battery or the device. Intensive
care units should have temporary generators that can pace the atrium and ventricle sequentially.
Battery depletion can be recognized by doing an ECG with a magnet placed on the generator
site. Most pacemakers pace at a fixed rate of 100 beats per minute on being exposed to a
magnet in such a fashion (the so-called “magnet rate”). Battery depletion will cause the “magnet
rate’’ to drop.
Rarely, there can be a total loss of output, which can be catastrophic for the pacemaker-
dependent patient. When a small voltage decrement occurs, end-of-life characteristics (such
as a change in pacing rate) are triggered, signaling the need for pacemaker replacement.
Once these initial end-of-life changes appear, there is usually a period of several months before
the pacemaker battery reaches a critically low voltage and pacing fails. In other words, total
battery depletion results in failure to output, while failure to capture occurs at a lower level of
battery depletion.
Capture Failure
Capture failure is defined as the absence of ventricular depolarization when the pacemaker
delivers an output. The surface ECG shows the pacing artifacts but no corresponding QRS
Fig. 4.1: Intermittent pacing spikes (circled) not followed by ventricular complexes
complexes (Fig. 4.1). The most common cause of failure to capture is dislodgment of the
pacemaker lead from the endocardial surface; such dislodgment usually occurs in the first few
weeks after implantation. The newer designs for active and passive fixation leads have been
associated with a greatly decreased incidence of dislodgment.1,2 Although the acceptable rate
for lead dislodgment is difficult to define, it should be less than 2% for ventricular leads and
less than 5% for atrial leads. After the first few weeks, fibrosis at the electrode tip can cause
an exit block, again leading to capture failure. Metabolic derangements, such as hypo or
hyperkalaemia (by hyperpolarizing the cell membrane), severe acidosis, or antiarrhythmic
drugs (e.g. long term use of Amiodarone) may also cause failure to capture (Table 4.2). Failure
to capture leads to loss of pacing, with possible heart failure, presyncope, syncope or rarely
bradycardia–induced ventricular premature contractions or even ventricular tachycardia.
Table 4.2: Causes of capture failure
Lead related problems Dislodgment

Insulation break (rare)
Electrolyte abnormalities Hyperkalaemia
Hypokalaemia
Severe metabolic acidosis
Anti-arrhythmic drugs Amiodarone
Propafenone
The intensivist, apart from recognising potentially life threatening capture failure, needs to
ascertain the cause of the same. Reprogramming the generator to a higher output (which
usually needs the assistance of the EP/ industry personnel) or inserting a temporary pacemaker
lead will suffice in conditions where lead dislodgement or fibrosis appears to be the cause.
Urgent correction of metabolic parameters would be needed in other cases.
Abnormalities of Pacemaker Sensing

Sensing issues can be subclassified as being due to undersensing or oversensing. Oversensing
is defined as the unexpected sensing of an intracardiac or extracardiac signal.3,4 Oversensing
should be suspected, if there are inappropriate pauses in a paced patient (occasionally,
oversensing in a dual chamber system can also cause tachycardia, as explained later).
Oversensing induced pauses may be intermittent, resulting in irregularly delayed pacemaker
stimulation; or constant, leading to a decreased pacing rate or total inhibition of pacemaker
output. Electrical signals that may cause oversensing include myopotentials, T waves, and P
waves. Atrial-channel oversensing may occur when “far-field” R waves are sensed (i.e.
ventricular depolarization is sensed by the atrial lead, and interpreted as atrial depolarization).
In a dual-chamber pacemaker, the pacemaker stimulus in one chamber is sometimes sensed
in the other chamber (chamber cross-talk). This can lead to inhibition of pacing output (pauses
or bradycardia). However, oversensing by the atrial lead may sometimes lead to ventricular
pacing at the upper rate (see pacemaker-mediated tachycardia in next section). Oversensing
induced pauses are sinister and require urgent intervention.
Undersensing, as the word implies, is failure to sense. Undersensing is rarely a medical
emergency (Fig. 4.2) However it may contribute to decompensation of heart failure.
Undersensing may be caused by lead insulation break, fibrosis at the lead tip, injury to adjacent
myocardium, or improper programming of the device’s sensing parameters. Atrial undersensing
may result in atrial pacing too soon after spontaneous atrial activity (P waves). In susceptible
Fig. 4.2: The QRS complexes marked by the arrows are not sensed by the pacemaker, leading to
inappropriate delivery of pacing
patients, this may induce supraventricular tachycardia, atrial flutter, or atrial fibrillation. In an
analogous fashion, ventricular undersensing may induce ventricular tachycardia.
Inappropriate Pacemaker Rate

Patient with a pacemaker and fast heart rates invariably leads to a suspicion of ventricular
tachycardia or tracking of a rapid atrial arrhythmia. Recognition of these arrhythmias is
important, as prompt management is often needed. Pacemaker re-entrant tachycardia or
pacemaker-mediated tachycardia (PMT) are rare causes of tachycardia which occurs when
sensing of a retrograde atrial depolarization initiates ventricular pacing, which in turn leads to
retrograde conduction and repetition of the same cycle.5 PMT, although not a emergency, will
need reprogramming of the refractory periods for terminating it. “Runaway pacemaker” occurs
infrequently with current pulse generators. The problem used to result from battery failure,
random component failure, or component failure induced by therapeutic radiation. Treatment
aimed at disabling the faulty pacemaker output must be initiated promptly if there is
haemodynamic compromise. This can be rapidly achieved by applying a magnet on the
pacemaker pocket.
Lead Malfunctions
Pacemaker lead malfunction can also result in failure of the pacemaker to pace appropriately.
A fracture or an insulation defect can cause this, although the occurrence is low with new
technology incorporated into the latest generation of pacemakers. Lead fracture is diagnosed
by a marked increase in lead impedance and/or visible lead discontinuity on chest radiograph.
Fracture of a pacing lead may cause capture failure. Lead insulation break is diagnosed solely
by reduced lead impedance because there are no clues visible on a chest radiograph. It results
in two potential problems. First, during pacing, current may leak out the lead body, thereby
reducing the current that is delivered to the lead tip. This may result in capture failure. Second,
and more commonly, the insulation break allows the device to sense chest wall myopotentials
and lead to oversensing problems.
IS A NORMALLY FUNCTIONING PACEMAKER THE REASON FOR ADMISSION
Pacing-induced Heart Failure

Permanent pacing can sometimes be the cause of otherwise unexplained heart failure, atrial
fibrillation and atrioventricular regurgitation. Thackray et al6 observed that 27% of the patients
developed LV dysfunction and symptoms of heart failure after implantation of a permanent
pacemaker. Data from the UKPACE trial,7 in which 2021 patients with permanent pacemakers
were monitored over a three-year period for onset of heart failure, revealed an yearly incidence
close to 3.3%.
A properly timed atrial systole improves stroke volume by providing greater left ventricular
end-diastolic fibre stretch, and, consequently, enhanced end-systolic fibre shortening.8 However,
large randomized clinical trials of pacing in both sinus node dysfunction and atrio-ventricular
block have shown conflicting results in this regard. In spite of maintaining AV synchrony, dual
chamber (DDD/R) pacing, as compared to single-chamber ventricular pacing (VVI/R) provides
surprisingly modest or negligible benefit in preventing heart failure (Table 4.3).
Table 4.3: Summary of RCTs comparing onset of heart failure among

patients with AV synchronous or ventricular based pacing
Clinical trial No of patients Pacemakers Results
UKPACE trial7 2021 DDDR vs. VVI/R No diff

CTOPP study9 2568 DDDR vs. AAIR No diff
MOST study10 2010 DDDR vs. VVI/R DDDR better
AAIR – Atrial sensed, Atrial paced rate-adaptive pacing mode

DDD/R – Dual chamber sensed and paced mode, with or without rate adaptation to exertion
VVI/R – Right ventricular sensed and paced pacing mode with or without rate-adaptation
Right ventricular (RV) apical pacing results in an asynchronous contraction as the electrical
wave front proceeds slowly through myocardial tissue instead of the more rapidly conducting
His-Purkinje system. This leads to a situation similar to left bundle branch block (LBBB) with
a wide QRS, wherein the left ventricular septum contracts first, followed much later by the
free (lateral) wall, leading to an asynchronous contraction and thus to a reduction in forward
cardiac output. Rosenqvist et al11 measured the left ventricular ejection fraction (LVEF) during
atrial pacing with a conducted QRS, AV sequential pacing, and RV pacing. LVEF was highest
during atrial pacing with a conducted QRS, intermediate during AV sequential pacing, and
lowest during ventricular pacing. The first large scale evidence in favour of this concept came
from a retrospective analysis of the Mode Selection Trial (MOST), in which risks of hospitaliza-
tion due to heart failure were linked directly to right ventricular apex pacing regardless of
pacing mode.12-14 Although AV synchrony is better than no AV synchrony, right ventricular
pacing is worse than ventricular activation via the conduction system. The benefit of the AV
synchrony may be offset by the impact of direct ventricular pacing in the long-term. Thus, it
is possible that a higher amount of ventricular pacing even with conventional DDDR pacing
systems has adverse long-term effects on ventricular performance that mitigate the benefit of
AV synchrony, and lead to a greater incidence of heart failure, in addition to an increased
incidence of atrial fibrillation.
In addition to the above mechanisms, inappropriate programming of the A-V delay (the
pacemaker’s equivalent of the P-R interval) to either too short or too long a value for an
individual patient may occasionally lead to diastolic mitral regurgitation causing raised
pulmonary capillary wedge pressures; such patients can also present with otherwise unexplained
heart failure.15
From a therapeutic standpoint, it is important to recognize that these forms of pacing-

induced heart failure are treatable. Upgrading a dual chamber pacing system to a triple chamber
one, i.e. with LV pacing in addition to RA and RV pacing in order to maintain interventricular
synchrony during pacing (Cardiac Resynchronisation Therapy, CRT) may correct the heart
failure. Similarly, upgrading a VVI/R system to a DDDR or CRT system may be beneficial.16
In other patients, simply changing the pacemaker programming to minimize the amount of
pacing or the A-V delay might show benefit.
Pacing-induced Atrial Fibrillation

Most, but not all studies comparing atrial based (i.e. atrioventricular/single chamber atrial –
AAI/R) pacing with ventricular based (i.e. single chamber right ventricular – VVI/R) pacing
show an increased incidence of atrial fibrillation (AF) with ventricular based pacing.7,9-10,17
Healey et al,18 in a meta-analysis of RCTs comparing atrial based vs ventricular based
pacing concluded that atrial based pacing decreases the risk of atrial fibrillation. AAI/R pacing
is used infrequently because patients with sinus node dysfunction have a yearly risk of
0.6-2% of developing new onset atrio-ventricular conduction block, a risk considered by many
operators to be too high to implant only an atrial lead in them.19 In practice, therefore, the
intensivist is more likely to encounter either VVI/R or DDD/R pacemakers. It is worthwhile to
remember that if a patient with a VVI/R pacemaker presents with otherwise unexplained AF
and/ or its attendant complications, the pacemaker may need to be looked at carefully.
Pacing–induced Atrioventricular Regurgitation

Lack of co-ordination between atrial and ventricular contractions in VVI/R pacing usually
causes trivial to mild systolic mitral regurgitation (MR); rarely, this can lead to haemodynamically
significant regurgitation, especially if the patient has a pre-existing mitral disease. The role of
DDDR pacing in causing significant diastolic mitral regurgitation, albeit rare, has already been
explained. Both of these can present as unexplained heart failure following a pacemaker
implantation. Further, patients presenting with signs of new onset right heart failure after a
pacemaker implantation should be screened for a tricuspid regurgitation secondary to
pacemaker lead–related tricuspid valve infective endocarditis.
Device-related Infection20
Pacemaker pocket infection can occasionally be the cause of pyrexia of unknown origin (PUO).
Though uncommon (incidence of 0.5-1.2% over 35000 patient follow-up years),18 this can
occur in three settings. Acute (within days to few weeks post-implant) infections are possible
although they are rare. More commonly, infections are seen to occur weeks to months after
the procedure, and present as a subacute local abscess. If left untreated, this can eventually
erode through the skin; or less commonly, track to the heart and cause right heart endo-
carditis. Septic emboli from an infected lead can occasionally embolise to the lung. Thus, the
clinical presentation of device-related infection could be one of unexplained pneumonia or

pulmonary embolism. Finally, device-related infection can also occur months to years after
implantation. This scenario, which is more common in subjects with relatively less subcutaneous
fat (as in the elderly), the pacemaker gradually erodes through its subcutaneous pocket and
becomes adherent to the overlying skin. Subsequently, the whole assembly gets infected. This
process should be readily obvious on local examination.
To summarise, device-related infection should be suspected in a patient with a pacemaker
if the patient presents with fever weeks to months after device implantation or the patient has
recurrent unexplained episodes of pneumonitis. In both these instances, if a transthoracic
echocardiogram is non-diagnostic, a trans-esophageal echocardiogram (TOE) should be done
to rule out right heart endocarditis. In a series of 500 consecutive pacemaker infections,
Staphylococcus species were identified as the culprit in 88%, while gram-negative bacilli were
implicated in only 9%. While this may serve as a useful guide to start antibiotic therapy,
removal of the device with the leads may be required, if the infection is not controlled with
antibiotics.
CAN THE PACEMAKER CONTRIBUTE TO THE DIAGNOSIS OR TREATMENT

Current day pacemakers are capable of recording and storing data. Examination of intracardiac
electrograms, obtained by interrogating the device with manufacturer-specific programmer,
may allow diagnosis of spontaneous arrhythmia mechanisms that are not apparent on surface
electrocardiogram. The device memory stores heart histogram data that provide insight into
chronotropic competence, heart rate in atrial fibrillation, and spontaneous arrhythmia burden.
Arrhythmia logs (often with stored electrograms) gives details regarding past rhythm
disturbances.
Critically ill patients in the ICU, particularly those admitted with fever or sepsis, often
require a higher cardiac output. These patients need a heart rate, which is appropriate and
matched to their metabolic needs. A lower background pacemaker rate can be reprogrammed
to higher rate which, in turn increases the cardiac output. In addition, pacing threshold may
rise in these patients, thus a change in the pacing output may also be needed. Higher pacing
rates may be required in patients with heart failure and acute decompensation due to
precipitating reasons other than pacing. Simply increasing the rate of the pacemaker may
help tide over the crisis. The AV interval may also be altered to improve cardiac output. In
some cases, it is appropriate to lengthen this interval to promote intrinsic conduction and
minimize right ventricular pacing that may contribute to heart failure, as described in earlier
sections.
A lower rate may also need to be adjusted to suppress bradycardia-dependent arrhythmias.
Potentially fatal arrhythmia like Torsades de pointes may be heart rate or bradycardia
dependent, so by increasing the pacemaker rate to 80-90 per min, the problem can be
averted.21
Urgent reprogramming may occasionally be needed to decrease the heart rate. If tracking
of atrial arrhythmias is causing inappropriately rapid pacing, reprogramming to single chamber
mode will prevent this until the arrhythmia can be controlled. Patients with a rate-adaptive
pacemaker that tracks physiologic parameters (e.g., minute ventilation) may experience
inappropriately rapid pacing, which can be prevented by deactivating this rate-adaptive feature.
Occasionally, implanted devices may be used to terminate spontaneous arrhythmias. Using
the pacemaker programmer, atrial burst stimulation may be delivered to terminate atrial
tachycardia or flutter (overdrive pacing). Ventricular burst stimulation may be delivered through
a permanent pacer to terminate organized ventricular tachycardia. But this should always be
done with the defibrillator back up, as some of these attempts in the ventricle can accelerate
Ventricular Tachycardia or convert it to fibrillation.
WILL THE PACEMAKER POTENTIALLY INTERFERE WITH ICU PROCEDURES

Critically ill patients with an permanent pacemaker may require other invasive procedures.
So, it is important to be aware of the potential problems during these procedures.
Central Line Placement

Insertion of central venous lines and pulmonary artery catheters may require special care in
patients with permanent pacemakers. Subclavian venipuncture on the same side as an
implanted device should be avoided because of a substantially high risk of needle puncture
through the insulation of indwelling leads and device malfunction. Internal jugular venipuncture
should be safe on either side. Care must be exercised while floating pulmonary catheters
through the right heart, where repeated attempts may cause the catheter to dislodge pacing
leads. This is a concern mainly for passive fixation leads, especially left ventricular leads placed
via the coronary sinus that are particularly prone to dislodgement. In such patients, pulmonary
artery catheters should be avoided, if at all possible. Optimally, leads should be passed through
the right heart under fluoroscopic guidance
External Cardioversion or Defibrillation

Patients with implanted pacemakers, defibrillators can undergo external cardioversion or
defibrillation in an emergency. During elective or emergent external cardioversion or
defibrillation, the external paddles / electrodes should be placed distant from the permanent
pacemaker so that the external energy discharge does not harm the device. Defibrillation
paddles should be positioned anteroposteriorly and as far away from the pacemaker or lead
as possible. After external energy discharge, pacing thresholds may rise, leading to a loss of
pacing capture. Thus, whenever possible, the pacemaker should be interrogated before external
energy delivery, and the pacing output should be increased for the procedure. After the
procedure, the pacemaker should always be interrogated once more to ensure that device
function and/or programmed parameters have not been altered.
Electromagnetic Interference
Use of electrocautery may interfere with pacemaker functioning and it is advisable to
programme the pacemaker to a non-demand mode (VOO/DOO). One must remember that
MR imaging is contraindicated in patients with an implanted device and an alternative imaging
modality should be chosen.
CONCLUSION
Over the years, the number of pacemakers being implanted has increased as their indications
have continued to grow. Along with this, there has been an increase in complexity of these
devices, often to the point at which doctors without formal training are not able to handle
them knowledgeably any more. It is important that Emergency Medicine and Critical Care
specialists are conversant with pacemaker related issues, in order to manage more
comprehensively, patients with an implanted pacemaker.
Resources for Further Reading

• Ellenbogen KA, Kay GN, Lau CP, Wilkoff B. Clinical Cardiac Pacing, Defibrillation and
Resynchronisation Therapy. 3rd edition, WB Saunders, 2006.
• Kenny T. Nuts and bolts of ICD therapy. Blackwell Futura, 2006.
REFERENCES
1. Hayes DL, Vlietstra RE, Trusty JM, Downing TP, Cavarocchi NC. A shorter hospital stay after cardiac
pacemaker implantation. Mayo Clin Proc 1988;63:236-40.
2. Mugica J, Ripart A. Twelve years’ experience with cardiac pacing leads: clinical conclusions for 8,004
cases. Clin Pacing Electrophysiol 1984;2:513-32.
3. Gabry MD, Behrens M, Andrews C, Wanliss M, Klementowicz PT, Furman S. Comparison of myopotential
interference in unipolar-bipolar programmable DDD pacemakers. Pacing Clin Electrophysiol 1987;10:1322-
30.
4. Barold SS, Falkoff MD, Ong LS, Heinle RA. Oversensing by single-chamber pacemakers: mechanisms,
diagnosis, and treatment. Cardiol Clin 1985;3:565-85.
5. Furman S, Fisher JD. Endless loop tachycardia in an AV universal (DDD) pacemaker. Pacing and Clin
Electrophysiol 1982;5:486-9.
6. Thackray SDR, Witte KKA, Nikitin NP, Clark AL, Kay GC, Cleland JGF. The prevalence of heart failure
and asymptomatic left ventricular systolic dysfunction in a typical regional pacemaker population. Eur
Heart J 2003;24:1143-52.
7. Toff WD, Camm AJ, Skehan JD. For the United Kingdom Pacing and Cardiovascular Events Trial (UK-
PACE) Investigators. Single chamber versus dual-chamber pacing for high-grade atrioventricular block. N
Engl J Med 2005;353:145-55.
8. Lamas GA. Physiological consequences of normal atrioventricular conduction: applicability to modern
cardiac pacing. J Card Surg 1989;4:89-98.
9. Connolly SJ, Kerr CR, Gent M, et al. For the Canadian Trial of Physiological Pacing Investigators. Effects
of physiological pacing versus ventricular pacing on the risk of stroke and death due to cardiovascular
causes. N Engl J Med 2000;342:1385-91.
10. Lamas GA, Lee KL, Sweeney MO et al for the MOST Investigators. Ventricular pacing or dual chamber
pacing for sinus node dysfunction. N Engl J Med 2002;346:1854-62.
11. Rosenqvist M, Isaaz K, Botvinick EH, et al. Relative importance of activation sequence compared to
atrioventricular synchrony in left ventricular function. Am J Cardiol 1991;67:148-56.
12. Sweeny MO, Hellkamp AS, Ellenbogen KA, et al. Adverse effect of ventricular pacing on heart failure and
atrial fibrillation among patients with normal baseline QRS duration in a clinical trial of pacemaker therapy
for sinus node dysfunction. Circulation 2003;107:2932-7.
13. The DAVID Trial Investigators. Dual-Chamber Pacing or Ventricular Backup Pacing in Patients with an
Implantable Defibrillator (DAVID) Trial. JAMA 2002;288:3115-23.
14. Sharma AD, Rizo-Patron C, Hallstrom AP et al for the DAVID Investigators. Percent right ventricular pacing
predicts outcomes in the DAVID Trial. Heart Rhythm 2005;2:830-4.
15. Barold SS, Ilercil A, Herweg B. Echocardiographic optimization of the atrioventricular and interventricular
intervals during cardiac resynchronization. Europace 2008;10:88-95.
16. Pipes RR, Nanthakumar K, Parker JD. Biventricular Pacemaker Upgrade in Previously Paced Heart Failure
Patients—Improvements in Ventricular Dyssynchrony. J Card Fail 2006;12:199-204.
17. Andersen HR, Thuesen L, Bagger JP, Vesterlund T, Thomsen PE. Prospective randomised trial of atrial
versus ventricular pacing in sick-sinus syndrome. Lancet 1994;344:1523-8.
18. Healey J S, Toff WD, Lamas GA, Andersen HR, Thorpe KE, Ellenbogen KA, et al. Cardiovascular Outcomes
With Atrial-Based Pacing Compared With Ventricular Pacing: Meta-Analysis of Randomized Trials, Using
Individual Patient Data. Circulation 2006;114:11-7.
19. Andersen HR, Nielsen JC, Thomsen PEB, Thuesen L, Mortensen PT, Vesterlund T, Pedersen AK. Long-
term follow-up of patients from a randomised trial of atrial versus ventricular pacing for sick-sinus syndrome.
Lancet 1997;350:1210-6.
20. Byrd C L. Managing Device-Related Complications and Transvenous Lead Extraction. In: Ellenbogen KA,
Kay GN, Lau CP, Wilkoff B (Eds): Clinical Cardiac Pacing, Defibrillation and Resynchronization Therapy,
3rd Edition, WB Saunders, New York, 2006.
21. Viskin S, Fish R, Roth A, Copperman Y. Prevention of Torsades de pointes in the congenital long QT
syndrome: use of a pause prevention pacing algorithm. Heart 1998;79:417-9.
Five
Role of Amiodarone in Cardioversion
of Recent Onset Atrial Fibrillation
Sumesh Arora, Vineet Nayyar
INTRODUCTION
Atrial fibrillation (AF) is the most common sustained arrhythmia seen in the general population.1
In the Framingham heart study,2 the risk of development of AF among people older than
40 years was estimated to be as high as 1 in 4. This represents a huge burden of disease in
the community, especially among the elderly. New onset AF is also common in hospitalised
patients.
In recent years, it has been recognized that occurrence of AF in ICU patients may be a
marker if increased severity of illness at admission. It has also been suggested that patients
who develop AF have a higher mortality during their hospitalisation. Whether AF is a cause
of increased mortality or is just associated with increased severity of illness is still unknown.
What is also not known is whether treatment of new onset AF in ICU has any impact on
reducing mortality or length of stay in ICU or long-term morbidity.
EPIDEMIOLOGY OF AF IN ICU
A number of studies have looked at the occurrence of arrhythmias in the ICU (Table 5.1).
Annane et al3 recently published the results of a multi-centre observational study done in 26
European ICUs. Over a period of one month, 12% of ICU patients were found to have
sustained arrhythmias. Supraventricular arrhythmias were more common than ventricular
arrhythmias and AF was documented to be the most common sustained arrhythmia, with a
prevalence of 6.5% during the study period.
The risk factors associated with AF in ICU have been studied in single centre studies. In
a prospective observational study in a surgical ICU, Seguin et al4 estimated the prevalence
of AF to be 5.3%. The study found advanced age, shock [particularly septic shock], blunt
chest trauma, pulmonary artery catheter use and withdrawal from calcium channel blockers
were risk factors for occurrence of AF in ICU. Patients who developed AF had significantly
higher mortality and longer hospital length of stay. Most of the episodes (71%) of AF occurred
in patients over the age of 69 years. Seguin5 et al also studied the epidemiology of AF in
Table 5.1: Prevalence of arrhythmias and atrial fibrillation in ICU
Authors Year Type of study Patients Number Setting Preva- Mortality RR Risk factors identified for AF
lence death
A. Studies looking specifically at Atrial Fibrillation
Arora 1 2007 Prospective ICU pts 61 Medical / 29.5% 56%* 2.7 Sepsis, age > 75 years, APACHE score
age >50 surgical > 20
non-cardiac
Shankar2 2007 Retrospective All ICU 110 NR 26.3% … NR Need for inotropes, sepsis, hypertension,
patients
Seguin 3 2006 Prospective All trauma 293 Surgical ICU 5.5% 25% NR Age >40, SAPS > 30, need for vaso-
patients pressors, >3 body regions traumatized,
SIRS
Seguin 4 2004 Prospective All ICU 429 Surgical ICU 5.3% 37.5%* NR Older age, high SAPS, shock, chest
admissions trauma, PA catheter use, therapy with
Ca channel blocker
B. Studies looking at Arrhythmias in general
Annane 7 2008 Prospective, All in ICU 1341 Medical / 6.5% … NR Age, higher SAPS, Cardiovascular or
Multi-centre >24 hrs surgical, endocrine co-morbidity, need for
mixed ICUs ventilation/ inotropes
Goodman8 2007 Prospective All ICU 611 Medical- 8.5% … NR Sepsis, low dose dopamine, advanced
patients surgical ICU age, hypothyroidism, pulmonary disease [for
all supraventricular arrhythmias]
Reinelt5 2000 Prospective All admissions 756 Medical/cardiac, 8.3% …. NR …
cardiac surgical
Artucio 6 1990 Retrospective All admissions 2820 Medical/surgical, 14.8% …. NR Severity of illness
non-cardiac
Key: NR: not reported, APACHE: Acute physiology and chronic health evaluation score, SAPS: Simplified acute physiology score, SIRS: Systemic inflammatory
response syndrome, PA: Pulmonary artery catheter.
Role of Amiodarone in Cardioversion of Recent Onset Atrial Fibrillation
43
patients with patients with trauma. The prevalence of AF in this group of patients was estimated
to be 5.5%. Advanced age and increased severity of illness were once again identified as risk
factors for occurrence of AF in ICU.
Arora et al6 prospectively studied the prevalence of AF among patients admitted to a
tertiary care, medical-surgical ICU. This study included only patients over the age of 50 years
and not surprisingly, found a higher prevalence rate (29.5%) in this population of patients.
Age >75 years, higher severity of illness [APACHE > 20] and sepsis as part of admission
diagnosis were independent predictors of development of AF. Shankar et al7 reported a similar
prevalence rate in a retrospective review of 110 patients.
So far, most epidemiological studies have been small and have been conducted over a
period ranging from a few weeks to months in single ICUs with a varying case-mix and illness
severity. Some studies have reported increased risk of hospital mortality in patients who develop
arrhythmias or AF in ICU.3,4,6 It is unclear, if AF by itself has an impact on mortality. Multi-
centre studies are therefore needed to quantify the exact burden of disease, risk factors
associated with onset of AF and the impact on ICU morbidity and mortality.
Annane3 et al reported an increased risk of hospital mortality for patients who had
supraventricular arrhythmias in ICU. Most of these patients had AF. The odds ratio for the risk
of mortality associated with AF was 1.95 [95% CI 1.27-3.01]. When adjusted for age, SAPS
II score, medical history and ICU admission diagnosis, the adjusted OR was 0.69 (95% CI
0.38-1.26) suggesting that supraventricular arrhythmias described in his cohort of patients
were not directly related to mortality. Instead, they were a marker of increased risk of death
in ICU.
AF After Cardiac Surgery

Atrial Fibrillation is common after cardiac surgery. The prevalence of AF after coronary artery
bypass surgery (CABG) is more than 30%.9 The risk is 60% or higher for patients undergoing
valve surgery.10 Postoperative AF usually occurs between day 2-4 post op, and is often short
lived and self limiting. When sustained, it can result in patient discomfort, hypotension, heart
failure, stroke and thrombo-embolic complications. This, in turn, is associated with prolonged
ICU and hospital length of stay, increased risk of ventricular arrhythmias and the need for
permanent pacemaker, increased resource utilisation and repeat hospital admissions.11
The epidemiology of postop AF following cardiac surgery has been extensively reviewed
earlier (Update 2006) and is not the purpose of this appraisal. However, it is worth emphasising
that in keeping with studies done in general ICU patients, sicker patients with peri-operative
myocardial ischaemia, sepsis, renal failure, need for intra-aortic balloon pump or re-operation
or readmission to ICU are more likely to develop AF in peri-operative period. Such a trend
has, as discussed above, been observed in general non-cardiac ICU12 patients.
Role of Amiodarone in Cardioversion of Recent Onset Atrial Fibrillation 45
AF After Major Non-cardiac Thoracic Surgery

Atrial fibrillation is a common complication after pulmonary, oesophageal and other non-
cardiac thoracic surgery. In a large prospective observational study of 2588 patients who
underwent major thoracic surgery on lung, oesophagus, mediastinum or chest wall, the
prevalence of AF was estimated to be 12.3%.13 Among other risk factors identified in this
study, oesophagectomy and pneumonectomy were found to be independently associated
with a risk of developing AF.
The prevalence of AF after lung resection surgery has been reported to vary between
12-21%.14-17 Factors that have been associated with onset of AF include advanced age, history
of hypertension, pneumonectomy, upper lobectomy, intra-pericardial surgery and lymph node
dissection. Video assisted thoracic surgery does not reduce the risk of AF following lung
surgery18. Excessive sympathetic activity following pulmonary surgery has been suggested as
a reason for the onset of post-op AF. Oka et al19 compared epidural morphine with epidural
bupivacaine after lung surgery. The risk of supraventricular arrhythmias was much less in the
bupivacaine group compared to morphine.
AF has also been observed to occur frequently after oesophagectomy. In a study by Murthy20
et al, AF was observed to occur in 22% of patients who underwent oesophagectomy.
Postoperative pulmonary complications and sepsis were associated with the onset of AF, which
when sustained, was associated with increased risk of hospital mortality and length of stay.
In a small case series of patients who underwent minimally invasive oesophagectomy, AF
was observed to occur in 16% of patients.21 Postoperative hypoxia, history of COPD,
postoperative thoracic-gastric dilatation and history of cardiac disease were associated with
onset of AF.
TREATMENT OPTIONS FOR MANAGEMENT OF RECENT ONSET AF IN ICU

Rhythm control is presumed to be the management of choice for all episodes of new onset
AF in the ICU. Primary prophylaxis of AF in high-risk patients is also an attractive option, but
this has not been systematically studied so far. This review focuses upon treatment options
following the onset of AF with particular emphasis on the use of amiodarone in ICU patients.
Approximately 40-60% of patients who develop AF will revert spontaneously to sinus
rhythm within 24 hours.22,23 As the duration of AF increases, electrical remodelling occurs,
which makes it hard for patients to revert to sinus rhythm. Thus, treatment to control rhythm
in an ICU patient, if used, should be started early. However, the manner in which rhythm
control can be best achieved is a matter of some debate. AF can be converted to sinus rhythm
electrically or pharmacologically. Electrical cardioversion is needed immediately, if AF is
associated with hypotension, heart failure or myocardial ischaemia. There is limited evidence
based data on pharmacological conversion and maintenance of sinus rhythm in ICU and
post cardiac surgery patients.
Upto 45%24 relapse within 24-48 hours of a successful cardioversion. Inflammation and
increased sympathetic tone have an important etiological role. Underlying heart disease and
atrial size >40 mm also contribute to an increased risk of relapse but additionally, to a lack
of response to cardioversion attempts in the first place.
The proportion of patients leaving hospital with sustained AF who, at the time of their ICU
admission, were in sinus rhythm is unknown. It is also not known whether active intervention
following onset of AF can or does improve this number. The natural history of AF in the ICU
is therefore, not clear. It is also not clear, if treatment used at the onset of AF in the ICU needs
to continue beyond the point at which the rhythm returns to normal.
ROLE OF AMIODARONE IN CARDIOVERSION OF RECENT ONSET AF

Amiodarone is a structural analogue of thyroid hormone. In heart, it reduces sinus and AV
nodal automaticity, reduces the conduction across the AV tract and bypass tracts, increases
action potential duration and prolongs the refractory period of myocardial tissue. It also has
weak beta blocking activity.
Amiodarone is the most commonly used antiarrhythmic drug for the treatment of
arrhythmias in ICU.25 A recent systematic review26 on the treatment of AF in ICU reviewed
papers published from 1966 to 2006. Over four decades, only four randomised controlled
trials of high quality were identified - most of them for treatment of supraventricular tachycardias
rather than AF alone. The authors concluded that there was insufficient evidence to support
the use of one agent over another for the treatment of AF in ICU. Thus, the best anti-arrhythmic
therapy for the management of new onset AF is, a matter of some debate.
Amiodarone however, continues to be used frequently for AF, almost as a knee jerk response
to the onset of this arrhythmia in the ICU. Surprisingly, amiodarone is not even approved for
this indication in the USA.27 A number of studies published over the last 10 years have
compared amiodarone with placebo or other drugs in clinical trials (Table 5.2). Amongst these,
only 4 studies have compared amiodarone to placebo22,23,28,29 in a clinical trial for the treatment
of new onset AF. In a prospective, placebo controlled randomised trial, Cotter22 compared
amiodarone to placebo for treatment of recent onset AF. 92% of patients in the amiodarone
group reverted to sinus rhythm within 24 hrs, compared to 64% in the placebo group
(p=0.0017). Of the patients in placebo group who did not convert to sinus rhythm, 83%
reverted to sinus rhythm after amiodarone was administered. In another prospective
randomised study by Cybulski et al,23 amiodarone restored sinus rhythm in 83% of patients
compared to 44% in placebo group (P<0.0001). Two out of 106 patients in amiodarone
group had asystole during treatment. This spontaneously resolved itself to sinus rhythm.
Bradycardia was observed in 6 (5.6%) patients without clinical symptoms. These results indicate
that amiodarone is relatively safe in patients with recent onset atrial fibrillation. Heart rate
however needs to be closely monitored during administration of amiodarone. Bradycardia
often occurs after restoration of sinus rhythm. Thus, the period of observation should extend
to a few hours after restoration of sinus rhythm.
Table 5.2: Amiodarone compared to other modes of treatment for AF of recent onset
Author Year Study design N Amiodarone Comparison Cardioversion Success Median time Comments
Bolus Infusion/Oral Amiodarone Others
Xandos32 2007 223 Bolus 20 mg/kg over Procainamide 81% 84% … Hypotension with
24 hrs amiodarone
Kochiadakis 28 2007 P, R, SB 362 300 mg IV 20 mg/kg over Procainamide, 89% Pc 68% 9 hrs Hypotension and
24 hrs Propafenone, Pf 80% phlebitis with
placebo amiodarone
Xu 46 2007 P, R, UB 80 - - Acupuncture–PC6, 67.5% 85% <1 hr Both paroxysmal
HT7 and CV17 atrial flutter and AF
points studied
Bobbio 47 2007 P, NR, UB 30 600 mg over 300 mg over Diltiazem 80% 80% … High recurrence rate–
3 hrs 21 hrs rhythm restored with
repeat
Kafkas38 2007 P, R 108 5 mg/kg IV 1500 mg over Ibutilide 69% 77% … Both atrial flutter
24 hrs and fibrillation studied
Thomas 36 2003 P, R 140 10 mg/kg IV No Sotalol, Digoxin 51% St 44%, 4.5 hrs 20% patients had AF
Dg 50% for >48 hr
Cybulski 23 2003 P, R, 160 5 mg/kg IV 10 mg/kg over Placebo 83% 44% 8.2 hrs 2 patients had asystole
20 hrs with amiodarone
Bernard24 2003 P, R DB, 40 5 mg/kg IV 15 mg/kg over Ibutilide 50% 45% 8 hrs 2 patients had hypo-
cross-over 24 hrs tension with
amiodarone
Karth 42 2001 P, R 60 300 mg IV + 45 mg/hr over Diltiazem 45% 30% … …
300 mg IV 24 hrs
Martinez- 2000 P, R, SB 150 5 mg/kg IV 50 mg/hr, Flecainide, 64% Fl 90%, <1 hr Slower time to revert
Marcos33 duration N/a Propafenone Pf 72% with amiodarone
Joseph 35 2000 P, R, UB 120 5 mg/kg IV Oral 400 mg Digoxin, Sotalol 77% St 88% <1 hr
Q8h
Cotter 22 1999 P, R 100 None 125 mg/kg to Placebo 92% 64% …
total dose 3 g
Boriani 29 1998 P, R 417 5 mg/kg IV 1800 mg over Placebo, 57% 37%, …
24 hrs Propafenone Pf 75%
Flecainide
Key: N: Number; P:Prospective, R: Randomized, DB: Double blind, SB: Single blind, UB: Unblinded, Pc: procainamide, Pf: Propafenone, St: sotalol, Dg: Digoxin,
Role of Amiodarone in Cardioversion of Recent Onset Atrial Fibrillation
Fl: Flecainide,
47
EFFICACY OF AMIODARONE COMPARED TO OTHER DRUGS

If a decision is made to control rhythm, antiarrhythmics certainly increase the chance of
successful conversion to sinus rhythm. The choice of an agent is influenced, in large measure,
by the clinical situation but also by practices prevalent in a particular unit or the attending
physicians’ understanding of literature. A summary of clinical trials done in patients with new
or recent onset AF is presented in Table 5.2 and discussed below.
DC Cardioversion for AF
Mayr30 et al studied the efficacy of electrical cardioversion for AF in ICU. Most patients
responded to one or two shocks. However, at 24 hours, only 16% of the patients remained
in sinus rhythm. Therefore by itself, electrical cardioversion is unlikely to be useful for restoring
sinus rhythm. Cardioversion needs to be combined with drug therapy to maintain sinus rhythm.
There are no studies comparing amiodarone to electrical cardioversion for AF of recent onset.
In a case series by Neumayr31 et al, electric cardioversion was successful in 67 out of 69 ICU
patients. There were no major side effects. Pre and post-cardioversion measurement of cardiac
enzymes did not suggest any myocardial injury from electric shock.
Amiodarone vs. Procainamide

In a recent study, Xandos32 compared efficacy of amiodarone and procainamide for
cardioversion of acute AF. Both drugs achieved similar success rate of over 80% in 24 hrs.
Procainamide was quicker during the loading phase of the drug but there was no difference
in the rate of cardioversion after that. Side effects were rare for both. In a placebo-controlled
trial, Kochiadakis28 compared amiodarone, procainamide and propafenone for efficacy in
converting AF to sinus rhythm within 24 hours. Amiodarone [89%] was more effective than
procainamide [68%] or placebo [61%].
Amiodarone vs. Flecainide

Flecainide has not been studied for acute onset AF in recent years. Boriani29 compared
Amiodarone, Flecainide and propafenone in a placebo controlled trial in AF of up to 7 days
duration. Cardioversion rates were higher with flecainide compared to amiodarone. However
in 4 out of 69 patients, flecainide converted AF to atrial flutter with 1:1 conduction rate and
a very fast ventricular rate. No serious side effects were observed in the amiodarone group.
In a randomised study by Marco-Martinez,33 flecainide was once again, found to be more
effective than amiodarone for cardioversion at 12 hours. Donovan34 compared flecainide to
amiodarone in a placebo-controlled trial in 98 patients with AF. The conversion rate was
higher with Flecainide at 2 hrs, but the effect disappeared at 8 hours. In this study, amiodarone
bolus dose was not followed by infusion. It is possible that flecainide may prove superior to
amiodarone for acute AF, but this has not been rigorously studied so far.
Amiodarone vs. Sotalol

Joseph35 et al compared amiodarone and sotalol for pharmacological cardioversion of new
onset AF. Both amiodarone and sotalol were equally effective in converting AF to sinus rhythm.
Patients treated with sotalol had lower heart rate on conversion to sinus rhythm. Thomas36
compared amiodarone to sotalol and digoxin. The study enrolled 140 patients with AF among
whom, nearly 20% had AF >48 hours. The efficacy of amiodarone, sotalol and digoxin was
similar in this study. In the Sotalol Amiodarone Atrial Fibrillation Efficacy Trial37 (SAFE-T),
amiodarone as sotalol were equally efficacious in converting AF to sinus rhythm. This trial
was done in patients with chronic AF who were anticoagulated. The results may not be
applicable for patients who are in ICU and have acute onset of AF.
Amiodarone vs. Ibutilide

In the last 10 years, 2 studies have compared amiodarone and ibutilide for recent onset AF
(Table 5.2). Kafkas et al38 compared ibutilide, another class III antiarrhythmic, to amiodarone
for recent onset AF and atrial flutter [within previous 3-48 hrs]. While ibutilide proved to be
superior to amiodarone for conversion of atrial flutter, both drugs were equivalent in their
efficacy to convert AF to sinus rhythm. Ibutilide was associated with increased risk of torsades
de pointes. Bernard39 in a double blind, randomised cross over trial found similar efficacy of
amiodarone and ibutilide at 4 and 29 hrs.
In another study by Hannersdorf,40 ibutilide was used for cardioversion of AF or atrial
flutter after amiodarone had been used for 2 hrs. 27% of patients failed to achieve sinus
rhythm after 2 hrs of amiodarone infusion. In these patients, Mg was administered and high
normal potassium levels were achieved. This was followed by use of ibutilide, which led to
cardioversion in nearly 80% of patients still in AF. Amiodarone is a highly lipophilic drug with
slow onset of action. Therefore, cardioversion in those patients who received ibutilide could
very well be the effect of amiodarone or magnesium. The number of patients in the study
was too small [26 patients with only 7 patients in AF] to draw much meaningful conclusions
regarding the usefulness of ibutilide for cardioversion in AF. There remain safety concerns
with use of ibutilide in conjunction with amiodarone. In this study, 3 out of 27 patients
developed non-sustained torsades, although none had sustained VT. QTc was also significantly
prolonged after use of ibutilide.
In a clinical study by Dilaveris et al,41 the order of administration of the two drugs was
reversed. Ibutilide achieved 68% cardioversion at 4 hrs, whereas combination of ibutilide
followed by amiodarone in those who have not cardioverted with ibutilide alone led to
cardioversion in 96% of patients. No major side effects were observed in this study. In both
the studies, which used combination of amiodarone and ibutilide, the second drug was given
within few hours of the administration of the first drug. This led to smaller cardioversion rates
compared to studies in which only one drug was given and the successful cardioversion was
defined at 24 hours. Both amiodarone and ibutilide may prolong QT interval. Sequential
administration of both drugs in ICU may lead to increased risk of ventricular arrhythmias.
Amiodarone vs. Propafenone

Two studies have compared amiodarone with propafenone in recent years. Marco Martinez33
found both amiodarone and propafenone to be equally effective for cardioversion at 12 hours.
Boriani et al29 found propafenone to be superior to amiodarone for cardioversion at 8 hrs.
One patient had pulmonary oedema during treatment with intravenous propafenone.
Propafenone was effective both as oral and intravenous (IV) treatment, but the effect was
quicker with parentral administration. With intravenous propafenone, most patients achieved
sinus rhythm in less than 3 hours. It is possible that propafenone may prove superior to
amiodarone, particularly when rapid cardioversion is needed in hemodynamically unstable
patients. The effect has however not been rigorously studied.
Amiodarone vs. Diltiazem

In a study done in ICU,42 amiodarone was compared to diltiazem. The success of cardioversion
was similar with two drugs. Rate control was better with diltiazem, but in 20% patients diltiazem
had to be discontinued for hypotension. Diltiazem therefore may not be a good choice for
patients with hypotension associated with new onset AF.
Amiodarone vs. Magnesium

There are no recent studies comparing amiodarone to Magnesium (Mg). In a small, randomised
ICU trial, Moran et al43 studied 42 critically ill patients with AF. Compared to amiodarone, Mg
was more likely to cardiovert AF to sinus rhythm. In an observational study, Sleeswijk44 reported
successful treatment of acute onset AF with Mg followed by amiodarone if no response was
seen within an hour. 90% of patients converted to sinus rhythm with the use of combination
of the two drugs.
Onalan et al45 reported a meta-analysis on the use of Mg in AF with rapid ventricular rate.
Compared to placebo, Mg was found to be a useful drug for both rate control [OR 2.97, 95%
CI 1.78-4.97] and rhythm control [OR 1.6, 95% CI 1.07-2.39]. There were no major side
effects associated with the use of Mg.
Mg is relatively safe drug and does not have any significant drug interaction with
amiodarone. Therefore, combination of two drugs may be useful strategy as has been reported
in the series by Sleeswijk et al above. The comparative or combined effect of amiodarone
and Mg has also not been studied in a clinical trial.
ADMINISTRATION OF AMIODARONE
Dose of Amiodarone
Clinical studies have used different doses of amiodarone ranging from a bolus of 5 mg/kg to
10 mg/kg intravenously. Bolus dose is normally followed by an infusion over 24-48 hours. In
a study comparing low dose [50 mg/hr] vs. high dose amiodarone [100 mg/hr] for AF and
atrial flutter, high dose amiodarone was associated with higher rate of success at cardioversion48
(80% vs. 60%). The difference between the two groups was more apparent after 10 hours of
infusion.
Route of Administration
Peripheral intravenous administration of amiodarone is painful and may lead to phlebitis.
The risk of phlebitis following peripheral administration of therapeutic doses of amiodarone
has been estimated to be approximately 14%.49 Xanthos et al50 compared oral vs. intravenous
administration of amiodarone in patients who had concurrent administration of digoxin. Both
oral and intravenous amiodarone were equivalent for cardioversion of AF to sinus rhythm,
with success rate of over 80%. Intravenous administration was associated with quicker
cardioversion but a higher risk of hypotension during administration. It can be concluded
from the study that for patients who develop AF in ICU and who do not have central venous
access, oral administration may be equally effective and may have less side effects.
Toxicity of Amiodarone
Hypotension may occur in response to intravenous administration of amiodarone. However,
in hemodynamically unstable patients, the beneficial effects of amiodarone on heart rate and
cardioversion may be far more useful than hypotension that may occur transiently. Holt et
al51 studied 10 critically ill patients, who were on inotropes and required amiodarone for
management of AF. Catecholamine dose was monitored while a loading dose of amiodarone
was given over 2 hrs. A significant reduction in heart rate and improvement in stroke volume
index was observed. Mean arterial pressure and cardiac index did not change significantly. In
another small study on 8 patients, Kumar52 reported that loading dose of amiodarone is well
tolerated.
Pulmonary toxicity is a well-established complication of long-term use of amiodarone and
occurs in approximately 3% of patients.53 It is dependent upon total cumulative dose of
amiodarone administered and occurs more frequently in patients who have pre-existing renal
disease. It is unknown if use of amiodarone in patients with pre-existing pulmonary disease
increases the risk of pulmonary toxicity. Recognition of pulmonary toxicity in ICU is challenging
because of multiple reasons for which patients may have respiratory failure. There have been
several isolated case reports of pulmonary toxicity from as little as 1500 mg dose of amiodarone.
In the AFFIRM study,54 use of amiodarone in patients with pre-existing pulmonary toxicity
increased the rate of diagnosis of amiodarone induced pulmonary toxicity, but did not increase
the risk of pulmonary cause mortality or all cause mortality. Therefore, amiodarone induced
pulmonary toxicity should be considered in the differential diagnosis of respiratory failure
when other common causes of respiratory failure have been excluded in critically ill patients.
Prolonged QT interval leading to torsades de pointes (TdP) is a potentially fatal complication
of amiodarone. In large cardiology studies where amiodarone was prescribed for AF or
ventricular arrhythmias, the risk of TdP with amiodarone was reported to be less than 1%.
In the AFFIRM study, prolonged QT interval was seen frequently, but the risk of TdP was
<0.5%, much less than sotalol. However, TdP have been reported frequently with acute,
short-term use of amiodarone. Amiodarone may interact with several other drugs that may
prolong QT interval, thus increasing the risk of TdP. Drug interaction have been reported with
concomitant use of beta blockers, digoxin55 [when used concomitantly for rate control in AF].
metronidazole,56 loratidine, trazodone57 and other drugs that may prolong QT interval, e.g.
macrolides, quinolones, azole antifungals may increase the risk of TdP58. Other factors observed
in small case series, that may prolong the risk of TdP include hypokalemia, bradycardia and
reduced LV ejection fraction.57 Overall, even though the risk of TdP with amiodarone may
be low, QT interval should be monitored and use of other drugs that may prolong QT interval
should be limited.
Clinically significant hepatotoxicity is usually a rare complication of chronic amiodarone
therapy, but has also been reported after intravenous bolus of amiodarone.59 Although there
are isolated case reports, large case series on use of amiodarone after cardiac surgery have
not reported any cases of hepatotoxicity. It is possible that patients who have other risk factors,
like hepatotoxic drugs or hypotension, may be more prone to develop hepatotoxicity from
amiodarone. It is also possible that hepatotoxicity is mediated by polysorbate 80—a vehicle
used in solutions of amiodarone.
Acute low backache has been reported following acute administration of amiodarone. 60
The reason for this is not known.
CONCLUSION
In summary, there is not much evidence in the intensive care literature on the use of amiodarone
for the management of new onset AF. Extrapolating data from the cardiology and emergency
medicine literature, it seems reasonable to assume that amiodarone is a useful and effective
drug for cardioversion of AF in ICU. It is better than placebo and most other anti-arrhythmics.
It has a low risk profile and is a drug that can be easily changed from intravenous to the oral
route, when needed. Amiodarone appears to work well in patients with poor LV function and
those who do not tolerate beta-blockers or calcium channel blockers.
Combination of Mg and amiodarone is an attractive option and deserve some more
research. The side effects with acute short-term treatment with amiodarone are rare. Therefore,
until more data is available, use of amiodarone for rhythm control in acute onset AF in ICU
should continue until more data comparing it to other antiarrhythmics become available.
More studies are required in the ICU setting to look at and compare the efficacy of cardioversion
and rate control with magnesium, sotalol and calcium channel antagonists.
REFERENCES
1. Kopecky SL, Gersh BJ, McGoon MD, et al. The natural history of lone atrial fibrillation. A population-
based study over three decades. N Engl J Med 1987;317:669-74.
2. Lloyd-Jones Donald M. Lifetime risk of development of atrial fibrillation. The Framingham heart study.
Circulation. 2004;110:1042-6.
3. Annane D, Se´bille V, Duboc D, Heuzey J, Sadoul N, Bouvier E, Bellissant E. Incidence and prognosis
of sustained arrhythmias in critically ill patients. Am J Respir Crit Care Med 2008;178:20-5.
4. Seguin P, Signouret T, Laviolle B, et al. Incidence and risk factors of AF in surgical intensive care unit. Crit
Care Med 2004;32:722-6.
5. Seguin P, Laviolle B, Maurice A, et al. Atrial fibrillation in trauma patients requiring intensive care. Intensive
Care Med 2006;32:398-404.
6. Arora S, Lang I, Nayyar V, Stachowsky E, Ross DL. Atrial Fibrillation in a tertiary care multidisciplinary
ICU – incidence and risk factors. Anesth Intensive Care 2007;35:707-13.
7. Shankar R, Welters I. Incidence and diagnosis of patients developing AF in ICU. Critical Care 2007;11
(suppl 2):221.
8. Goodman S, Shirov T, Weissmann C. Supraventricular arrhythmias in intensive care unit patients: short
and long-term consequences. Anesth Analg 2007;104:880-6.
9. Mathew J, Fonte M, Tudor I, Ramsay J, Duke P, Mazer C, et al. A Multicenter Risk Index for Atrial Fibrillation
After Cardiac Surgery. JAMA 2004:291;1720-9.
10. Creswell LL, Schuessler RB, Rosenbloom M, et al. Hazards of postoperative atrial arrhythmia. Ann Thorac
Surg 1993;56:539-49.
11. Juneja R, Mehta Y. Atrial fibrillation in post-cardiac surgical patient. In Nayyar V (Ed). Critical Care Update
2006, Jaypee Brothers Medical Publishers, New Delhi.
12. Aranki SF, Shaw DP, Adams DH, et al. Predictors of atrial fibrillation after coronary artery surgery: Current
trends and impact on hospital resources. Circulation 1996;94:390-7.
13. Vaporciyan AA, Correa AM, Rice DC, Roth JA, Smythe WR, Swisher SG, Walsh GL, Putnam JB Jr. Risk
factors associated with atrial fibrillation after non-cardiac thoracic surgery: Analysis of 2588 patients. J
Thorac Vasc Surg. 2004;127:779-86.
14. Gomez-Caro A, Moradiellos EJ, Ausin P, Diaz-Hellin V, Larru E, Perez-Anton JA, Martin de Nocolas JL,
et al. Risk factors for atrial fibrillation after thoracic surgery. Arch Broncopneumol 2006;42:9-13.
15. Cardinale D, Martinoni A, Cipolla CM, Civelli M, Lamantia G, Fiorentini C, Mezzetti M. Atrial fibrillation
after operation for lung cancer. Clinical and prognostic significance. Ann Thorac Surg 1999;68:1827-31.
16. Foroulis CN, Kotoulas C, Lachanas H, Lazopoulos, Konstantinou M, Lioulias AG. Factors associated with
cardiac rhythm disturbances in early post pneumonectomy period: a study of 259 pneumonectomies. Eur
J Cardiothorac Surg 2003;23:384-9.
17. Rena O, Papalia E, Oliaro A, Casadio C, Ruffini E, Filloso PL, Sacerdotge C, Maggi G. Supraventricular
arrhythmias after resection surgery of the lung. Eur J Cardiothorac Surg 2001;20:688-93.
18. Park BJ, Zhang H, Rusch VW, Amar D. Video-assisted thoracic surgery does not reduce the incidence of
postoperative atrial fibrillation after pulmonary lobectomy. J Thorac Cardiovasc Surg 2007;133:775-9.
19. Oka T, Ozawa Y, Ohkubo Y. Thoracic epidural bupivacaine attenuates supraventricular tachyarrhythmias
after pulmonary resection. Anaesth Analg 2001;93:253-9.
20. Murthy SC, Law S, Whooley BP, Alexandrou A, Chu KM, Wong J. Atrial fibrillation after esophagectomy
is a marker for postoperative morbidity and mortality. J Thorac Cardiovasc Surg 2003;126:1162-7.
21. Collins DJ, Johnson E, Kroshus T, Ganz Q, Batts K, Seng J, Nwaneri O, Dunn D. Experience with minimally
invasive oesophagectomy. Surg Endosc 2006;20:298-301.
22. Cotter G, Blatt A, Kaluski E, Metzkor-Cotter E, Koren M, Litinski I, et al. Conversion of recent onset
paroxysmal atrial fibrillation to normal sinus rhythm: the effect of no treatment and high-dose amiodarone.
A randomized, placebo controlled study. Eur Heart J 1999;20:1833-42.
23. Cybulski J, Ku³akowski P, Budaj A, Danielewicz H, Maciejewicz J, Kawka-Urbanek T, Ceremuzyñski L.

Intravenous amiodarone for cardioversion of recent-onset atrial fibrillation. Clin Card 2003;26:329-35.
24. Bernard EO, Schmid ER, Schmidlin D, et al. Ibutilide versus amiodarone in atrial fibrillation: A double-
blinded, randomized study. Crit Care Med 2003;31:1031-4.
25. Reinelt P, Karth GD, et al. Incidence and type of tachyarrhythmias in critically ill patients: a single centre
experience in a medical-cardiological ICU. Intensive Care Med 2001;27:1466-73.
26. Kanji S, Stewart R, Fergusson DA, McIntyre L, et al. Treatment of new-onset atrial fibrillation in non-
cardiac intensive care unit patients: a systematic review of randomized trials. Crit Care Med 2008;36:1620-
4.
27. United States Food and Drug Administration. Amiodarone hydrochloride prescribing information. Available
at: http://www.fda.gov/cder/drug/infopage/amiodarone/default.htm. Accessed December 2008.
28. Kochiadakis GE, Igoumenidis NE, Hamilos ME, Marketou ME, Chlouverakis GI, Vardas PE. A comparative
study of the efficacy and safety of procainamide versus propafenone versus amiodarone for the conversion
of recent-onset atrial fibrillation. Am J Cardiol. 2007;99:1721-5.
29. Boriani G, Mauro B, Capucci A, Botto G, Broffoni T, et al. Conversion of recent-onset atrial fibrillation
to sinus rhythm: Effect of different drug protocols. Pacing Clin Electrophysiol 1998;21:2470-4.
30. Mayr A, Ritsch N, Knotzer H, Dunser M, Schobersberger, Ulmer H, Mutz N, Haslbeder W. Effectiveness
of direct-current cardioversion for the treatment of supraventricular tachyarrhythmias, in particular atrial
fibrillation, in surgical intensive care patients. Crit Care Med 2003;31:401-5.
31. Neumayr G, Schratzberger P, Friedrich G, Gänzer H, Wiedermann CJ. Effect of electrical cardioversion on
myocardial cells in patients in intensive care. BMJ 1998;316(7139):1207-10.
32. Xanthos T, Prapa V, Papadimitriou D, Papadimitriou L. Comparative study of intravenous amiodarone
and procainamide in the treatment of atrial fibrillation of recent onset. Minerva Cardioangiol 2007;55:433-
41.
33. Martínez-Marcos FJ, García-Garmendia JL, Ortega-Carpio A, Fernández-Gómez JM, Santos JM, Camacho
C. Comparison of intravenous flecainide, propafenone, and amiodarone for conversion of acute atrial
fibrillation to sinus rhythm. Am J Cardiol 2000;86:950-3.
34. Donovan KD, Power BM. Intravenous flecainide versus amiodarone for recent-onset atrial fibrillation. Am
J Cardiol 1995;75:693-7.
35. Joseph AP, Ward MR. A prospective, randomized controlled trial comparing the efficacy and safety of
sotalol, amiodarone, and digoxin for reversion of new-onset atrial fibrillation. Ann Emerg Med 2000;36:1-
9.
36. Thomas SP, Guy D, Wallace E, Crampton R, Kijvanit P, Eipper V, Ross DL, Cooper MJ. Rapid loading of
sotalol or amiodarone for management of recent onset symptomatic atrial fibrillation: a randomized, digoxin-
controlled trial. Am Heart J 2004;147(1):E3.
37. Singh BN, Singh SN, Reda DJ, et al. Sotalol Amiodarone Atrial Fibrillation Effi cacy Trial (SAFE-T)
Investigators. Amiodarone versus sotalol for atrial fi brillation. N Engl J Med 2005;352:1861-72.
38. Kafkas NV, Patsilinakos SP, Mertzanos GA, Papageorgiou KI, Chaveles JI, Dagadaki OK, Kelesidis KM.
Conversion efficacy of intravenous ibutilide compared with intravenous amiodarone in patients with recent-
onset atrial fibrillation and atrial flutter. Int J Cardiol 2007;118:321-5.
39. Bernard EO, Schmid ER, Schmidlin D, et al. Ibutilide versus amiodarone in atrial fibrillation: A double-
blinded, randomized study. Crit Care Med 2003;31:1031-4.
40. Hennersdorf MG, Perings SM, et al. Conversion of recent-onset atrial fibrillation or flutter with ibutilide
after amiodarone has failed. Intensive Care Med 2002;28:925-9.
41. Dilaveris P, Synetos A, Giannapoulos G, Gialafos E, Stefanadis C. Conversion of recent onset atrial
fibrillation or flutter with amiodarone after ibutilide has failef: a rapid, efficient and safe algorithm. Ann
Noninvasive Electrocardiol 2005;10:382-6.
42. Karth D, Geppert A, Neunteufl T, Priglinger U, Haumer M, Gschwandtner M, Siostrzonek P, Heinz G.
Amiodarone versus diltiazem for rate control in critically ill patients with atrial tachyarrhythmias. Crit Care
Med 2001;29:1149-53.
43. Moran JL, Gallagher J, Peake S, Cunningham D, Salagras M, Leppard P. Parenteral magnesium sulphate
versus amiodarone in therapy of atrial tachyarrhythmias: A prospective randomized study. Crit Care Med
1995;23:1816-24.
44. Sleeswijk ME, Tulleken JE, Noord TV, Meertens JHJM, Lightenberg JJM, Zijlstra JG. Efficacy of magnesium-
amiodarone step-p scheme in critically ill patients with new onset atrial fibrillation: A prospective
observational study. J Intensive Care Med 2008;23:61-7.
45. Onalan O, Crystal E, Daoulah A, Lau C, Crystal A, Lashevsky I. Meta-analysis of magnesium therapy for
acute management of rapid atrial fibrillation. Am J Card 2007;99:1726-32.
46. Xu HK, Zhang YF. Comparison between therapeutic effects of acupuncture and intravenous injection of
amiodarone in the treatment of paroxymal atrial fibrillation and atrial flutter. Zhongguo Zhen Jiu.
2007;27:96-8.
47. Bobbio A, Caporale D, INternullo E, Ampollini L, Bettati S, Rossini E, et al. Postoperative outcome of
patients undergoing lung resection presenting with new-onset atrial fibrillation managed by amiodarone
or diltiazem. Eur J Cardio-thoracic Surg 2007;31:70-4.
48. Tuseth V, Jaatun J, Dickstein K. Amiodarone infusion in the treatment of acute atrial fibrillation or flutter:
high versus low dose treatment. Heart 2005;91:964-5.
49. Slim AM, Roth JE, Boyd SY, Rubal BJ. The incidence of phlebitis with intravenous amiodarone at guideline
dose recommendations. Mil Med. 2007;172:1279-83
50. Xanthos T Intravenous and oral administration of amiodarone for the treatment of recent onset atrial
fibrillation after digoxin administration. Int J Cardiol 2007;121:291-5.
51. Holt AW. Hemodynamic response to amiodarone in critically ill patients receiving catecholamine infusions.
Crit Care Med. 1989:17;1270-6.
52. Kumar A. Intravenous amiodarone for therapy of atrial flutter and fibrillation in critically ill patients with
severely depressed left ventricular function. South Med J. 1996;89:779-85.
53. Skroube T, Galliatsou E, et al. Amiodarone induced acute lung toxicity in an ICU setting. Acta Anaesthesia
Scand 2005;49:569-71.
54. Olshansky B, Magdi S, Rubin A, Kostis J, Shorofsky S, Slee A, Greene L and NHLBI AFFIRM investigators.
Use of amiodarone for atrial fibrillation in patients with pre-existing pulmonary disease in AFFIRM study.
Am J Card 2005;95:404-5.
55. Schrickel JW, Schwab JO, Yang A, et al. “Torsade de pointes” in patients with structural heart disease
and Atrial Fibrillation Treated with Amiodarone, â-Blockers, and Digitalis. Pacing Clin Electrophysiol.
2006;29:363-6.
56. Kounas SP, Letsas KP, et al. QT interval prolongation and torsades de pointes due to a coadministration
of metronidazole and amiodarone. Pacing Clin Electrophysiol. 2005;28:472-3.
57. Antonelli D, Atar S, et al. Torsade de pointes in patients on chronic amiodarone treatment: contributing
factors and drug interactions Isr Med Assoc J. 2005;7:163-5.
58. Vassallo P, Trohman RG, et al. Prescribing amiodarone: an evidence based review of clinical indications.
JAMA 2007;298:1312-22.
59. Hsieh HJ, Fatal amiodarone-induced hepatotoxicity: a case report and literature review. Int J Clin Pharmacol
Ther. 2008;46:96-101.
60. Korantzopoulos P, Pappa E. Acute low back pain during intravenous administration of amiodarone: a
report of two cases. Int J Cardiol. 2005;98:355-7.
Six
Defibrillation during CPR—Is a
Delayed, Single, Biphasic Shock Better
Ram E Rajagopalan, T Prithviraj
INTRODUCTION
Defibrillation is a life-saving procedure in the resuscitation of the victim of sudden cardiac
arrest presenting with either ventricular fibrillation or pulseless ventricular tachycardia (VF/
VT). Typically, cardiopulmonary resuscitation (CPR) guidelines published before 2005
emphasized the time-dependence of successful defibrillation and advised a shock-first strategy
for all VF/VT arrest. It was emphasized that performing routine CPR before defibrillation would
only delay the delivery of shock and reduce the probability of successful restoration of a
perfusing rhythm. Similarly, with the emphasis being on early defibrillation, it was typical to
recommend multiple (stacked) shocks to enhance the restoration of perfusion, based on the
belief that there was a significant incremental benefit from sequential shocks, mediated by a
progressive reduction in thoracic impedance with each stacked shock.
In the last decade, with increased public awareness of bystander CPR, better engineering
of the defibrillators and the availability of automated options, early access to defibrillation has
increased dramatically and a number of basic science and clinical studies have refined our
understanding of this procedure. As a result, the 2005 guidelines of the ILCOR, American
Heart Association and European Resuscitation Council, among others, have made a radical
departure from each of the traditional recommendations in handling VF/VT arrest. Defibrillation
is now delayed to allow CPR especially in unwitnessed events and the putative benefits of
consecutive shocks are now recognized to be so marginal that a single defibrillatory shock is
considered to be adequate, especially while using a biphasic defibrillator. This review is an
evaluation of the science behind these new proposals.
CPR BEFORE DEFIBRILLATION – SHOULD DEFIBRILLATION BE DELAYED

The traditional argument to support a shock-first approach to VF/VT has been based on the
observation that responsiveness to defibrillation is time-sensitive.1 As each minute of delay in
delivering a defibrillatory shock reduces the probability of attaining a perfusing rhythm by
about 10%, defibrillation after 10 minutes will not restore the rhythm in any patient. Given
Defibrillation during CPR—Is a Delayed, Single, Biphasic Shock Better 57
this temporal relationship it is presumed that any process that delays defibrillation, including
the administration of cardiac compressions, will reduce the chance of a good outcome and
defibrillation is recommended as the first procedure when VF/VT is recognized. Many recent
critical assessments view this one-size-fits-all approach of immediate defibrillation of all patients
with a VF/VT (irrespective of the duration of circulatory arrest) as inflexible and incorrect.
Weisfeldt and Becker eloquently highlight this concern in their 2002 clinical commentary,2
arguing that VF/VT arrest must be considered a three-phase time-sensitive event, in which
the temporal evolution of the pathophysiology itself warrants a variation in resuscitative
approaches as time progresses. Based on strong experimental data, their model proposes
that the first 4 minutes after VF/VT arrest be considered an “Electrical Phase”. Animal studies
imply that circulatory arrest for such brief duration of time carry no significant metabolic
consequences and defibrillation alone enhances the probability of a return of spontaneous
circulation (ROSC) in a time-dependent manner. This contrasts with the observations in
prolonged arrest (4-10 minutes), when the consequences of tissue oxygen deprivation become
relevant and defibrillatory success pivots on the ability to improve circulation and tissue
oxygenation. Thus, in this “Circulatory Phase” after cardiac arrest, defibrillation alone is
unsuccessful, unless it is preceded by atleast 2 to 3 minutes of good cardiopulmonary
resuscitation aimed at resolution of tissue hypoxia. When the arrest interval exceeds this time
limit (>10 minutes) the victim enters a “Metabolic Phase” when both the circulatory arrest
and the ensuing reperfusion may propagate a widespread inflammatory process that amplifies
tissue injury beyond focal ischemia and may progress to an irreversible state. Experiments
indicate that neither CPR nor defibrillation is effective beyond this point, though small clinical
studies imply that heroic attempts at controlled reperfusion using cardiopulmonary bypass
and aggressive metabolic manipulation may be of some benefit.
While the primary difficulty in applying these concepts to clinical practice may lie in our
inability to estimate arrest duration accurately, this model is clinically relevant in that it suggests
a flexible approach in which witnessed or in-hospital events may be presumed to be in the
“electrical phase” necessitating immediate defibrillation, while delayed, unwitnessed or out-
of-hospital events require CPR first to enhance the success of defibrillation. There has been
an increasing amount of data to support such an approach in humans in the last 5 years.
The initial observational study of Cobb3 and colleagues was an evaluation of outcomes
before and after a change of resuscitation protocol in the emergency medical services in
Seattle. The study revealed that survival and early neurological recovery was better after the
standard ALS shock-first protocol was modified to include 90 seconds of CPR prior to
defibrillation. Subgroup analysis demonstrated that the differences were significant only in
patients in whom the paramedic response was delayed by 4 minutes or longer. This study
justifies the need to alter the resuscitation strategy based on the time of its initiation.
Confirmatory evidence was provided by a randomized controlled trial (Wik et al, 20034)
performed on 200 patients with out-of-hospital ventricular fibrillation. Though the 22% survival
to hospital discharge in the 104 patients randomized to receive CPR before defibrillation did
not differ significantly from the 15% survival in patients on the traditional defibrillation-first
protocol (p = 0.17), patients in whom response times exceeded 5 minutes had higher rates
of ROSC (58% vs. 38% p=0.04), survival to hospital discharge (22% vs. 4%; p=0.006) and
1 year survival (20% vs 4%; p=0.01) when CPR preceded defibrillation. Three minutes (180
seconds) of CPR was performed before defibrillation in this study in contrast to the
90 second protocol employed by Cobb.
Unfortunately, the value of pre-defibrillation CPR was not corroborated by a subsequent
randomized controlled trial from Australia (Jacobs et al 20055) that evaluated the effect of
90 seconds CPR prior to defibrillation, on survival to discharge, ROSC and survival at 1 year
in 256 non-paramedic witnessed VF/VT arrests. There was no difference in survival (4.2 vs.
5.1%) or in ROSC. While it is tempting to postulate that this lack of success is attributable to
the shorter duration of pre-defibrillation CPR, factors related to the quality of chest compression
may have been equally important. Evidence for this comes from a large range of experimental
and clinical studies.
Interruption of chest compression is frequent during CPR for the provision of ventilation
and for the evaluation of the pulse and rhythm. In a preclinical study done on swine by Yu
et al,6 frequent interruptions in chest compression for rhythm analysis was associated with a
greater chance of failure to resuscitate. Longer interruptions were associated with poor coronary
flow during resuscitation, as indicated by lower coronary perfusion pressures and an increased
frequency of myocardial dysfunction after ROSC, emphasizing the need to keep interruptions
of CPR for pulse and rhythm checks to a minimum.
Interruption of CPR for ventilation also appears to impact outcome negatively in this setting.
A series of animal studies by Berg, Ewy and associates7,8 starting with demonstrations of the
inefficacious artificial respiration in CPR,7 went on to establish that neurologically intact survival
was dependent on the continuity of circulatory assistance and not pH, oxygen saturation or
the absolute levels of minute ventilation.8 These studies suggest that pre-defibrillation CPR
focusing on continuous chest compression is more efficacious than traditional CPR that is
characterized by interruptions for ventilation, pulse checks and rhythm evaluations.
Clinical confirmation of these experimental data comes from the studies of Eftestol9,10 and
colleagues evaluating fibrillatory waveforms obtained during CPR. They showed that during
CPR favourable effects on VF waveform (which predict successful defibrillation) were
maximized when CPR was administered for durations greater than three minutes, for a scenario
where ambulance response time exceeded five minutes and when interruptions of CPR were
minimal.
Taken together, these studies strongly suggest that the success of pre-defibrillation CPR in
the “circulatory phase” is highly dependent on the quality of chest compression, which in
turn, is primarily influenced by a minimization of interruptions. It is entirely possible that the
heterogeneous effects on survival seen in clinical studies of pre-defibrillation CPR4,5 are directly
influenced by these qualitative disparities. In addition, clinical studies (Van Alem et al11 2003)
have also demonstrated that a perfusing rhythm is seldom obtained in the first few minutes
after defibrillation, which makes CPR after defibrillation an essential means of preserving
organ perfusion. Moreover, as the immediate resumption of chest compressions after shock
is unlikely to provoke adverse rhythms,12 resumption of chest compressions should not be
delayed.
In the light of the above facts, Ewy and colleagues have proposed a protocol for minimally
interrupted cardiac resuscitation (MICR) also termed as cardio-cerebral resuscitation13 as a
means of standardizing the quality of CPR before and after defibrillation. This protocol includes
“an initial series of 200 uninterrupted chest compressions, rhythm analysis with a single shock,
200 immediate post-shock chest compressions before pulse check or rhythm reanalysis, early
administration of epinephrine, and delayed endotracheal intubation”. In a recent prospective
observational study, comparing outcome before and after the introduction of an MICR protocol,
these authors demonstrated a clear benefit in survival to hospital discharge especially in patients
with ventricular fibrillation (4.7% vs. 17.6% after MICR training; OR = 8.6; 95% CI,
1.8-42.0). Neurological function amongst survivors was also improved in patients administered
MICR.
In summary, clinical evidence supporting delayed defibrillation to allow pre-shock CPR is
growing. The dichotomous results of early clinical trials could be explained by qualitative
differences in the CPR administered before defibrillation. A growing body of evidence implies
that improving CPR quality by reducing interruptions for rhythm/pulse checks or even for
artificial ventilation will translate into improved survival. It would be difficult to claim that
current evidence is overwhelmingly in favour of pre-defibrillation CPR, though it appears to
be evolving in that direction.
SINGLE SHOCK VS. 3 STACKED SHOCKS

With the current emphasis on the need to minimize interruptions during CPR, it would appear
logical that ROSC could be maximized by reducing the “hands-off” time during defibrillation.
In direct contradistinction to this perspective, it has been felt for many years that multiple
(usually three) consecutive shocks may significantly enhance the success of defibrillation.
Consecutive shocks progressively reduce transthoracic impedance (TTI) and result in the
delivery of larger currents to the fibrillating myocardium. As there are no trials in humans that
clearly compare the value of one versus multiple defibrillations; this recommendation is simply
extrapolated from other experimental and laboratory data.
Most studies demonstrate that the first shock provides the greatest chance of converting
VF/VT to a perfusing rhythm and the incremental benefit of subsequent shocks is small. For
example, in the ORCA comparison of monophasic and biphasic defibrillation,14 though the
initial monophasic shock was successful in only 66%, 3 consecutive shocks increased the
probability of success by a meagre additional 7%, while with biphasic defibrillation success
increased from 92 to 96% only. Such studies imply that the marginal benefits obtained from
consecutive defibrillation attempts may not translate into adequate changes in outcome to
justify the extra hands-off time that the process entails.
Similarly, in a retrospective analysis of cardiac arrest data from two EMS systems15 questions
were raised about the influence of TTI on defibrillation success, ROSC and survival to discharge
after out-of-hospital cardiac arrest. Short arrest-to-shock intervals rather than variations in
TTI correlated with success in these studies. It is, thus fair to conclude that defibrillation success
is unrelated to TTI and is not significantly enhanced by “stacking” the shocks.
More recently, a one-shock protocol was compared with 3 consecutive shocks in a porcine
model.16 In comparison to the conventional 3-stacked shocks, the one-shock protocol reduced
CPR interruption from 45 to 34% of the total resuscitation time and increased survival from
64% to 100%. So, despite the absence of the highest quality clinical evidence, secondary
data does, in fact, favour the use of single-shock protocols, especially when biphasic defibril-
lation assures high initial rates of success.
IS BIPHASIC BETTER THAN MONOPHASIC DEFIBRILLATION

Introduced more than 10 years ago, the biphasic defibrillator has slowly replaced monophasic
equipment. Although the physiological benefits of biphasic defibrillation were obvious, it was
unclear if they offered a significant advance over monophasics in the clinical setting. The
optimal energy required for biphasic defibrillation is significantly lower than that required for
monophasics and the resultant attenuation of myocardial injury was postulated to translate
into better post-resuscitation cardiac function.17 However, few clinical trials showed an
unequivocal benefit of biphasic defibrillation in human resuscitation both in the basic-life
support / AED setting or when used in advanced life-support scenarios.
More recently, in the first published randomized controlled trial,14 150J biphasic shocks
were compared to 200-360 J monophasic shocks in out-of hospital cardiac arrest in 115
patients. All patients were evaluated for defibrillation success, ROSC, survival to hospital
admission and discharge and neurological status at discharge. Though defibrillatory success,
ROSC and hospital admission rates were higher with biphasic shocks, survival to discharge
was not significantly different. However, among survivors, biphasic shocks resulted in better
preservation of neurological status. Similar results (showing defibrillatory efficacy, but no
difference in survival to discharge) were demonstrated in a later study18 in the same setting
of BLS/AED use. In the setting of advanced life support resuscitation, the ORBIT19 and
TIMBER20 trials could only demonstrate superior first-shock success rates with the use of
biphasic defibrillation, but no effect on any of the clinical end points of ROSC or survival to
hospital discharge.
The lack of effect on live discharge rates is probably because these trials are small and
underpowered. Despite its consistent efficacy in defibrillation at lower energy levels, it is difficult
to draw universal conclusions about the ideal biphasic energy settings or the ideal waveform
pattern, as there are no comparative studies of the various biphasic devices. Additionally,
many of the monophasic-biphasic comparisons have been studied under what might be
considered suboptimal resuscitation protocols that neither paid attention to pre-resuscitation
CPR nor to the qualitative issues discussed earlier in this paper. Furthermore, one must always
be concerned about biased reporting when there are very few industry-independent evaluations
in the literature.
CONCLUSION
After viewing all the current evidence one can make a reasonable case in favour of a flexible
approach to a VF/VT arrest, with attention being paid to pre-defibrillation CPR in select indivi-
duals. However, more evaluations are needed to confirm the genuine role of low interruption
high quality CPR in this setting. It is rather unfortunate that there are many questions left
unanswered about the “superiority” of biphasic technology, even as monophasic machines
are being prematurely withdrawn from the market. In conclusion, it must be conceded that
the current literature is insufficient to draw hard conclusions about the advantages of biphasic
defibrillation, though there are reasonable physiological arguments to favour this new
technology.
REFERENCES
1. Valenzuela T, Roe D, Cretin S, Spaite D, Larsen M. Estimating effectiveness of cardiac arrest interventions.
Circulation 1997;96:3308-13.
2. Weisfeldt ML, Becker LB. Resuscitation after cardiac arrest: a 3-phase time-sensitive model. JAMA
2002;288:3035-8.
3. Cobb LA, Fahrenbruch CE, Walsh TR, et al. Influence of Cardiopulmonary Resuscitation prior to
defibrillation in patients with out-of-hospital ventricular fibrillation. JAMA 1999;281(13):1182-8.
4. Wik L, Hansen TB, Fylling F, et al. Delaying defibrillation to give basic cardio-pulmonary resuscitation to
patients with out-of-hospital ventricular fibrillation: A randomized trial. JAMA 2003;289(11):1389-95.
5. Jacobs IG, Finn JC, Oxer HF, Jelinek GA. CPR before defibrillation in out-of-hospital cardiac arrest: A
randomized trial. Emergency Medicine Australasia 2005;17:39-45.
6. Yu T, Weil MH, Tang W, et al. Adverse outcomes of interrupted precordial compression during automated
defibrillation. Circulation 2002;106:368-72.
7. Robert A Berg, Arthur B Sanders, Karl B Kern, Ronald W Hilwig, Joseph W Heidenreich, Matthew E
Porter, et al. Adverse Hemodynamic effects of interrupting chest compressions for rescue breathing during
cardiopulmonary resuscitation for ventricular fibrillation cardiac arrest. Circulation 2001;104:2465-70.
8. Karl B Kern, Ronald W Hilwig, Robert A Berg, Arthur B Sanders, Gordon A Ewy. Importance of continuous
chest compressions during cardiopulmonary resuscitation: Improved outcome during a simulated single
lay – rescuer scenario. Circulation 2002;105:645-9.
9. Eftestol T, Wik L, Sunde K, Steen PA. Effects of cardiopulmonary resuscitation on predictors of ventricular
fibrillation defibrillation success during out-of-hospital cardiac arrest. Circulation 2004; 110:10-5.
10. Eftestol T, Sunde K, Steen PA. Effects of interrupting precordial compressions on the calculated probability
of defibrillation success during out-of-hospital cardiac arrest. Circulation 2002;105: 2270-3.
11. van Allem AP, Sanou BT, Koster RW. Interruption of Cardiopulmonary Resuscitation with the use of the
automated external defibrillator in out-of-hospital cardiac arrest. Ann Emerg Med 2003;42:449-57.
12. Erik P. Hess, Roger D. White. Ventricular fibrillation is not provided by chest compression during post-
shock organized rhythms in out-of-hospital cardiac arrest. Resuscitation 2005;66:7-11.
13. Bobrow BJ, Clark LL, Ewy GA, Chikani V, Sanders AB, Berg RA, et al. Minimally Interrupted Cardiac
Resuscitation by Emergency Medical Services for Out-of-Hospital Cardiac Arrest JAMA 2008;299(10):1158-
65.
14. Schneider T, Martens PR, Paschen H, et al. Multicenter, randomized, controlled trial of 150J biphasic
shocks compared with 200 to 360J monophasic shocks in the resuscitation of out-of-hospital cardiac
arrest victims. Circulation 2000;102:1780-7.
15. White RD, Blackwell TH, Russell JK, et al. Transthoracic impedance does not affect defibrillation,
resuscitation or survival in patients with out-of-hospital cardiac arrest treated with a non-escalating biphasic
waveform defibrillator. Resuscitation 2005;64(1):63-9.
16. Tang W, Snyder D, Wang J, et al. One-shock versus three-shock defibrillation protocol significantly improves
outcome in a porcine model of prolonged ventricular fibrillation cardiac arrest. Circulation 2006;113:2683-
9.
17. Faddy SC, Powell J, Craig JC. Biphasic and monophasic shocks for transthoracic defibrillation: a meta
analysis of randomised controlled trials. Resuscitation 2003;58:9-16.
18. van Alem AP, Chapman FW, Lank P, et al. A prospective, randomized and blinded comparison of first
shock success of monophasic and biphasic waveforms in out-of-hospital cardiac arrest. Resuscitation
2003;58(1):17-24.
19. Morrison LJ, Dorian P, Long J, et al. Out-of-hospital cardiac arrest rectilinear biphasic to monophasic
damped sine defibrillation waveforms with advanced life support intervention trial (ORBIT). Resuscitation
2005;66(2):149-57.
20. Kudenchuk PJ, Cobb LA, Copass MK, et al. Transthoracic incremental monophasic versus biphasic
defibrillation by emergency responders (TIMBER): A randomized comparison of monophasic with biphasic
waveform ascending energy defibrillation for the resuscitation of out-of-hospital cardiac arrest due to
ventricular fibrillation. Circulation 2006;114:2010-8.
Importance of Assessing the Microcirculation in an ICU Patient 63
Seven
Importance of Assessing the
Microcirculation in an ICU Patient
Saxon Ridley
INTRODUCTION
Oxygen delivery is crucial for cellular homeostasis. A deficient oxygen supply quickly leads
to cellular and then organ dysfunction. A properly functioning microcirculation is the last vital
stage in the transport of oxygen from the inhaled gases to the mitochrondria. The aim of this
chapter is to review the physiology and pathophysiology of the microcirculation, recognise
dysfunction (signs and symptoms) and review the methods for measuring microcirculatory
flow and the treatment options.
PHYSIOLOGY
The microcirculation has two main roles, the delivery of oxygen and nutrients and regulation
of immune function. It comprises of a network of arterioles, capillaries and venules; all of
these vessels are less than 100 mm in diameter. The endothelium lining the microcirculation
is the largest surface area in the body. The exact structure and function of the microcirculation
is very variable depending upon the individual organ but the general physiology of the
microcirculation has three important aspects. First are the components, which are the
endothelial, smooth muscle and red blood cells, leucocytes and plasma elements (e.g.
coagulation factors, cytokines and chemokines). The endothelial cells are covered by a layer
of glycocalyx which forms an important barrier and transduction system between the lumen
of the capillaries and the endothelium. Second is the flow control by arteriolar tone determined
by smooth muscle contraction, cellular rheology and capillary patency. Third is regulation,
which may be myogenic (sensing strain and stress), metabolic (based upon oxygen and CO2
tensions, lactate and pH) or neurohumoral (both autocrine and paracrine). The microcirculation
is a complex network of resistance and exchange vessels, where perfusion is dependent upon
numerous factors, the most important of which are the local driving pressure, arteriolar tone,
cellular rheology and capillary patency.
PATHOPHYSIOLOGY
Systemic inflammation, of which sepsis is the most severe, affects every component of the
microcirculation, impairing its regulatory function and resulting in regional mismatch of oxygen
supply and demand.1 These alterations are more severe in non-survivors than survivors2 and
persistent microvascular alterations are associated with the development of multiple organ
failure and death.3 Depending upon the type and magnitude of the primary insult (e.g. trauma,
sepsis, haemorrhage etc), the presence or absence of accompanying co-morbidities and the
timing of initial effective resuscitation, circulatory shock and inflammation can trigger activation
of the microcirculation. The endothelial cells increase their production of inducible nitric oxide
synthase. As this induction is variable, the production of nitric oxide is different in various
organ beds so resulting in pathological shunting of blood flow. By contrast, those vascular
beds where nitric oxide synthase is not expressed do not vasodilate and consequently become
under perfused. The endothelium also expresses various adhesion molecules such as P selectin
initially and then later intracellular and vascular cell adhesion molecules. The endothelial cells
lose their anionic charge and glycocalyx and so become leaky. Disturbed gap junctions interfere
with endothelial cell-to-cell communication and so interrupt upstream regulation. There is
disturbed signal transduction and electrophysiological breakdown resulting in the loss of smooth
muscle control. Smooth muscle cells also lose their sensitivity to adrenergic stimuli and hence
their tone. However, the microcirculation may still respond appropriately to some signals such
as acetylcholine. Red blood cells lose their usual deformability and become prone to aggregation
both of which increase viscosity and lead to microvascular plugging. Red blood cells also
release nitric oxide and ATP when hypoxic.4 Activated leucocytes release cytokines and reactive
oxygen species, which disrupt microvascular structures, cellular components and coagulation
function. They also become less deformable and increase their levels of cognate integrin
receptors to improve adhesion. Tissue cells lose their cell junctions and in the plasma there
is activation of coagulation and fibrin deposition. All these contribute to microcirculatory
dysfunction and shunting which precipitates cellular distress and organ failure (Table 7.1).
At the capillary level, loss of regulation reduces the functional capillary density leading to
microvascular derecruitment. Individual capillaries become either functionally occluded by
cellular aggregates, intense vasoconstriction or inadequate perfusion pressure. Heterogeneous
flow abnormalities in capillary beds cause vulnerable vascular areas to become hypoxic (where
the local tissue oxygen tension (PtO2) is less than PvO2). ‘Lethal corners’ may develop towards
the venous end of the capillary bed where the diffusion distance for oxygen between adjacent
capillaries becomes too great. As the capillary bed becomes increasingly truncated and so
functionally deficient, the diffusion distance for oxygen increases further so extending the
area of the ‘lethal corner’ (Fig. 7.1).
The precise origin of the oxygen extraction deficit in sepsis is not clear but it may be a
combination of pathological shunting, hypoxic areas in the ‘lethal corners’ or an inability of
the mitochondria to use oxygen. Under normal conditions, ATP production by mitochondrial
Table 7.1: Pathophysiological consequences of microvascular dysfunction

Endothelial activation Altered red and white cell rheology
Viscosity alteration
Disseminated intravascular coagulation
Altered microvascular flow and resistance Vasoplegia

Opening of arteriovenous shunts
Intravascular pooling
Redistribution of organ blood flow
Cardiorespiratory dysfunction
Increased microcirculatory permeability Oedema formation

Vascular and tissue congestion
Fig. 7.1: The increasing size of the ‘lethal corner’ where PtO2 becomes critical as microcirculatory derecruitment
increases the diffusion distance for oxygen (Modified from Keogh A. The number and distribution of capillaries
in muscles with calculations of the oxygen pressure head necessary for supplying the tissue. J Physiol 1919 May
20; 52:409-15)
oxidative phosphorylation accounts for more than 90% of total oxygen consumption. In muscle
biopsies from septic patients, Brealey found nitric oxide overproduction, glutathione (an
intracellular antioxidant) depletion, mitochondrial dysfunction (as measured by respiratory
chain activity) and decreased ATP concentrations suggesting that there may be a primary
failure of biological fuel generation5. These changes were most notable in patients with organ
failure and eventual non-survivors. Persistent regional and microcirculatory dysfunction leading
to local hypoxic areas has been termed ‘Microcirculation and Mitochondrial Distress Syndrome
(MMDS)’.6
SIGNS AND SYMPTOMS

The macrocirculatory haemodynamic variables of hypotension (due to arterial vasodilation
and loss of vascular resistance) and hypovolaemia (as endothelial leakage leads to oedema
formation and vascular contraction) are not sensitive enough for the detection of microvascular
dysfunction. Heterogeneity of flow uncouples regional and local flow from systemic measures
of flow and oxygen consumption such that microcirculation dysfunction is poorly reflected in
global measures of oxygen delivery and consumption. If severe, SvO2 will fall but a normal
SvO2 does not exclude microvascular decruitment. Ince has proposed a shunting theory to
explain this apparent paradox whereby the shutting down of vulnerable vascular beds shunts
oxygen rich blood from the arterial to the venous system so maintaining the SvO2 despite
microcirculatory hypoxia.7 The venous blood returning from the hypoxic areas will contain
higher CO2 and lactate levels. Paradoxically vasodilation in the hypoxic capillary beds may
improve the driving capillary perfusion pressure and so improve oxygen delivery (see later).
The signs and symptoms of microcirculatory dysfunction are non-specific; they generally
reflect organ dysfunction and therefore include hypotension, tachycardia, oliguria,
encephalopathy, cool peripheries, slow capillary refill and lactate acidosis. Lactate acidosis is
a non-specific measure of an anaerobic metabolism as it is also elevated in diabetes, liver
dysfunction, tissue reperfusion, catecholamine infusion and pyruvate dehydrogenase inhibition.
However, it does have some prognostic value and can be used as a treatment trigger (e.g. in
early goal directed therapy). Spronk6 has developed a scoring system to warn that the
conditions for microcirculatory dysfunction are present. The scoring system is based on
haemodynamic variables, the state of the peripheral circulation, microvascular variables (e.g.
tonometric CO2 measurement), systemic markers of tissue oxygenation and organ dysfunction
(Table 7.2).
ASSESSING MICROVASCULAR FLOW ABNORMALITIES

The development of orthogonal polarization spectral imaging has allowed direct observation
of microcirculatory haemodynamics in humans. Orthogonal polarization spectral imaging is
an optical technique, which uses a microscope for visualizing the microcirculation in the surface
mucosa using polarized green light and crossed-polarized images. The technique is based on
the principle that green light (wavelength 530 nm) illuminates tissue to a depth of upto 3 mm.
In the superficial vessels, the scattered green light is absorbed by the haemoglobin in the red
cells (which so appear dark) and scattered by the leucocytes (which appear green). This allows
both the capillaries and venules to be visualized because they both contain red blood cells.
Table 7.2: Score for presence of microvascular dysfunction

(a score over 2 suggests microvascular derecruitment)
Variable Score Parameter thresholds
Haemodynamic variables 2 Heart rate >100 bpm

MAP < 50 mmHg or CVP <2 or > 15 mmHg
Cardiac index <2.2 l/min/m2
Peripheral circulation 2 Mottled skin

Core–peripheral Temperature difference >5oC
Impaired capillary refill
Microvascular variables 1 Increased tonometric CO2 gap

Increased sublingual CO2 gap
Systemic markers of tissue 1 Lactate >6 mmol/l

oxygenation
SvO2 <60%
Organ Dysfunction 1 Oliguria at <0.5 ml/kg/min

1 Decreased mental state
Using this technique, De Backer8 has shown that in health the capillaries are equally distributed
between the tissues (i.e. a uniform and adequate functional capillary density) whereas in
diseased states there is marked heterogeneity in microcirculatory flow with some beds having
preserved, sluggish or even no flow.
An improved optical modality in terms of technology and image quality is sidestream
dark-field imaging. This uses light-emitting diodes placed around the tip of a light guide with
a centre core optically isolated from the outer rim. When the light guide is placed on tissues
surfaces, light from the outer rim penetrates the tissues illuminating the microcirculation from
the interior. This dark field illumination completely removes reflections from the tissue surfaces
and so yields clearer images of the microcirculation including flowing red and white cells.
CLASSIFYING MICROCIRCULATORY DISTURBANCES

Elbers and Ince9 have used orthogonal polarization spectral and sidestream dark field imaging
to investigate and classify capillary flow under different conditions. Although the capillary
abnormalities are caused by different mechanisms, they all have a common distributive defect
caused by the functional shunting of capillaries in the presence of normal or hyperdynamic
venular flow. Qualitatively describing the flow in capillaries and venules, they proposed five
classes of abnormality. Class I abnormalities define all capillaries being stagnant despite normal
or sluggish flow in the venules. This is seen in patients resuscitated with vasopressors to
normalise the blood pressure. Class II abnormalities are characterized by empty capillaries
adjacent to capillaries with normal flow. This decrease in capillary density increases the diffusion
distance between the remaining capillaries (i.e. leading to the development of ‘lethal corners’).
The Sa02 in the capillaries with flow is high, suggesting poor oxygen off-loading as a result of
decreased exchange area. Class II abnormalities are most commonly seen during the use of
extracorporeal circuits such as cardiac bypass and membrane oxygenation. Class III
abnormalities describe capillaries with stagnant flow next to those with normal flow; this
abnormality is most commonly seen in patients with sickle cell disease and cerebral malaria.
Class IV abnormalities occur when there are capillaries with stagnant flow next to those with
hyperdynamic flow. The venules may also have hyperdynamic flow and this condition is
frequently seen in resuscitated hyperdynamic septic patients. Class V abnormalities occur
when there is hyperdynamic flow in all the vessels of the microcirculation such that individual
cells are travelling so fast they cannot be separately identified. These abnormalities occur in
septic patients but also during extreme exercise.
The various classes of abnormal capillary and venular flow are not mutually exclusive and
can occur simultaneously in different parts of the microcirculation. These abnormalities arise
as the combined result of regional pathological changes in the vessel wall and red and white
cells, the systemic haemodynamic profile at the time and any cardiovascular moderating
treatments applied. For example, vasopressors will raise the systemic arterial blood pressure
but will compromise microcirculatory flow10 (Type I abnormality). Hyperoxia can lead to
arteriolar vasoconstriction and hence a possible reduction in capillary functional density11
(Class II abnormality). Haemodilution not only decreases blood viscosity but also altered red
blood cell rigidity. Decreased viscosity reduces the bloods resistance to flow and consequently
the pressure gradient along the capillary may decrease which paradoxically leads to a fall in
capillary flow (Class II abnormality). Alterations in cellular rigidity can also lead to obstruction
in capillary flow leading to Class I, II and IV abnormalities12. The variable expression of inducible
nitric oxide synthase will cause local vasodilation and consequently local hyperdynamic flow
(Class IV and/or V abnormalities).
QUANTIFYING MICROCIRCULATORY FLOW

Acquiring images of the microcirculation is undoubtedly the best way to test the effects of
therapeutic interventions. However, to differentiate between normal and diseased states and
test the effects of treatments, it is important to standardise the methods for image acquisition
and then generate accurate measurements of microcirculatory flow. Leading experts in
microcirculatory flow13 have suggested that when acquiring images five sites of interest should
be examined to ensure a comprehensive study. For human investigations a 5 x microscope
objective should be used so that the field of interest is not too narrow. Pressure of artefacts,
which may interfere with flow should be minimised by allowing the probe to only just make
contact with the organ surface. High quality video images are suitable for subsequent analysis.
Microcirculatory flow can then be quantified using a number of valid and reliable measures.
Vessel density can be calculated as the number of vessels crossing a grid of equidistant horizontal
and vertical lines divided by the total length of the lines. Using 20 seconds of flow as a
threshold, perfusion can then be categorised as continuous, absent or intermittent (i.e. flow
for atleast 50% of the time). The proportion of perfused vessels can then be calculated as
100 x (the total number of vessels minus those vessels with no or intermittent flow) / the total
number of vessels. The perfused vessel density, which is an estimate of functional capillary
density, can be calculated by multiplying the vessel density by the proportion of perfused
vessels. The microvascular flow index score is based on determining the predominant type of
flow in four quadrants of the microscope field. Flow is scored as absent (0), intermittent (1),
sluggish (2), or normal (3). The average value of the four quadrants is the microvascular flow
index. As heterogeneity in microcirculatory flow is one of the key pathological findings,
quantifying such heterogeneity is valuable. If the microvascular flow index is measured in
three to five sites, the difference between the highest and lowest value divided by the mean
flow velocity gives a good index of heterogeneity as it takes into effect extreme deviation.
Rather than directly observing microvascular flow, the physiological effects of changes in
microvascular flow can be quantified by measuring tissue PCO2 (PtCO2). Arterial CO2 (PaCO2),
regional blood flow, microcirculatory flow and tissue CO2 production are the main factors
controlling PtCO2. Under stable conditions when PaCO2 is constant, tissue CO2 reflects the
balance between its production and clearance. Adequate perfusion results in a constant tissue
CO2 because it is washed out as it is produced by aerobic metabolism. However, because
PaCO2 may vary with cardiorespiratory performance, the difference between regional and
PaCO2 (known as the PCO2 gap (PgapCO2)) more accurately reflects tissue CO2. If oxygen
delivery decreases to a critical level, metabolism switches from aerobic to anaerobic and at
this stage, tissue CO2 will start to climb as CO2 is generated by the buffering of lactic acid by
bicarbonate ions. Using mathematical models, Gutierrez14 suggested that raised tissue CO2 is
a marker of tissue hypoperfusion. In hypoxic hypoxia, where blood flow is preserved but
tissue oxygenation is inadequate, tissue CO2 remains relatively constant because of adequate
blood flow. However, in stagnant hypoxia where flow decreases, increases in tissue CO2 should
be expected. This hypothesis proposing that microcirculatory flow is a major factor in
determining tissue CO2 has been supported by animal15 and human studies.16
Monitoring gastric tissue CO2 was developed as a clinically useful measure of global micro-
circulatory flow as the splanchnic circulation is more vulnerable to ischaemia and has a higher
critical oxygen requirement than other parts of the body. Tonometry is based upon the relatively
simple principle that measurement of the luminal PCO2 reflects the tissue CO2 as CO2 is
considered freely diffusible. However, methodological limitations (such as the requirement for
gastric acid suppression and discontinuing enteral feed) have limited the uptake of gastric
tonometry into general critical care practice despite the fact that trends in gastric intramucosal
pH measurements are useful in predicting outcome.17 Furthermore, the calculation of
intramucosal pH requires inclusion of an estimate of bicarbonate ion concentration. It has
been assumed that the arterial bicarbonate ion concentration is identical to the intramucosal
concentration, which may or may not be the case. As a result, the intragastric PCO2 has
superseded measurement of the intramucosal pH and the gap between the gastric and arterial
PCO2 is now a more reliable measure of prognosis.18 Lebuffe et al18 measured gastric to end
tidal PCO2 gap (and so removing the requirement for arterial sampling) in high risk surgical
patients undergoing surgery exceeding two hours duration. He found that the maximum
intraoperative difference (above a threshold value of 21 mmHg) could predict postoperative
morbidity.
However, promising the advances in tonometry technology, the passage of a gastric (or
other) tube is still required. Tissue CO2 can be measured less invasively at the skin and
sublingual mucosa. In critically ill patients, Marik and Bankov20 showed that sublingual
capnography correlated well with gastric tonometry results and in fact were better predictors
of patient outcome that other global haemodynamic markers such as arterial lactate levels or
mixed venous saturation. Patients with initial sublingual-arterial PCO2 gaps greater than 25
mmHg had a higher mortality rate. Despite correction of traditional haemodynamic parameters,
the sublingual-arterial PCO2 gap remained higher in non-survivors compared to survivors.
Creteur21 combined sublingual tonometry and orthogonal polarization spectral imaging and
showed that tissue CO2 corresponded with the proportion of non-perfused capillaries and
sublingual microvascular flow. Using a dobutamine infusion he showed that the sublingual
PCO2 decreased as the sublingual perfusion improved and that the falls in sublingual PCO2
paralleled the increase in proportion of perfused capillaries. However, despite these encouraging
results, commercially produced equipment for measuring sublingual PCO2 is not widely
available.
OPTIMISING THE MICROCIRCULATION

Optimising the microcirculation involves reversing arteriolar hyporesponsiveness, increasing
perfused capillary density and reversing venular obstruction by microthrombi.
Volume expansion will recruit vascular beds if the circulating volume is deficient and the
body’s autoregulatory mechanisms are intact. It will restore the microcirculatory barrier function
and promote oxygen transfer as haemoglobin acts as an O2 sensor and O2 responsive nitric
oxide signal transducer, mediated by S-nitroso (SNO) haemoglobin.22 However, paradoxically
simple haemodilution may divert blood flow from weak microvascular beds by altering the
capillary driving pressure secondary to rheological alterations (see above). Despite concerns
about the immunosuppressive effect of blood transfusion and the consequent increases in
morbidity and mortality, in the resuscitative phase of sepsis management fresh blood will
improve microcirculatory oxygen delivery.
Because of the pivotal role of variable expression of inducible nitric oxide synthase, nitric
oxide inhibitors might seem a logical therapy for improvement in microcirculatory flow.
Although animal based work with nitric oxide inhibitors is encouraging, a clinical trial failed
to improve mortality23 and attempts to clinically use such inhibitors seem to have been
abandoned for the present. Nitric oxide stimulates soluble guanylate cyclase in vascular cells
which in turn induces relaxation and hyporeactivity. Methylene blue inhibits guanylate cyclase
and its administration in patients with septic shock increases mean arterial pressure and systemic
vascular resistance while decreasing vasopressor requirements. However, its effect on mortality
has not been conclusively proven. Despite steroids having multiple potentially beneficial effects
(see later), their advantage in suppressing the pathological mechanisms in the microcirculation
(including inhibiting nitric oxide synthase) of the septic patients has not been clearly proven.
According to the shunting theory of sepsis, vasodilators such as prostacyclin, dobutamine
and glycerine trinitrate (a nitric oxide donor) will improve the driving pressure at the arterial
end of the microcirculation so long as the circulating volume has been restored. Vasopressors
such as norepinephrine and vasopressin will increase the arterial pressure but this may not
necessarily be matched by improvements in the markers of microcirculatory function, especially
if vasopressors are used in high doses. The use of agents is therefore a balance between
systemic and microcirculation perfusion.
Agents such as Activated Protein C and steroids have several different actions on the
microcirculatory flow. Activated Protein C is both an anticoagulant (Factor V and VIII inhibition)
and profibrinolytic (inhibits plasminogen activator inhibitor-1 and limits thrombin-activatible
fibrinolysis inhibitor). It also protects the endothelial barrier from thrombin-induced disruption
and decreases white cell chemotaxis and adhesion. Steroids suppress a number of key functions
of the immune system. They inhibit macrophage and endothelial cell function and reduce the
migration of white cells to inflammatory sites. The adhesion of neutrophils to endothelial cells
is reduced as is the release of humoral factors. In addition to their anti-inflammatory effects,
steroids also increase cardiac output and blood pressure and correct relative adrenal
insufficiency and potentiate adrenergic receptor sensitivity.
QUANTIFYING THERAPEUTIC EFFECT

Unless the sophisticated monitors for orthogonal polarization spectral and sidestream dark-
field imaging or tonometry are available, it is very difficult to directly assess the effects of any
therapeutic manoeuvres on microvascular flow. The equipment required for routine polarization
spectral and sidestream dark-field imaging and sublingual tonometry is not widely available.
Automated gastric tonometers are available but the practicalities of their use may limit their
application to most critically ill patients.
Most intensive care units will be limited to measurement and manipulation markers that
are upstream of the microcirculation (e.g. preload, contractility and afterload) and downstream
(e.g. SvO2, lactate, base deficit and inflammatory markers). The therapeutic goals that have
been published for early goal directed therapy and in the Surviving Sepsis Campaign (and
subsequently incorporated into care bundles) are appropriate for improving microcirculatory
flow even if their direct action cannot be assessed. For example, the guidelines recommend
that:
“Dopamine and norepinephrine are both effective for increasing arterial blood pressure.
It is imperative to ensure that patients are adequately fluid resuscitated. Dopamine raises
cardiac output more than norepinephrine, but its use may be limited by tachycardia.
Norepinephrine may be a more effective vasopressor in some patients.”24
Most of these guidelines will improve or reverse the systemic conditions that predispose
towards microcirculatory flow abnormalities (Table 7.2).
CONCLUSION
The importance of the microcirculation relates to its vital role in delivering oxygen to the
mitochondria. Unfortunately without advanced imaging or monitoring techniques, it is part
of the circulation that cannot be easily monitored. Our understanding of what beneficial
influence intensive care support might bring to the microcirculation is therefore limited.
However, further research is vital as this will not only benefit the patients but perhaps resolve
the unanswered question as to whether it is oxygen delivery impairment or failure of oxygen
usage that causes the lethal corners in capillary beds and so predisposes to organ failure.
REFERENCES
1. Ince C. The microcirculation is the motor of sepsis. Crit Care 2005;9(Suppl 4):S13-9.
2. Trzeciak S, Dellinger RP, Parrillo JE, Guglielmi M, Bajaj J, Abate NL, et al. Microcirculatory Alterations in
Resuscitation and Shock Investigators. Early microcirculatory perfusion derangements in patients with
severe sepsis and septic shock: relationship to hemodynamics, oxygen transport, and survival. Ann Emerg
Med 2007;49(1):88-98.
3. Sakr Y, Dubois MJ, De Backer D, Creteur J, Vincent JL. Persistent microcirculatory alterations are associated
with organ failure and death in patients with septic shock. Crit Care Med 2004; 32(9):1825-31.
4. Cosby K, Partovi KS, Crawford JH, et al. Nitrite reduction to nitric oxide by deoxyhemoglobin vasodilates
the human circulation. Nat Med 2003;9(12):1498-505.
5. Brealey D, Brand M, Hargreaves I, et al. Association between mitochondrial dysfunction and severity and
outcome of septic shock. Lancet 2002;360(9328):219-23.
6. Spronk PE, Zandstra DF, Ince C. Bench-to-bedside review: sepsis is a disease of the microcirculation. Crit
Care 2004;8(6):462-8.
7. Ince C, Sinaasappel M. Microcirculatory oxygenation and shunting in sepsis and shock. Crit Care Med
1999;27(7):1369-77.
8. De Backer D, Creteur J, Preiser JC, Dubois MJ, Vincent JL. Microvascular blood flow is altered in patients
with sepsis. Am J Respir Crit Care Med 2002;166(1):98-104.
9. Elbers PW, Ince C. Mechanisms of critical illness-classifying microcirculatory flow abnormalities in distributive
shock. Crit Care 2006;10(4):221-8.
10. Taylor AE, Moore TM. Capillary fluid exchange. Am J Physiol 1999;277:S203-10.
11. Tsai AG, Cabrales P, Winslow RM, Intaglietta M. Microvascular oxygen distribution in awake hamster
window chamber model during hyperoxia. Am J Physiol Heart Circ Physiol 2003;285(4): H1537-45.
12. Baskurt OK, Temiz A, Meiselman HJ. Red blood cell aggregation in experimental sepsis. J Lab Clin Med
1997;130(2):183-90.
13. De Backer D, Hollenberg S, Boerma C, Goedhart P, Büchele G, Ospina-Tascon G, et al. How to evaluate
the microcirculation: report of a round table conference. Crit Care 2007;11(5):R101-9.
14. Gutierrez G. A mathematical model of tissue-blood carbon dioxide exchange during hypoxia. Am J Respir
Crit Care Med 2004;169(4):525-33.
15. Vallet B, Teboul JL, Cain S, Curtis S. Venoarterial CO(2) difference during regional ischemic or hypoxic
hypoxia. J Appl Physiol 2000;89(4):1317-21.
16. Creteur J. Gastric and sublingual capnometry. Curr Opin Crit Care 2006;12(3):272-7.
17. Friedman G, Berlot G, Kahn RJ, Vincent JL. Combined measurements of blood lactate concentrations
and gastric intramucosal pH in patients with severe sepsis. Crit Care Med 1995; 23(7):1184-93.
18. Levy B, Gawalkiewicz P, Vallet B, Briancon S, Nace L, Bollaert PE. Gastric capnometry with air-automated
tonometry predicts outcome in critically ill patients. Crit Care Med 2003;31(2):474-80.
19. Lebuffe G, Vallet B, Takala J, et al. A European, multicenter, observational study to assess the value of
gastric-to-end tidal PCO2 difference in predicting postoperative complications. Anesth Analg
2004;99(1):166-72.
20. Marik PE, Bankov A. Sublingual capnometry versus traditional markers of tissue oxygenation in critically
ill patients. Crit Care Med 2003;31(3):818-22.
21. Creteur J, De Backer D, Sakr Y, Koch M, Vincent JL. Sublingual capnometry tracks microcirculatory changes
in septic patients. Intensive Care Med 2006;32(4):516-23.
22. Singel DJ, Stamler JS. Chemical physiology of blood flow regulation by red blood cells: the role of nitric
oxide and S-nitrosohemoglobin. Annu Rev Physiol 2005;67:99-145.
23. Lopez A, Lorente JA, Steingrub J, et al. Multiple-center, randomized, placebo-controlled, double-blind
study of the nitric oxide synthase inhibitor 546C88: effect on survival in patients with septic shock. Crit
Care Med 2004;32(1):21-30.
24. Hollenberg SM, Ahrens TS, Annane D, et al. Practice parameters for hemodynamic support of sepsis in
adult patients: 2004 update. Crit Care Med 2004;32(9):1928-48.
Section C: Renal Problems in ICU
Eight
Biomarkers in Acute Kidney Injury
RM Chakravarthi
INTRODUCTION
It is common knowledge that over the last three decades, outcome of acute myocardial infarction
(AMI) has improved and the mortality has reduced from about 50% to the present rate of about
6% or so. In contrast, the outcome of acute kidney injury (AKI) during the same period has not
undergone any major improvement. A review published in 2005, analysed studies of AKI
published between 1970 and 2004 (80 studies, N=15,897) and found that reported mortality
of acute renal failure (ARF) had not changed over these years and was nearly constant at
around 50%.1 Many other studies have confirmed this high mortality, ranging between 40 to
60% associated with AKI in the critical care setting.2-5 Several factors account for unchanged
high mortality figures; they are discussed below, not necessarily in any order of importance.
Firstly, a change in the demographics of ICU patients who are diagnosed and treated for
AKI has been evident over this period. In more recent years, patients are older and have
multiple co-morbidities and have a greater number organ system failing along with AKI.6
Secondly, a lack of commonly accepted definitions used for diagnosing AKI has hampered
progress.7 Serum creatinine, a surrogate for glomerular filtration rate (GFR) and popularly
used as a marker of renal dysfunction, is unreliable especially in the acute setting. A large
decline in GFR may manifest only as a small change in serum creatinine level, thereby delaying
the recognition of AKI and initiation of appropriate therapy.8,9 Lastly, the lack of specific
biomarkers for early recognition of AKI has meant a delayed onset of treatment. Unlike
biomarkers for AMI like troponins that facilitate an early diagnosis and therapy (and have a
positive impact on outcome), patients with AKI have none
This chapter aims to review the current state of early diagnosis of AKI by using new
biomarkers of kidney injury.
BIOMARKERS OF AKI
Biomarkers – shortened form of ‘Biological markers’ is a term that was introduced in 1989
to describe any measurable diagnostic indicator that is used to asses the risk or presence of
Biomarkers in Acute Kidney Injury 75
disease.10 Until recently and notwithstanding significant advances in our understanding of

AKI, research in this area has been slow to catch-on. Pathophysiological processes leading to
the spectrum of AKI from subclinical injury to complete organ failure are more clearly
understood now.11 However, progress into the identification of a reliable biomarker of acute
kidney injury has been sluggish. Pathologically, processes that cause AKI include persistent
vasoconstriction, tubular obstruction, cellular, structural, metabolic alterations and
inflammation.11 To find a single marker or a test that can safely and reliably indicate kidney
injury from a wide variety of processes has been difficult.
Traditionally used markers of AKI including blood urea, serum creatinine and urinary indices
are non-specific and there is a time lag between the onset of kidney injury and abnormalities
of these markers in blood or urine. This is because these markers are metabolic waste products,
which accumulate many hours or days after onset of kidney injury. These markers vary with
patient’s physical state, e.g. blood urea may be elevated in dehydration and serum creatinine
is sex and muscle mass dependent. Intervention after these markers are elevated may be too
late to improve outcome. Thus, there is an urgent need for the identification of alternative
markers of kidney injury that can be detect AKI early and prevent evolution of this syndrome
into a full blown case of dialysis requiring renal failure. The last decade of research has
fortunately yielded biomarkers and therapeutic targets, which may enable us to identify AKI
early enough (long before changes in traditional markers of kidney function e.g., urea or
creatinine).
Biomarkers by definition are components of blood or urine that can be quantified. An
ideal biomarker of AKI is one that can be obtained non-invasively at the bedside, is easy to
measure and is relatively inexpensive. Most importantly, a biomarker should predict kidney
injury before histological damage sets in and also identify the segment of nephron most affected
by the pathological process.12
Various biomarkers have been studied as markers of kidney injury (Table 8.1).13,14 Many
have been tested in clinical situations for early detection of target population and for the
application of preventative and treatment strategies. Each of the markers used in AKI may be
discussed in the context of four distinct phases of AKI, risk identification, surveillance, diagnosis
and prognosis though some of them are useful in more than one phase (Table 8.2).15,16 For
the purposes of this review, only novel biomarkers that appear to be the most promising for
clinical application and early detection of AKI have been included.
SERUM BIOMARKERS OF AKI
Serum Creatinine
Though popularly used as a marker of kidney injury, it is a late marker of kidney injury,
increasing in blood several hours, sometimes as long as 48 hours after an insult to renal
homoestatsis. Creatinine is an amino acid derivative of creatine metabolism and has a molecular
weight of 113 kDa. Under steady state conditions, it is produced at a constant rate and is
Table 8.1: Biomarkers for early detection of AKI
Body fluid Biomarkers* Marker for Time for detection Assay
Serum Cystatin C GFR 12-14 hrs Available

NGAL Prox tubular injury 2-4 hrs Not available
Urine Cystatin C Prox tubular injury NA Not available
NGAL Prox tubular injury 2-4 hrs Available
KIM -1 Prox tubular injury 12-24 hrs Not available
IL - 18 Prox tubular injury 4-6 hrs Not available
NHE - 3 Prox tubular injury NA Not available
*see text for abbreviations
Table 8.2: Clinical phase of AKI and types of biomarkers that are useful in each phase
Marker Risk identification Surveillance Diagnosis Prognosis
Clinical Congestive heart Low BP Low urine output Dialysis

failure Low urine output
Liver disease
Diabetes mellitus
CKD
Sepsis
Functional High creatinine GFR Cystatin C GFR
Low GFR Renal blood flow Fractional excretion Na
Low Renal blood Fractional excretion Urea
flow GFR
Structural Proteinuria NHE 3 KIM-1 KIM-1
NAG NHE 3
NAG
GST
Inflammatory Reactive O2 species NGAL NGAL
FGF 23 IL-18 IL-18
C-reactive protein
Pentraxins
Abbreviations: CKD, chronic kidney disease; FGF 23 Fibroblast growth factor 23; GFR glomerular filtration
rate; GST glutathione S-transferase; IL-18 interleukin 18; KIM-1 kidney injury molecule 1; NAG N-acetyl-b-
(D)-glucosaminidase; NGAL neutrophil gelatinase-associated lipocalin; NHE 3 sodium hydrogen exchanger 3
filtered freely without reabsorption or metabolism by the kidney. However, between 10-40%
of creatinine is excreted into the urine by tubular secretion, making it unreliable as a marker
of glomerular filteration. Serum creatinine values generally rise only after >50% of the kidney
function is lost; hence, an initial decline in kidney function goes undetected. The production
of creatinine is variable according to age, race, sex, muscle mass, metabolism, nutritional
status and concomitant co-morbid conditions, hydration status and medication use. In critically
ill patients, the production of creatinine is also altered. All these factors make it a very unreliable
marker of kidney injury especially in the acute setting.9,13
Cystatin C
Human Cystatin ‘C’ is a non-glycosylated, endogenous protein of low molecular weight
(13 kDa) belonging to the cystatin superfamily. This protein is secreted from all nucleated
cells of the body and is thus, found in detectable amounts in most body fluids. In the kidneys,
it is removed from plasma by glomerular filtration. There is a linear relationship between the
reciprocal cystatin C concentration in plasma and GFR, making it superior to creatinine
clearance for the estimation of GFR. Cystatin C levels are unaffected by patient’s age, body
size or diet, but are influenced by abnormalities in thyroid function, immunosuppressive use
and in the presence of systemic inflammation like SIRS, trauma or sepsis. More than 99% of
cystatin produced by the body is flitered by the kidney and metabolized almost exclusively by
the proximal tubular cells. As a result, none appears in urine unless the tubular cells are
damaged.
Many clinical studies have investigated the usefulness of cystatin C as a marker of AKI. It
has been reported to have a higher sensitivity to detect reduced GFR especially early in the
so-called ‘creatinine blind’ range.17 In addition, as an estimate of GFR, it performs better
than the standard creatinine clearance in patients who have low serum creatinine levels to
begin with (malnourished, cirrhotics, children. elderly etc). In a small study on critically ill
patients,18 a 50% or greater elevation in cystatin C levels were found, on average 1-2 days
ahead of elevation in serum creatinine levels. Likewise, urinary detection of cystatin C in a
separate but small study of ICU patients was predictive of the need for subsequent dialysis,
forecasting the need for this intervention better than other markers. More studies are needed
to further refine the role of cystatin C as a diagnostic test in ICU patients.
URINARY BIOMARKERS OF AKI

Urinary biomarkers can further be classified as inflammatory biomarkers or tubular proteins.19
Among the inflammatory biomarkers, NGAL and urinary IL-18 seem to hold some promise
for clinical application in the future. They are discussed in some more detail in the following
paragraphs.
Neutrophil Gelatinase-associated Lipocalin (NGAL)

NGAL is a 25 kDa protein that is covalently bound to gelatinases from neutrophils. It is also
known as human neutrophil lipocalin, lipocalin-2 or siderocalin and belongs to the super-
family of proteins called lipocalins. It is one of the earliest and robustly elevated proteins in
the setting of ischaemic and nephrotoxic AKI.20-22 In kidney tubules, NGAL mRNA is
upregulated within a few hours of a noxious stimulus, indicating the possibility that this protein
belongs to a panel of stress-induced renal biomarkers involved in the pathophysiogical process

of acute renal damage.22 In a small cohort study, serum NGAL levels and urinary NGAL
levels were significantly increased from baseline within 2 hours after on-pump coronary bypass
surgery in individuals who later developed acute kidney injury.23 In this study, multivariate
analysis confirmed the amount of NGAL in urine as the most powerful, independent predictor
of severe kidney damage, and receiver operator characteristic curves showed good diagnostic
value for a cutoff level of 50 mg/l, suggesting that NGAL in this context represented an early,
sensitive, and specific predictor of acute kidney injury.23 In another study on patients
undergoing elective cardiac surgery,24 all patients showed elevated values of NGAL in the
postop period. Interestingly, patients who developed AKI showed increasing values of NGAL
beyond 2 hours post-op, while others who did not develop AKI showed an early high value
that declined over time. Those who developed AKI showed a peak in creatinine only by
day 4 postop.
In a study involving 635 individuals with normal renal function or various types of renal
dysfunction, Nickolas et al25 found that a single measurement of NGAL helped to distinguish
AKI from all other forms of kidney dysfunction. More recently, urinary NGAL has been assessed
as a potential marker of contrast induced nephropathy in patients undergoing cardiac
interventions.26
Overall, the role of NGAL in the diagnosis of early AKI has yet to be established. While
early studies have reported promising results, more data is clearly needed to add to our current
approach (Table 8.3).
Interleukin-18 (IL-18)
IL-18 is a pro-inflammatory cytokine induced in the proximal tubules during AKI. Active forms
of IL-18 exit the cells and enter urine after being activated in proximal tubules.27 In mice,
urinary IL-18 was increased in ischemic AKI compared to sham operated mice.27 Parikh et
al28 reported a study of 72 patients with various renal diseases where urinary IL-18 had a
high sensitivity and specificity for the diagnosis of acute tubular necrosis (ATN) versus pre-
renal azotaemia, urinary tract infection, chronic kidney disease and normal renal function in
healthy subjects.
In a case control study of ICU patients with ARDS, elevated levels of IL-18 (>100 pg/ml)
were associated with an estimated > 6 fold increase in the odds of developing AKI.29 Following
cardiopulmonary bypass, IL-18 levels are elevated early in the post-op period and levels
obtained at 4 hours have a predictive value for the development and duration of AKI.30
Kidney Injury Molecule-1 (KIM-I)

KIM-1 is a transmembrane glycoprotein that is normally expressed in kidney tissue. In animal
studies, KIM-I is found to be over expressed in proximal tubular cells in both ischaemic and
nephrotoxic acute kidney injury.31,32 In human studies, urinary levels of KIM-I have been
Table 8.3: Predictive value of NGAL in AKI-results of major recent studies
Study Year Subjects/ Methods Findings
Mishra et al 2006 71 Children undergoing uNGAL and sNGAL 2-h values are independent predictors of
cardiopulmonary bypass AKI. Using a cutoff value of 50mg/L: AUC, 0.99; sensitivity,
100%; specificity, 98% in predicting AKI
Hirsch et al 2007 91 Children undergoing Significant increase in uNGAL and sNGAL within 2 h after
elective cardiac catheteriza- intervention in those subsequently developing CIN. Using a
tion and contrast adminis- cutoff value of 100 ng/ml of s uNGAL; AUC, 0.92: sensitivity,
tration (loversol) 73%; and specificity, 100%.
Wagener et al24 2006 81 adults undergoing major Patients developing postoperative AKI had greater uNGAL levels
cardiac surgery until 18 h after surgical intervention. Using a cutoff value of 213
ng/ml for uNGAL at 18 h: AUC, 080; sensitivity, 73%; and
specificity, 78% in predicting postoperative AKI.
Bachorzewska- 2006 Patients undergoing Significant Increase in sNGAL 2 and 4 h after PCI; significant
Gajewska et al26 elective PCI increase in uNGAL 2 and 12 h after PCI. Significant association
between s NGAL and serum creatinine.
Trachtman et al 2006 34 Children affected by Children with greater baseline uNGAL (>200 ng/ml) showed
diarrhoea-associated HUS greater peak blood urea and creatinine levels and required
dialysis more often than others (9/20 v 1/14).
Mishra et al 2006 Biopsy specimens from In cadaveric specimens, NGAL staining intensity strongly
13 cadaveric and 12 living- correlated with cold ischaemia time (R = 0.87; P < 0.001) and
related kidney allografts postoperative serum creatinine peak (R = 0.86; P <0.001).
Patients developing delayed graft function showed the most
intense NGAL staining in biopsy specimens.
Parikh et al 2006 Urine samples collected Greater NGAL values in delayed graft function group with
day 0 from recipients of respect to others. Using a cutoff value of 1,000 ng/mg creatinine:
living donor and deceased AUC, 0.90; sensitivity, 90%; and specificity, 80% in predicting
donor kidneys with or delayed graft function.
without delayed graft
function
Abbreviations: NGAL, neutrophil gelatinase associated lipocalin; sNGAL, serum NGAL; uNGAL, urinary NGAL; AUC, area under the curve;
AKI, acute kidney injury; CIN, contrast induced nephropathy; PCI, percutaneous coronary intervention.
Biomarkers in Acute Kidney Injury
79
found elevated in patients with ischaemic ATN compared to patients with contrast-induced
nephropathy, patients of CKD and normal individuals.33 Kidney biopsies from patients also
show marked upregulation of KIM-1 molecule in patients with AKI compared to other types
of acute renal insults.
Na/H Exchanger Isoform 3 (NHE3)

NHE3 is a sodium transporter protein located in the apical membrane of the tubular cells of
proximal tubules and thick ascending limb of loop of Henle. It is liberated in to the lumen and
excreted in urine when there is ischemic necrosis.34 du Cheyron et al35 demonstrated a six-
fold increase in urinary NHE3 in patients with AKI compared with control subjects. However,
the clinical utility of this marker is still unknown.
OTHER MARKERS OF KIDNEY INJURY

Many enzyme markers have been identified in urine and have been found to be elevated in
patients with AKI. These include g-glutamyl transpeptidase (gGT); glutathione transferase-pi
(GST-pi), glutathione transferase-a (GST-a), alkaline phosphatase (AlkP) and N-acetyl
glucoaminidase (NAG).36 Although, these markers are sensitive for tubular damage, their
predictive value in AKI is poor as elevated values are known to occur even in patients who
do not eventually develop AKI. Situations where early and minor tubular damage occurs as
in elderly patients or those after coronary bypass surgery may be associated with elevated
levels of these enzyme markers.37
Albumin in urine is a marker of kidney injury but is non-specific, doesn’t discriminate
between patients with or without AKI and drops to near normal levels within 24 hrs of injury
even in patients with AKI.38
CONCLUSION
Unfortunately, biomarkers of kidney injury are not universally available as yet for use by the
bedside. Until that happens, we must rely on the traditional, easily measurable markers like
creatinine and urine output. The acute dialysis quality initiative (ADQI) have developed a set
of criteria for defining AKI based upon serum creatinine and urine output called the RIFLE
criteria (Risk, Injury, Failure, Loss and End stage) criteria.39 These have been discussed in an
earlier issue of the Update.
Presently several large studies have validated RIFLE criteria in different clinical situations.40,41
These studies are important as they indicate that identification and prevention of AKI or
attenuation of the injury could result in marked decreases in morbidity, mortality and cost of
therapy of AKI.
In the future, RIFLE criteria used in combination with biomarkers that are sensitive and
specific in AKI could revolutionize the management of AKI and, possibly improve outcome
of AKI in the critically ill. Further multi-centre research, is necessary to come up with validation
data of biomarkers and identification of the ideal biomarker which is easily measurable,
accurate, reproducible, cost effective and easy to interpret by the clinician. NAGL is probably
the best available at present; however, meaningful research is needed before any of the
biomarkers can be recommended for routine clinical use.
REFERENCES
1. Ympa YP, Sakr Y, Reinhart K, Vincent JL. Has mortality from acute renal failure decreased? a systematic
review of the literature. Am J Med 2005;118:827-32.
2. Liano F, Pascual J: Epidemiology of acute renal failure: A prospective, multicentre community based study.
Madrid Acute Renal Failure Study Group. Kidney Int 1996;50:811-8.
3. Metnitz PG, Krenn CG, Steltzer H, et al. Effect of acute renal failure requiring renal replacement therapy
on outcome in critically ill patients. Crit Care Med 2002;30:2051-8.
4. Mehta RL, Pascual MT, Soroko S, et al. Spectrum of acute renal failure in the intensive care unit: the
PICARD Experience Kidney Int 2004;66:1613-21.
5. Uchino S, Kellum JA, Bellomo R, et al. Acute renal failure in critically ill patients: a multinational, multicenter
study. JAMA 2005;294:813-18.
6. Turney JH, Marshall DH, Brownjohn AM, et al. The evolution of acute renal failure 1956-1988. Q J Med
1990;74:83-104.
7. Kellum JA, Levin N, Bouman C, Lameire N. Developing a consensus classification system for acute renal
failure. Curr Opin Crit Care 2002;8:509-14.
8. Lameire N, Hoste E. Reflections on definition, classification and diagnostic evaluation of acute renal failure.
Curr Opin Crit Care 2004;10:468-75.
9. Star RA, Treatment of acute renal failure. Kidney Int 1998;54:817-31.
10. Gutman S, Kessler LG. The US Food and Drug Administration perspective on cancer biomarker
development. Nat Rev Cancer 2006;6:565-71
11. Bonventre JV, Weinberg JM. Recent advances in pathophysiology of ischemic acute renal failure. J Am
Soc Nephrol 2003;14:2199-210.
12. Shah HS, Mehta R.L. Acute Kidney injury in critical care: time for a paradigm shift? Curr Opin Nephrol
Hypertens 2006;15:561-65.
13. Zhou H, Hewitt SM, Yuen PS, et al. Acute Kidney injury biomarkers: Needs, present status, future promise.
Neph Self Assess Program 2006;5:63-71.
14. Han WK, Bonventre JV, Biologic markers for the early detection of acute kidney injury. Curr Opin Crit
Care 2004;10:476-82.
15. Mehta RL, Kellum JA, Shah SV, et al Acute Kidney Injury Network: report of an initiative to improve
outcomes in acute kidney injury. Crit Care 2007;11:R31.
16. Coca SG, Parikh CR. Urinary Biomarkers for Acute Kidney Injury: perspectives on translation: Clin J Am
Soc Nephrol 2008;3:481-90.
17. Newman DJ, Thakkar H, Edwards RG, et al. Serum Cystatin ‘C’ measured by automated immunoassay:
A more sensitive marker of changes in GFR than serum creatinine. Kidney Int 1995;47:312-8.
18. Herget-Rosenthal S, Marggraf G, Hsuing J, et al. Early detection of acute renal failure by serum cystatin
c. Kidney Int 2004;66:1115-22.
19. Molitoris BA, Melnikov VY, Okusa MD, et al. Technology insight: biomarker development in acute kidney
injury—what can we anticipate? Nat Clin Pract Nephrol 2008;4:154-65.
20. Supavekin S, Zhang W, Kucherlapati R, et al. Differential gene expression following renal ischemia/
reperfusion. Kidney Int 2003;63:1714-24.
21. Mishra J, Ma Q, Prada A, et al. Identification of neutrophil gelatinase-associated lipocalin as a novel early
urinary biomarker for ischemic renal injury. J Am Soc Nephrol 2003:14:2534-43.
22. Mishra J, Mori K, Ma Q, et al. Neutrophil gelatinase-associated lipocalin: a novel early urinary biomarker
for cisplatin nephrotoxicity. Am J Nephrol 2004;24:307-15.
23. Bolignano D, Donato V, Coppolino G, et al. Neutrophil gelatinase-associated lipocalin (NGAL) as a marker
of kidney damage. Am J Kidney Dis 2008;52:595-605.
24. Wagener G, Jan M, Kim M et al. Association between increases in urinary neutrophil gelatinase-associated
lipocalcin and acute renal dysfunction after adult cardiac surgery. Anaesthesiology 2006;105:485-91.
25. Nickolas TL, O’Rourke MJ, Yang J, et al. Sensitivity and specificity of a single emergency department
measurement of urinary neutrophil gelatinase-associated lipocalin for diagnosing acute kidney injury. Ann
Intern Med 2008:148:810-9.
26. Bachorzewska-Gajewska H, Malyszko J, Sitniewska E, et al. Neutrophil gelatinase-associated lipocalcin
and renal function after percutaneous coronary interventions. Am J Nephrol 2006;26:287-92.
27. Melinkov VY, Ecder T, Fantuzzi G, et al. Impaired IL-18 processing protects caspase-1-deficient mice from
ischemic acute renal failure. J Clin Invest 2001;107:1145-52.
28. Parikh CR, Jani A, Melnikov VY, et al Urinary interleukin -18 is a marker of human acute tubular necrosis.
Am J Kidney Dis 2004;43:405-14.
29. Parikh CR, Abrahams E, Anculkiewicz M, Edelstein CL. Urine IL 18 is an early marker for acute kidney
injury and predicts mortality in intensive care unit. J Am Soc Nephrol 2005;16:3046-52.
30. Parikh CR, Mishra J, Thiiessen-Philbrook H, et al. Urinary IL 18 is an early predictive biomarker of acute
kidney injury after cardiac surgery. Kidney Int 2006;70:199-203.
31. Ichimura T, Bonventre JV, Bailly V, et al. Kidney injury molecule-1 (KIM-1), a putative epithelial cell adhesion
molecule containing a novel immunoglobulin domain, is up-regulated in renal cells after injury. J Biol
Chem 1998;273:4135-42.
32. Ichimura T, Hung CC, Yang SA, et al. Kidney injury molecule-1: a tissue and urinary biomarker of
nephrotoxicant-induced kidney injury. Am J Physiol Renal Physiol 2004;286:F552-63
33. Han WK, Bailly V, Abichandan R, et al. Kidney injury molecule (KIM-1): a novel biomarker for human
renal proximal tubular injury. Kidney Int 2002;62:237-44.
34. Kwon TH, Frokiaer J, Han JS, et al. Decreased abundance of major Na (+) transporters in kidneys of rats
with ischemia-induced acute renal failure. Am J Physiol Renal Physiol 2000;278:F925-39.
35. du Cheyron D, Daubin C, Poggiol J, et al. Urinary measurement of Na+/H+ exchanger isoform 3 (NHE3)
protein as new marker of tubule injury in critically ill patients with ARF. Am J Kidney Dis 2003;42:497-
506.
36. Westhuyzen J, Endre ZH, Reece G, et al. Measurement of tubular enzymuria facilitates early detection of
acute renal impairment in the intensive care unit. Nephrol Dial Transplant 2003;18:543-51.
37. Boldt J, Brenner T, Lang J, et al. Kidney-specific proteins in elderly patients undergoing cardiac surgery
with cardiopulmonary bypass. Anesth Analg 2003;97:1582-89.
38. Hynninen MS, Niemi TT, Pöyhiä R, et al. N-acetyl cysteine for the prevention of kidney injury in abdominal
aortic surgery: a randomized double blind, placebo-controlled trial. Anesth Analg 2006;102:1638-45.
39. Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure - definition, outcome measures, animal models,
fluid therapy and information technology needs: the Second International Consensus Conference of the
Acute Dialysis Quality Initiative (ADQI) group. Crit Care 2004;8:R204-12.
40. Ricci Z, Cruz D, Ronco C. The RIFLE criteria and mortality in acute kidney injury: A systematic review.
Kidney Int 2008;73:538-46.
41. Sajja LR, Mannam G, Chakravarthi RM, et al. Coronary artery bypass grafting with or without
cardiopulmonary bypass in patients with pre operative non-dialysis dependant renal insufficiency: a
randomized study. J Thoracic Cardiovasc Surgery 2007;133:378-88.
Septic Acute Kidney Injury: Epidemiology and Pathophysiology 83
Nine
Septic Acute Kidney Injury:
Epidemiology and Pathophysiology
Karim Somani, Sean M Bagshaw
INTRODUCTION
Acute kidney injury (AKI) is exceedingly common in critical illness, generally portends a worse
clinical outcome along with increased utilization of health resources.
Prior observational studies have suggested the occurrence of AKI in critically ill patients
can range from 1-25% depending on the specific definition used for AKI.1-4 Recently, several
large observational studies have suggested the occurrence of AKI is far more widespread than
previously appreciated.5-8 In these studies, AKI was defined by the RIFLE classification scheme,
a novel and consensus-driven definition for AKI9 (Table 9.1). Cruz et al in a prospective
observational study of 2164 critically ill patients found that when defined by the RIFLE criteria,
11% developed AKI, with 3.3% overall receiving renal replacement therapy (RRT).6 Two large
multi-centre retrospective studies found 36% of all ICU patients developed AKI.5,8 In a single-
centre retrospective study, Hoste et al found an estimated 67% of all ICU patients developed
AKI when defined by the RIFLE criteria.7 Importantly, this study found more than 50% of all
patients achieving the RISK class progressed to more severe kidney injury (Injury or Failure
class) and this was independently associated with higher in hospital mortality.
Moreover, additional data have emerged to suggest the incidence of AKI, in particular in
ICU settings, is increasing while mortality rates have only marginally improved.10-12 In a
10 year surveillance of early onset AKI in 20 ICUs across Australia, Bagshaw et al found the
incidence of AKI was increasing by approximately 3% per year.11 There are several plausible
explanations for these trends. First, the ‘modern’ critically ill patient is now more likely to be
elderly, to have significant co-morbid illnesses (including chronic kidney disease), and more
likely to develop AKI in the context of multi-system organ failure, invasive diagnostic testing,
and/or complex surgical interventions.13 These data indicate AKI is relatively common, is
increasing in incidence, and is associated with higher morbidity, mortality and health resource
utilization. Accordingly, the timely recognition and diagnosis of AKI is vitally important and
can directly impact patient outcome.
Table 91: Summary of operational definitions for AKI, sepsis and septic AKI
RIFLE classification scheme for AKI (Adapted from9)
RIFLE class Serum creatinine (SCr) criteria Urine output

criteria
Risk Increase in serum SCr > 1.5 × baseline or decrease in < 0.5 ml/kg/hr
GFR > 25% for > 6 hrs
Injury Increase in serum SCr > 2.0 × baseline or decrease in < 0.5 ml/kg/hr
GFR > 50% for > 12 hrs
Failure Increase in serum SCr > 3.0 × baseline or decrease in < 0.3 ml/kg/hr
GFR > 75% or an absolute serum Cr > 354 μmol/L with > 24 hrs or
an acute rise of at least 44 μmol/L anuria > 12 hrs
Loss Complete loss of kidney function >4 weeks
End stage Complete loss of kidney function >3 months
Sepsis/systemic inflammatory response syndrome (SIRS) (Adapted from 26)
Definition
Sirs Presence of > 2 of the following:

1. Temperature >38°C or <36°C
2. Heart rate >90 beats/min
3. Respiratory rate >20 breaths/min or PaCO2 <32 mmHg
4. White cell count >12,000 cells/mm3, <4,000 cells/mm3 or with >10% immature
(band) forms
Sepsis Presence of confirmed or suspected infection plus ³2 SIRS criteria
Severe sepsis Presence of sepsis plus e”1 organ system dysfunction
Septic shock Presence of severe sepsis plus fluid unresponsive hypotension
Septic AKI
Definition
Septic AKI AKI defined by RIFLE criteria and SEPSIS defined by the SIRS criteria
Sepsis is also a highly prevalent syndrome that precipitates critical illness and leads to ICU
admission.14,15 The multi-centre European SOAP study recently found 37.4% of all ICU patients
had sepsis, with 24.7% of patients presenting with a primary diagnosis of sepsis.15 In a large
observational study of approximately 750 million hospitalizations over a 25-year period, Martin
et al found an annualized increase in sepsis diagnoses of 8.7% that translated into significant
increases in the number of sepsis related deaths.14
While the etiology of AKI in critically illness is often multi-factorial, sepsis has consistently
been found to be a key contributing factor.4,15-20 The distinction of AKI of septic compared
with non-septic origin may have clinical relevance.21 Emerging data indicate that septic AKI
may be characterized by a distinct pathophysiology.22-25 For this reason, septic AKI may show
important differences in patient characteristics, response to interventions and clinical outcomes
when compared to AKI of non-septic origin. Accordingly, we believe it important that critical
care physicians have an appreciation of the epidemiology and pathophysiology of the
syndrome of septic AKI.
DEFINING SEPTIC AKI

In order to understand and describe the epidemiology and clinical outcomes of a syndrome
such as AKI, sepsis or septic AKI, the syndrome requires a standardized definition.
The syndrome of sepsis has benefited from a consensus developed standardized definition
for over 15 years.26 This has lead to numerous epidemiologic investigations of describing the
burden of illness attributable to sepsis and clinical trials of prophylactic and therapeutic
interventions.
Regrettably, this has not been the case for AKI. Inferences from epidemiologic studies of
AKI have been somewhat limited due to wide variations in how AKI has been defined.9 In-
fact, a recent report indicated that more than 35 different definitions of AKI are present in the
literature.27 This lack of agreement has been unfortunate and likely held up scientific progress
in the field of critical care nephrology. However, the Acute Dialysis Quality Initiative (ADQI)
group, comprised of experts in the fields of nephrology and critical care medicine, published
the RIFLE classification scheme, a novel consensus and evidence based definition for AKI as
described above9 (Table 9.1). The RIFLE classification defines three grades of severity of AKI
(Risk, Injury and Failure) based on changes to serum creatinine and urine output and two
clinical outcomes (Loss, End-Stage Kidney Disease). The RIFLE classification has now been
evaluated in a number of clinical studies of critically ill patients with AKI and generally found
to have clinical relevance for the diagnosis of AKI, classifying the severity of AKI and for
monitoring the progression of AKI, as well as having modest predictive ability for
mortality.5-8,10,17,28
Accordingly, the presence of validated consensus definitions for both sepsis and AKI provide
an opportunity for defining septic AKI as the simultaneous presence of both syndromes in the
absence of other clear and established, non–sepsis-related (i.e. radiocontrast media, other
nephrotoxins) precipitants of AKI (Table 9.1).
EPIDEMIOLOGY OF SEPTIC AKI
Incidence
The occurrence of septic AKI is surprisingly common in critically ill populations. To date, few
studies have specifically described the incidence of septic AKI. Most have reported the
proportions of patients with either sepsis or AKI among a cohort of ICU patients with one or
the other syndrome (Table 9.2).
In observational studies focused on critically ill patients with sepsis, the occurrence of
concomitant AKI has ranged between 11-64%, respectively.15-17,19,20,29,30 Rangel-Frausto et
al, first described the stepwise increase in occurrence of AKI, defined by having received RRT,
Table 9.2: Incidence estimates of septic AKI across observational studies
86
Primary observational studies of septic populations
Study Year N ICU Type(s) AKI Definition Incidence Mortality (%)

(%) No AKI AKI
Hoste16 2003 185 Surgical SCr >177 μmol/L 16 28§ 57§

Yengenaga19 2004 257 Mixed SCr >177 μmol/L 11 24‡ 72‡
Diaz de Leon29 2006 332 Mixed Based on serum/urine biochemical
not otherwise specified 32 - -
Lopes17 2007 182 Medical RIFLE criteria 37 10Ô 27-55Ô
Oppert30 2007 401 Mixed SCr × 2 or U/O 41 43§ 67§
<0.5 mL/kg × 4 hr
Bagshaw20 2008 4523 Mixed RIFLE criteria 64 35† 58†
Primary observational studies of AKI populations
Study Year N ICU Type(s) AKI definition Incidence Mortality (%)

(%) Non-septic Septic
Neveu18 1996 345 Mixed SCr >310 μmol/L or

Urea>36 mmol/L or
Increase SCr 100% 46 45§ 75§
Bagshaw33 2005 240 Mixed Need for RRT 58 - -
Soap15 2006 1177 Mixed SOFA score >2 51 23‡ 41‡
Best kidney34 2007 1753 Mixed Urea>30 mmol/L or
UO<200 mL/12 hr or RRT 48 52§ 70§
Bagshaw32 2007 120123 Mixed RIFLE criteria 32 22§ 30§
Cruz6 2007 2164 Mixed RIFLE criteria 26 - -
Abbreviations: SCr = serum creatinine; SOFA = sequential organ failure assessment; RRT = renal replacement therapy
§ Hospital mortality
‡ ICU mortality
Ô 60-day mortality
† 90-day mortality
in critically ill patients stratified by severity of sepsis.31 As patients progressed from sepsis to
severe sepsis and septic shock, the occurrence of AKI increased dramatically from 24% to
39% and 89%, respectively. Similarly, Lopes et al found the occurrence of AKI increased
from 29 to 51% as patients progressed from severe sepsis to septic shock.17 A recent multi-
centre point-prevalence study from Germany found 41% of patients with sepsis/septic shock
had AKI.30 In another multi-centre observational study of critically ill patients with septic shock,
Bagshaw et al reported 64% had evidence of AKI within 24 hours of onset of shock.20
For observational studies primarily focused on AKI in critical illness, sepsis has been found
a contributing factor in 26-58% of cases.6,18,32-34 In the BEST Kidney study, a prospective
observational surveillance of nearly 30,000 admissions to 54 ICUs across 23 countries, found
5.7% (n=1738) of critically ill patients developed AKI.4 Sepsis was found to be the primary
contributing factor in 47.5% of AKI cases.34 In the NEiPHROS-AKI study, Cruz et al found an
estimated 26% of AKI cases were attributable to sepsis.6 In a multi-centre study of early AKI
performed across 57 ICUs in Australia and New Zealand, Bagshaw et al found septic AKI was
evident in 11.7% (n=14,039) of all patients at the time of ICU admission.
Patient Characteristics/Risk Factors

In septic AKI, the predominant sources of sepsis have been shown to be chest, abdomen and
genitourinary.16,20,34 In addition, those developing AKI were more likely to have positive blood
cultures.20
Importantly, patients with septic AKI appear to have important differences in baseline
characteristics and have worse acute physiology and laboratory parameters. Septic AKI patients
are generally older and have a higher burden of co-morbid illnesses, including congestive
heart failure, chronic kidney disease, diabetes mellitus, liver disease, active malignancy and
chronic obstructive pulmonary disease.16,20,34 These patients generally have higher acuity of
illness, are more likely to be medical admissions, however, if surgical, more likely to have had
an emergency procedure.32 In addition, these patients have more hemodynamic instability,
worse pulmonary function, greater acidemia, higher white blood cell counts and a higher
likelihood to receive mechanical ventilation and vasoactive therapy.16,20,34
Several additional clinical and therapeutic factors may also contribute to development of
AKI in septic shock. For example, septic AKI patients have been shown to be more oliguric,
despite having received more fluid and diuretic therapy. As such, these patients are more
likely to accumulate fluid and have a positive fluid balance.16,20,32,34-36 In addition, Bagshaw
et al found that delays to effective antimicrobial therapy after the onset of shock in sepsis was
an independent predictor for development of AKI20 (Fig. 9.1).
Clinical Outcomes
These distinguishing features of septic AKI appeared to translate into relevant differences in
clinical outcomes when compared with non-septic AKI or sepsis alone.
Fig. 9.1: Cumulative effective antimicrobial therapy following onset of hypotension and associated
incidence of AKI (Adapted from20)
Numerous clinical studies have found septic AKI is associated with higher mortality
compared with either non-septic AKI or sepsis not complicated by AKI (Table 9.2). Oppert et
al reported hospital mortality of 67% vs. 43% for septic and non-septic AKI respectively.30
After adjustment for relevant co-variates, septic AKI remained a significant independent
predictor of death (odds ratio [OR] 2.1, 95% CI, 1.3-3.5). Similarly, in the BEST Kidney
study, septic (compared with non-septic AKI) was independently associated with higher risk
of hospital mortality in multi-variable analysis (OR 1.5, 95% CI 1.2-1.9).34 In addition, increased
severity of AKI in septic patients has also been shown to be associated with higher mortality
(Fig. 9.2). In a single centre retrospective study of 182 critically ill septic patients, Lopes et al
reported stepwise increases in mortality when patients progressed from no AKI to Risk, Injury
or Failure RIFLE classes (9.6%, 27.3%, 28.6%, 55%, respectively).17 Similarly, in a large
multicentre retrospective study of over 120,000 ICU patients, Bagshaw et al found higher
crude ICU and hospital mortality associated with increased severity of AKI.32 In another study
of 4532 critically ill patients with septic shock, Bagshaw et al found significantly higher ICU,
hospital and 90-day mortality when stratified by increasing RIFLE class of AKI, a finding that
persisted after adjustment in multivariable analysis.20
In critically ill patients surviving to hospital discharge, data have suggested septic compared
with non-septic AKI may be associated with greater renal recovery and independence from
RRT.34 Lengths of stay in both ICU and hospital for critically ill patients with septic AKI
compared with either non-septic AKI or sepsis alone have been shown to be significantly
longer.16,19,20,32,34 In summary, septic AKI has consistently been shown to be associated with
Fig. 9.2: Unadjusted ICU and in-hospital mortality stratified by the subgroups:
none; sepsis only; non-septic AKI; and septic AKI (Adapted from32)
greater morbidity and mortality, however, may also be associated with a higher rate of renal
recovery in survivors.
PATHOPHYSIOLOGY
Our understanding of the pathogenesis of human AKI in general and septic AKI in particular,
has been markedly affected by the paucity of histopathological data.21,37 In clinical settings,
lack of these data arise from the risks associated with performing routine and/or repeated
kidney biopsies in critically ill patients. In addition, this practice would likely be viewed as
unethical in the absence of a strong suspicion for kidney parenchymal disorders such as
vasculitis or glomerulonephritis, where findings on kidney biopsy may directly guide clinical
management (i.e. chemotherapy and/or immunosuppression).
Accordingly, we are often at a disadvantage in the ICU. In the absence of clinical histo-
pathologic correlation, we must rely on several indirect measures to evaluate what physiologic
processes are occurring in the kidney and early on in AKI.37 These measures include urine
output, evaluation of urinary biochemistry (i.e. urine sodium, fractional excretion of sodium,
fractional excretion of urea) and urinary microscopy. Regrettably, these data only represent
a surrogate for what processes may (or may not) be occurring at the level of the kidney
vasculature, interstitium and/or tubular cells. Thus, there have not been significant advances
in understanding the pathophysiology of septic AKI for several decades.
To overcome this lack of histopathologic correlation in humans, experimental animal models
of AKI have been developed that allow for more sophisticated, invasive and continuous
measurements to be made. Unfortunately, data from the majority of these experimental studies
have been based on ischaemia–reperfusion or drug-induced injury models of AKI. We contend
that these experimental models are not relevant to septic AKI, and that the data generated
from such models may in fact be misleading when applied by critical care clinicians to interpret
what might be occurring in the critically ill patient with septic shock developing AKI.37
Global Renal Blood Flow in Septic AKI

A longstanding paradigm in clinical nephrology derived largely from experimental and
observational data in hypodynamic shock (i.e., haemorrhagic, cardiogenic or even septic) is
that AKI is primarily a consequence of reduction in renal blood flow (RBF) leading to kidney
ischaemia.23,38,39 This construct would imply that simple timely restoration of adequate RBF
would be a method to provide kidney protection in all critically ill patients (septic or not).37
However, our understanding of RBF in critically ill patients, in particular those with sepsis is
limited. Specifically, we do not know whether, in critically ill septic patients, in presence of a
normal or increased cardiac output, RBF actually decreases, remains stable or even increases.
This is because RBF cannot be directly measures either continuously or intermittently (without
a significant invasive monitoring) in humans.
In selected experimental models of septic AKI, global RBF has been shown to decline
early after induction of sepsis or endotoxaemia.40 In these circumstances, a prolonged systemic
hypodynamic circulation and kidney ischaemia will contribute not only to a reduction in
GFR, but also metabolic failure and diminished content of high-energy phosphates, potentially
leading to renal tubular necrosis and severe AKI.37
Alternatively, other studies have shown that the kidney circulation participates in the systemic
vasodilatory and hyperdynamic state characteristic of severe sepsis/septic shock.41-46 As such,
RBF has been shown to be either preserved or augmented in sepsis and AKI occurs not in
the context of kidney hypoperfusion, but rather in the setting of adequate or increased global
RBF. In an experimental pig model of hyperdynamic sepsis, Ravikant et al found an increase
in global and medullary RBF.43 In a small cohort of critically ill septic patients with AKI, Brenner
et al estimated RBF by percutaneously placed renal vein thermodilution catheters and found
AKI had occurred in these patients despite near normal RBF.41 In a recent systematic review
of 160 studies of experimental models of sepsis evaluating kidney function measured by RBF,
Langenberg et al found, that in approximately one-third of studies, RBF was preserved or
increased.25,47 Moreover, by multivariable analysis, the only variable found to be independently
associated with RBF during experimental sepsis was cardiac output – where a preserved or
elevated cardiac output was associated with preserved or increased RBF and low cardiac
output was associated with reduced RBF.
In early and/or resuscitated septic shock, critically ill patients classically show a hyperdynamic
circulation characterized by high cardiac output and low systemic vascular resistance. Therefore,
evaluation of RBF and kidney function in hyperdynamic models of sepsis may be more
applicable and relevant to human septic shock. Recently, a large sheep model of septic shock
was developed, where both cardiac output and RBF were measured continuously and a high
cardiac output state was induced by a continuous infusion of E. coli that simulates the clinical
and haemodynamic state of human septic shock.46 In this study, Langenberg et al found,
during early hyperdynamic sepsis, RBF was markedly increased while renal vascular resistance
was concomitantly decreased.46 Despite this global increase in RBF, sheep developed AKI
characterized by progressive oliguria and a 3-fold increase in serum creatinine. In this model,
recovery of kidney function was associated with relative reductions in cardiac output and
RBF to near baseline.48 Moreover, during recovery indices of tubular function (i.e., urinary
sodium, fractional excretion of sodium, fractional excretion of urea), and conventional markers
of kidney function (i.e., serum creatinine) rapidly normalized and returned to baseline within
approximately 18 hours.48 These data suggest that early loss of kidney function may occur
through sepsis-induced changes to kidney vascular activity and may be an important
mechanism for the loss of glomerular filtration pressure early (i.e., 24-48 hours) after onset
of sepsis. More importantly, these data provide “proof of concept” that, in septic AKI, glomerular
filtration pressure (GFR) can be lost in the setting of markedly increased global RBF.37 As
such, septic AKI may represent a unique form of AKI distinctly characterized, at least in the
early phase, by marked hyperemia. This loss of GFR, in the setting of increased global RBF,
would theoretically require reduction in glomerular filtration pressure by preferential efferent
arteriolar dilation in excess of that occurring in the afferent arterioles.37 This would lead to
loss of GFR, reduced creatinine clearance and oliguria.
To test this hypothesis, we would need to measure RBF in critically ill patients with septic
shock and AKI. Regrettably, there are few data that have evaluated RBF in human septic
shock. Simply stated, we do not know what the characteristic profile of RBF in human septic
AKI is, yet, this is a critical issue to investigate in order to better understand the pathophysiology
of septic AKI. Kidney hypoperfusion may still be a significant contributor to AKI in sepsis
when associated with a hypodynamic circulation, however, hypoperfusion may not play a
significant role in during hyperdynamic sepsis.
Regional Renal Blood Flow and Bioenergetics in Septic AKI

While global RBF has been found in experimental studies to be either preserved or increased
in hyperdynamic sepsis, there theoretically could be preferential reductions in regional RBF
within the kidney vasculature (i.e., cortical vs. medullary). In a large sheep model of
hyperdynamic septic AKI using laser Doppler flowmetry to continuously monitor cortical and
medullary RBF, Di Giantomasso et al found flow to both regions was well preserved and that
administration of a vasoconstrictor (i.e., norepinephrine) actually augmented flow. 44,49
Interestingly, these findings would appear to challenge the prevailing view that the medulla
is ischaemic during hyperdynamic septic AKI.37
While both global and regional RBF appear preserved in hyperdynamic septic AKI,
theorectically, the kidneys could still be susceptible to bioenergetic failure due to maladaptive
adenosine triphosphate (ATP) utilization, uptake and/or depletion. In a series of studies applying
a magnetic resonance spectroscopy technique with simultaneous measurement of RBF, May
et al were also able to show that ATP content is preserved during hyperdynamic sepsis in a
sheep model.50,51 Importantly, these data suggest that global ischaemia or bioenergetic failure
are not likely to primary determinant of loss of GFR in hyperdynamic septic AKI.
Non-Hemodynamic Kidney Injury

As mentioned above, neither global nor regional changes to RBF are likely to contributor to
septic AKI alone. Additional factors must therefore play a role and are likely immunologic
and/or toxic in nature. Sepsis is characterized by the release of a vast array of inflammatory
cytokines, arachidonate metabolites, vasoactive mediators, thrombogenic agents, and other
biologically active compounds. Accumulating experimental data suggest that these various
mediators and neuroendocrine mechanisms might be involved in the pathogenesis of organ
dysfunction in sepsis.21,37,52 These immunologic and/or toxic mechanisms are likely to be
important, perhaps even more important than previously thought, for mediating and/or even
propagating kidney injury in sepsis, than the hemodynamic factors described above. Recent
data suggest renal cell apoptosis may have a contributory role in septic AKI.53-55 In addition,
several context-specific and patient related factors may also contribute to development of
septic AKI including baseline kidney function, co-morbid vascular disease, and administration
of nephrotoxic drugs and/or radiocontrast media.
Urinary Changes in Septic AKI

Urinary biochemistry and microscopy are among the oldest and most time-honored tests
used in nephrology. Numerous reviews, textbooks and articles have long suggested that it is
possible for various urinary tests to distinguish acute tubular necrosis (ATN) (i.e., structural
injury) from pre-renal azotemia (PRA) (i.e., functional injury).39,56,57 Yet, considering the
emerging data that suggest septic AKI may be somewhat unique, can this paradigm hold true
for septic AKI? Recently, Bagshaw et al completed a systematic review of the urinary findings
seen in experimental models of sepsis to assess their diagnostic and prognostic value. 22 No
urinary test of biochemistry, derived indices or microscopy appeared to have diagnostic
accuracy, prognostic value or clinical utility. Likewise, in a systematic review evaluating the
values of these urinary tests in humans, there was generally a significant paucity of data
along with a wide array of findings in septic AKI.58 These findings support the concept that
the biochemical analysis of urine (i.e., urinary sodium, urea, creatinine) and/or use of calculated
indices of tubular function (i.e., fractional excretion of sodium, fractional excretion of urea)
have limited diagnostic and/or prognostic value in septic AKI. Additional clinical investigation
is clearly needed to better understand the potential diagnostic and/or prognostic role of
urinalysis in septic AKI. In this regard, emerging biomarkers of AKI will likely prove more
valuable.59
Histopathology of Septic AKI

Obtaining histopathologic data in septic AKI would be a fundamental strategy for providing
valuable insight into the pathophysiology of septic AKI. In a recent systematic review on the
histopathology of septic AKI, Langenberg et al identified only six human and 14 animal studies
that described the relevant histopathologic changes in septic AKI.24 Among the 184 patients
in the six human studies, only 26 (22%) had histopathologic features suggestive of
ATN.29,60-62 In general, across both experimental and human studies, there was no consistent
or typical kidney histopathological pattern in septic AKI. While the majority of studies showed
some general but mild histopathological changes, ATN was an uncommon finding. These
findings fail to confirm the widely held assumption that ATN is the most common
histopathological pattern described in septic AKI. Moreover, these findings draw attention to
the need for additional experimental and clinical studies that investigate the histopathology
associated with septic AKI.
CONCLUSION
Sepsis is the most important contributing factor to AKI in critically ill patients. Distinguishing
septic from non-septic AKI has clinical relevance. Emerging data have suggested septic AKI
may be characterized by a distinct pathophysiology compared with non-septic AKI. Moreover,
critically ill patients with septic AKI show numerous differences in clinical, acute physiologic
and laboratory parameters when compared with non-septic AKI or sepsis alone. Importantly,
septic AKI appears to independently predict a more complicated course and higher mortality.
For these reasons, we believe it important that critical care physicians have an appreciation
of the epidemiology and pathophysiology of the syndrome of septic AKI.
REFERENCES
1. Brivet FG, Kleinknecht DJ, Loirat P, Landais PJ. Acute renal failure in intensive care units—causes, outcome,
and prognostic factors of hospital mortality; a prospective, multicenter study. French Study Group on
Acute Renal Failure. Crit Care Med 1996;24:192-8.
2. de Mendonca A, Vincent JL, Suter PM, Moreno R, Dearden NM, Antonelli M, et al. Acute renal failure
in the ICU: risk factors and outcome evaluated by the SOFA score. Intensive Care Med 2000;26:915-21.
3. Metnitz PG, Krenn CG, Steltzer H, Lang T, Ploder J, Lenz K, et al. Effect of acute renal failure requiring
renal replacement therapy on outcome in critically ill patients. Crit Care Med 2002; 30:2051-8.
4. Uchino S, Kellum JA, Bellomo R, Doig GS, Morimatsu H, Morgera S, et al. Acute renal failure in critically
ill patients: a multinational, multicenter study. JAMA 2005;294:813-8.
5. Bagshaw SM, George C, Dinu I, Bellomo R. A Multi-Centre Evaluation of the Rifle Criteria for Early Acute
Kidney Injury in Critically Ill Patients. Nephrol Dial Transplant 2007.
6. Cruz D, Bolgan I, Perazella MA, Bonello M, de Cal M, Corradi V. North East Italian Prospective Hospital
Renal Outcome Survey on Acute Kidney Injury (NEiPHROS-AKI): targeting the problem with the RIFLE
criteria. Clin J Am Soc Nephrol 2007;2:418-25.
7. Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, et al. RIFLE criteria for
acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit
Care 2006;10:R73.
8. Ostermann M, Chang RW. Acute kidney injury in the intensive care unit according to RIFLE. Crit Care
Med 2007;35:1837-43.
9. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P: Acute renal failure—definition, outcome measures,
animal models, fluid therapy and information technology needs: the Second International Consensus
Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care 2004;8:R204-12.
10. Ali T, Khan I, Simpson W, Prescott G, Townend J, Smith W, et al. Incidence and outcomes in acute kidney
injury: a comprehensive population-based study. J Am Soc Nephrol 2007;18:1292-8.
11. Bagshaw SM, George C, Bellomo R. Changes in the incidence and outcome for early acute kidney injury
in a cohort of Australian intensive care units. Crit Care 2007;11:R68.
12. Xue JL, Daniels F, Star RA, Kimmel PL, Eggers PW, Molitoris BA, et al. Incidence and mortality of acute
renal failure in Medicare beneficiaries, 1992 to 2001. J Am Soc Nephrol 2006;17:1135-42.
13. Bellomo R. The epidemiology of acute renal failure: 1975 versus 2005. Curr Opin Crit Care 2006;12:557-
60.
14. Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of sepsis in the United States from 1979
through 2000. N Engl J Med 2003;348:1546-54.
15. Vincent JL, Sakr Y, Sprung CL, Ranieri VM, Reinhart K, Gerlach H, et al. Sepsis in European intensive
care units: results of the SOAP study. Crit Care Med 2006;34:344-53.
16. Hoste EA, Lameire NH, Vanholder RC, Benoit DD, Decruyenaere JM, Colardyn FA. Acute renal failure
in patients with sepsis in a surgical ICU: predictive factors, incidence, comorbidity, and outcome. J Am
Soc Nephrol 2003;14:1022-30.
17. Lopes JA, Jorge S, Resina C, Santos C, Pereira A, Neves J, et al. Acute renal failure in patients with
sepsis. Crit Care 2007;11:411.
18. Neveu H, Kleinknecht D, Brivet F, Loirat P, Landais P: Prognostic factors in acute renal failure due to
sepsis. Results of a prospective multicentre study. The French Study Group on Acute Renal Failure. Nephrol
Dial Transplant 1996;11:293-9.
19. Yegenaga I, Hoste E, Van Biesen W, Vanholder R, Benoit D, Kantarci G, et al. Clinical characteristics of
patients developing ARF due to sepsis/systemic inflammatory response syndrome: results of a prospective
study. Am J Kidney Dis 2004;43:817-24.
20. Bagshaw SM, Lapinsky S, Dial S, Arabi Y, Dodek P, Wood G, et al. Acute kidney injury in septic shock:
clinical outcomes and impact of duration of hypotension prior to initiation of antimicrobial therapy. Intensive
Care Med 2008:Epub: Dec 9.
21. Wan L, Bellomo R, Di Giantomasso D, Ronco C. The pathogenesis of septic acute renal failure. Curr Opin
Crit Care 2003;9:496-502.
22. Bagshaw SM, Langenberg C, Wan L, May CN, Bellomo R. A systematic review of urinary findings in
experimental septic acute renal failure. Crit Care Med 2007;35:1592-8.
23. Bellomo R, Bagshaw S, Langenberg C, Ronco C. Pre-renal azotemia: a flawed paradigm in critically ill
septic patients? Contrib Nephrol 2007;156:1-9.
24. Langenberg C, Bagshaw SM, May CN, Bellomo R. The histopathology of septic acute kidney injury: a
systematic review. Crit Care 2008;12:R38.
25. Langenberg C, Bellomo R, May C, Wan L, Egi M, Morgera S. Renal blood flow in sepsis. Crit Care
2005;9:R363-74.
26. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, et al. Definitions for sepsis and organ
failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference
Committee. American College of Chest Physicians/Society of Critical Care Medicine. Chest 1992;101:1644-
55.
27. Hoste EA, Cruz DN, Davenport A, Mehta RL, Piccinni P, Tetta C, et al. The epidemiology of cardiac
surgery-associated acute kidney injury. Int J Art Organ 2008;31:158-65.
28. Ricci Z, Cruz D, Ronco C. The RIFLE criteria and mortality in acute kidney injury: A systematic review.
Kidney Int 2008;73:538-46.
29. Diaz de Leon M, Moreno SA, Gonzales Diaz DJ, Briones GJ. Severe sepsis as a cause of acute renal
failure. Nefrologia 2006;26:439-44.
30. Oppert M, Engel C, Brunkhorst FM, Bogatsch H, Reinhart K, Frei U, et al. Acute renal failure in patients
with severe sepsis and septic shock a significant independent risk factor for mortality: results from the
German Prevalence Study. Nephrol Dial Transplant 2007.
31. Rangel-Frausto MS, Pittet D, Costigan M, Hwang T, Davis CS, Wenzel RP: The natural history of the
systemic inflammatory response syndrome (SIRS). A prospective study. JAMA 1995; 273:117-23.
32. Bagshaw SM, George C, Bellomo R. Early acute kidney injury and sepsis: a multicentre evaluation. Crit
Care 2008;12:R47.
33. Bagshaw SM, Laupland KB, Doig CJ, Mortis G, Fick GH, Mucenski M, et al. Prognosis for long-term
survival and renal recovery in critically ill patients with severe acute renal failure: a population-based
study. Crit Care 2005;9:R700-R709.
34. Bagshaw SM, Uchino S, Bellomo R, Morimatsu H, Morgera S, Schetz M, et al. Septic acute kidney injury
in critically ill patients: clinical characteristics and outcomes. Clin J Am Soc Nephrol 2007;2:431-9.
35. Payen D, de Pont AC, Sakr Y, Spies C, Reinhart K, Vincent JL. A positive fluid balance is associated wiht
a worse outcome in patients with acute renal failure. Crit Care 2008;12:R74.
36. Van Biesen W, Yegenaga I, Vanholder R, Verbeke F, Hoste E, Colardyn F, et al. Relationship between fluid
status and its management on acute renal failure (ARF) in intensive care unit (ICU) patients with sepsis:
a prospective analysis. J Nephrol 2005;18:54-60.
37. Wan L, Bagshaw SM, Langenberg C, Saotome T, May C, Bellomo R. Pathophysiology of septic acute
kidney injury: what do we really know? Crit Care Med 2008;36:S198-203.
38. Abuelo JG. Normotensive ischemic acute renal failure. N Engl J Med 2007;357:797-805.
39. Schrier RW, Wang W. Acute renal failure and sepsis. N Engl J Med 2004;351:159-69.
40. Kikeri D, Pennell JP, Hwang KH, Jacob AI, Richman AV, Bourgoignie JJ. Endotoxemic acute renal failure
in awake rats. Am J Physiol 1986;250:F1098-F1106.
41. Brenner M, Schaer GL, Mallory DL, Suffredini AF, Parrillo JE. Detection of renal blood flow abnormalities
in septic and critically ill patients using a newly designed indwelling thermodilution renal vein catheter.
Chest 1990;98:170-9.
42. Lucas CE, Rector FE, Werner M, Rosenberg IK. Altered renal homeostasis with acute sepsis. Clinical
significance. Arch Surg 1973;106:444-9.
43. Ravikant T, Lucas CE. Renal blood flow distribution in septic hyperdynamic pigs. J Surg Res 1977;22:294-
8.
44. Di Giantomasso D, Morimatsu H, May CN, Bellomo R. Intrarenal blood flow distribution in hyperdynamic
septic shock: Effect of norepinephrine. Crit Care Med 2003;31:2509-13.
45. Langenberg C, Wan L, Bagshaw SM, Egi M, May CN, Bellomo R. Urinary biochemistry in experimental
septic acute renal failure. Nephrol Dial Transplant 2006;21:3389-97.
46. Langenberg C, Wan L, Egi M, May CN, Bellomo R. Renal blood flow in experimental septic acute renal
failure. Kidney Int 2006;69:1996-2002.
47. Langenberg C, Bellomo R, May CN, Egi M, Wan L, Morgera S. Renal Vascular Resistance in Sepsis.
Nephron Physiol 2006;104:1-11.
48. Langenberg C, Wan L, Egi M, May CN, Bellomo R. Renal blood flow and function during recovery from
experimental septic acute kidney injury. Intensive Care Med 2007;33:1614-8.
49. Di Giantomasso D, Morimatsu H, May CN, Bellomo R. Increasing renal blood flow: low-dose dopamine
or medium-dose norepinephrine. Chest 2004;125:2260-7.
50. May C, Wan L, Williams J, Wellard MR, Pell G, Jackson G, et al. A technique for the measurement of
renal ATP in a large animal model of septic shock. Int J Artif Organs 2005;28:16-21.
51. May C, Wan L, Williams J, Wellard MR, Pell G, Langenberg C, et al. A technique for the simultaneous
measurement of renal ATP, blood flow and pH in a large animal model of septic shock. Crit Care Resusc
2007;9:30-3.
52. Schumer M, Colombel MC, Sawczuk IS, Gobe G, Connor J, O’Toole KM, et al. Morphologic, biochemical,
and molecular evidence of apoptosis during the reperfusion phase after brief periods of renal ischemia.
Am J Pathol 1992;140:831-8.
53. Jo SK, Cha DR, Cho WY, Kim HK, Chang KH, Yun SY, et al. Inflammatory cytokines and lipopolysaccharide
induce Fas-mediated apoptosis in renal tubular cells. Nephron 2002;91:406-15.
54. Messmer UK, Briner VA, Pfeilschifter J. Tumor necrosis factor-alpha and lipopolysaccharide induce apoptotic
cell death in bovine glomerular endothelial cells. Kidney Int 1999;55:2322-37.
55. Rodriguez-Wilhelmi P, Montes R, Matsukawa A, Nariuchi H, Hurtado V, Montes M, et al. Tumor necrosis
factor-alpha inhibition reduces CXCL-8 levels but fails to prevent fibrin generation and does not improve
outcome in a rabbit model of endotoxic shock. J Lab Clin Med 2003;141:257-64.
56. Oxford Textbook of Clinical Nephrology. 3rd edn. Oxford: Oxford University Press 2005.
57. Perazella MA, Coca SG, Kanbay M, Brewster UC, Parikh CR. Diagnostic value of urine microscopy for
differential diagnosis of acute kidney injury in hospitalized patients. Clin J Am Soc Nephrol 2008;3:1615-
9.
58. Bagshaw SM, Langenberg C, Bellomo R. Urinary biochemistry and microscopy in septic acute renal failure:
a systematic review. Am J Kidney Dis 2006;48:695-705.
59. Bagshaw SM, Langenberg C, Haase M, Wan L, May CN, Bellomo R. Urinary biomarkers in septic acute
kidney injury. Intensive Care Med 2007;33:1285-96.
60. Hotchkiss RS, Swanson PE, Freeman BD, Tinsley KW, Cobb JP, Matuschak GM, et al. Apoptotic cell
death in patients with sepsis, shock, and multiple organ dysfunction. Crit Care Med 1999; 27:1230-51.
61. Mustonen J, Pasternack A, Helin H, Pystynen S, Tuominen T. Renal biopsy in acute renal failure. Am J
Nephrol 1984;4:27-31.
62. Rosenberg IK, Gupta SL, Lucas CE, Khan AA, Rosenberg BF. Renal insufficiency after trauma and sepsis.
A prospective functional and ultrastructural analysis. Arch Surg 1971;103:175-83.
Anticoagulation During CRRT—Does Citrate Offer an Advantage? 97
Ten
Anticoagulation During CRRT—
Does Citrate Offer an Advantage?
Shamik H Shah, Ravindra L Mehta
INTRODUCTION
In humans, a complex system of plasma, cellular, and endothelial factors maintains the delicate
balance between haemostasis and hemorrhage.1 Coagulation is the normal process occurring
when vascular injury results in formation of a fibrin clot; thrombosis refers to the pathological
formation of clot in response to injury, stasis, and hypercoagulability.1 This fine balance between
procoagulant and anticoagulant activity is disturbed when blood is passed through an
extracorporeal circuit (ECC). Clotting of continuous renal replacement therapy (CRRT) circuits
could be caused by increased activation of coagulation, initiated either by the (intrinsic) contact
activation pathway or the (extrinsic) tissue factor/activated factor VII (FVIIa) pathway, or by
low activity of the endogenous anticoagulant pathways, such as the anti-thrombin system,
the protein C/protein S system and the tissue factor pathway inhibitor system. In addition,
decreased fibrinolysis could also contribute to clotting of extracorporeal circuits (ECC).2
In addition to these, a number of factors are known to contribute to the initiation of
coagulation of the extracorporeal circuit in CRRT.3 These factors are described in Table 10.1.
NEED FOR ANTICOAGULATION

Anticoagulation is necessary despite the increased risk in critically ill patients. The term CRRT
suggests a treatment without interruption. However, interruptions do occur and the duration
of the circuit “down-time” may influence solute and fluid control. Fealy and colleagues
conducted a prospective study to ascertain the percentage of “down-time” in an ICU setting.4
They observed a mean circuit “down-time” of 22% or 5.27 hours per day with circuit clotting
as the main cause. There was a strong correlation between “circuit down-time” (e.g., treatment
interruptions) and increase in plasma urea and creatinine.4
In another prospective study, the authors observed a median down-time of 3 hours. There
was a significant inverse correlation between down-time and upward change in creatinine
and urea over each 24-hr time cycle.5
Table 10.1: Factors affecting coagulation of the extracorporeal circuit3
Patient factors Vascular access Extracorporeal circuit Treatment characteristics
Platelet count and Malposition and Filter characteristics Blood flow 70

function11 kinking (material, geometry,
surface, hollow-fibre
length, pore diameter,
heparin-coating)68,69
Activation of coagulation Vascular filling Tubing (material, coating)72 Filtration fraction73
via tissue factor pathway71
Decrease in concentration Changes in patient Blood-air contact in the Pre or Post-dilution23,76,77
of natural anticoagulants74 position bubble trap chamber75
Antithrombin, Heparin Negative thoracic Intermittent blood flow Nursing factors79
cofactor II, Tissue factor pressure reductions73
pathway inhibitor,
Activated protein C78
Inhibition of fibrinolysis15 Geometry, length and Reaction time to
diameter of the circuit alarms81
catheter80
Transfusion of blood
products2
In a prospective, randomized trial done at our center, the mean duration of CRRT treatment
was 16.1 hours, suggesting a down-time of nearly 8 hours every day.6
These findings underline the need for a safe and effective anticoagulation strategy for
patients undergoing CRRT.
The aim of anticoagulation during CRRT is to prolong filter and circuit longevity, thereby
improving solute clearance and reducing material7 and nursing cost. The goals of ideal
anticoagulation are the following:
i. Anticoagulation with high anti-thrombotic potential with low haemorrhagic risk;
ii. Brief anticoagulant action, limited to the CRRT circuit;
iii. Anticoagulation monitoring should be easy and suited for bedside use;
iv. Long-term use of anticoagulation should not be associated with severe systemic side
effects;
v. An antidote should be available in case of overdose.
COAGULATION DERANGEMENTS IN CRITICALLY ILL PATIENTS WITH AKI

In critically ill patients with sepsis and multi-organ failure, the imbalance in the normal clotting
mechanisms can be extreme, with activation of multiple inflammatory pathways, and down
regulation of anticoagulant pathways.8,9 Gando et al showed that extrinsic coagulation pathway
is activated in patients with severe sepsis and septic shock.10 They also showed enhanced
thrombin generation and activation, and fibrin formation in these patients.10
Various investigators have studied the coagulation abnormalities in patients with acute
kidney injury and critical illness. Both bleeding and clotting abnormalities may occur. While
platelet counts can be normal, altered platelet function is often encountered and can contribute
to increased bleeding risk. On the other hand, enhanced platelet activation is seen in patients
with AKI11 due to physical contact with the ECC or due to interaction with the activated
clotting cascade. Platelets also undergo transient morphological changes secondary to
circulating through the ECC. These changes are consistent with primary, reversible aggregation
and activation.11 Adherence can occur quickly when platelets are exposed to the ECC foreign
surface. Studies have shown that patients with a higher baseline platelet count require more
heparin to achieve comparable filter life.12,13 while patients with lower platelet counts have
a decreased risk of filter clotting.2
Patients with critical illness have been found to have increased levels of procoagulant
factors VII and VIII, decreased levels of coagulation inhibitors, such as anti-thrombin III (ATIII),
Proteins C and S, and impaired fibrinolysis.14 Voss and colleagues showed that the activation
of coagulation is accompanied by an activation of fibrinolysis in the microcirculation, but that
systemically the increased inhibitors of fibrinolysis (PAI, alpha 2-antiplasmin) induce a decrease
of the fibrinolytic capacity of the blood. The severity of the disease determines the extent of
the alterations.15 Dialysis can be associated with the release of other pro-coagulant substances
such as vWF, 6-keto prostaglandin F1-alpha, and tissue plasminogen activator from endothelial
cells.14,16 Additionally, during dialysis, there is an increase in the endothelial release of tissue
plasminogen activator causing an increased in fibrinolytic activity, primarily as a result of the
bioincompatibility of materials in the extracorporeal circuit.17 In study by Cardigan and
colleagues, it was shown that activation of FVII occurred during CVVH, which was related to
levels of TFPI, and was coincidental to thrombin generation.18
It is apparent that the complex interplay of underlying disease process and exposure to the
ECC determine the coagulant and bleeding risk for patients with AKI. Knowledge of these
events can help guide the optimum choice of anticoagulant for renal replacement therapy.
AVAILABLE ANTICOAGULATION STRATEGIES

(A list of available anticoagulation strategies is given in Table 10.2).
No Anticoagulation
Several trials have looked at the use of no anticoagulation in patients undergoing CRRT.
Median filter life has been found to vary from 19 to 53 hours; the prime variable for increased
filter life being a low platelet count (<60,000/ml).19-23 Current data suggests that no
anticoagulation be used in patients with high risk for bleeding complications (<48 hours
postoperative, platelet count < 60,000/ml, aPTT > 45s).24
Table 10.2: Available anticoagulation strategies
Strategy Filter Initial Maintenance Monitoring Target Filter life3 Advantages Disadvantages 100
priming dose dose
No anti- None or with NA NA Filter thrombosis, NA 3 to 70h No monitoring or Inadequate filter

coagulation heparinized Filter pressure risks associated with patency in all other
saline other anticoagulation patients.
(5-8 IU/ml) regimens; adequate
filter patency especially
in high risk patients
with decreased platelets
Heparin Heparinized 2000- 3-15 IU/Kg/hr aPTT/ACT 1-1.4 15 to 50h Familiarity, effective Variability in dosing
saline 5000 pre-filter times anticoagulation, ease between patients,
(5-8 IU/ml) IU pre- normal of monitoring in Increased risk of
filter catheterized patient, haemorrhagic
short half-life, effects complications, HIT,

quickly reversed with Adrenal
protamine insufficiency
Low dose None or None 500 IU/hr or Maintain 19 to 32h Familiarity, defensive Continued risk of
heparin as above 5 or 10 IU/ normal or monitoring of AC rather bleeding, ineffective
Kg/hr baseline aPTT than chasing the aPTT filter patency
Regional AC effect of heparin Protamine adminis-

heparinization restricted to the ECC, tered to reverse the
reduces the risk of effects of heparin may
systemic patient bleeding cause hypotension
LMWH Heparinized Bolus of Pre-filter Anti-Xa activity 0.25- 15.4 to AC response more More costly and
dalteparin Saline or 15-25 administration 0.35 46.8h reliable and predict- pitfalls of heparin
1.5-2.5 units/ IU/Kg at 10 units/Kg/hr. IU/mL able than heparin. persist. Special
mL Dalteparin Reduced risk of syste- coagulation assays
mic hemorrhage. required. Effects
Incidence of HIT and half-life vary
type II less likely
LMWH 0.15 0.05 mg/kg/hr Anti-Xa activity 0.25-0.35 18 to As above As above

enoxaparin mg/kg IU/mL 30 h
[range
0.75 to
72h]
Danaparoid AS above 750 U 1-2 U/kg/hr As above As above As above As above
Contd...
Table 10.2 contd...
Strategy Filter Initial Maintenance Monitoring Target Filter life 3 Advantages Disadvantages
priming dose dose
Argatroban 250 μg/Kg 0.5 to 2 μg/ aPTT 1 to 1.4 No signi- Safe in renal failure. No antidote for
Kg/min times ficant data Established as treatment bleeding and aPTT
normal for HIT may not absolutely
correlate with
anticoagulation
Citrate (TSC None or with 140-160 Vary rate of Maintain post-filter Post-filter 24 to 124h Ease of administration, Labor intensive,
or ACD-A heparinized ml/hr of citrate in ionized calcium at ionized ease of monitoring, low Difficulty in obtaining
administered saline if no 4% TSC 5-10 ml/hr 0.2-0.3 mmol/L calcium at incidence of adverse appropriate dialysate
alone pre-filter) HIT present given pre- increments to Use systemically 0.2.-0.3 events, ease of disconti- and citrate solution
filter. maintain desired administered mmol/L. nuation if complications (depends on
(3-8% of post-filter calcium to maintain arise. There may be pharmacy avail-
the blood ionized calcium systemic ionized improved biocompati- ability); potential for
flow rates) rates calcium within bility with citrate metabolic alkalosis,
the normal range hypernatremia, hyper-
Or ACD-A of the lab (1.0-1.2 /hypocalcaemia, or
citrate solu- mmol/L) citrate toxicity.
tion at 250
ml/hr
Citrate None or with Citrate Varied depend- Post-filtered ionized 48 to >72h Simpler than separate Combining the goal
(administered heparinized can be ing on prescribed calcium ranging citrate administration. of filter anticoagula-
in replacement saline if no given as effluent dose from 0.2-0.5 mmol/L. Ability to utilize standar- tion and dialysis dose
fluid pre-filter) HIT present a 0.4, 0.5 (i.e. 35 ml/kg/hr) Systemic ionized dized dialysate or leads to complexities
or 0.67% or desired hourly calcium in normal replacement fluids. in altering the rates of
solution fluid balance range for laboratory. Similar filter longevity various fluids. The
and metabolic stability metabolic complica-
as with citrate alone tions mentioned above
administration. are still a concern.
Hirudin None No loading 0.005-0.01 ECT 80-100s 11-45 h Can be used in Clinical experience
mg/kg/hr, patients having or limited. Half-life
after 1-2 days suspected of having considerably pro-
0.005 mg/kg/hr HIT type II longed in patients
or intermittent with renal
bolus 0.002 insufficiency
Anticoagulation During CRRT—Does Citrate Offer an Advantage?
g/kg (preferred)
Abbreviations: aPTT, activated partial Thromboplastin Time; HIT, heparin induced thrombocytopenia; TSC, tri-sodium citrate; ACD-A, anticoagulant citrate dextrose-
A; LMWH, low molecular weight heparin.
101
Fig. 10.1: Circuit Diagram for continuous venovenous renal replacement (CVVHF, CVVHD, CVVHDF)
systemic Heparin anticoagulation. Blood is pumped from a venous catheter into a dialysis filter at a rate
of 100-150 ml/min. If dialysate is used it is typically run countercurrent at approximately 1/6 the blood flow rate.
At our institution, 500ml of Normal saline is given as a dilutional agent pre-filter. If replacement fluid is given
it is typically done post-filter at our institution. The composition of the replacement fluid is varied by acid base
status of the patient
Unfractionated Heparin (UFH)

UFH is the most commonly used anticoagulant in CRRT.25 It is a mixture of heparin molecules
with different sizes (5-30kDa). It acts by a 1000-fold potentiation of antithrombin, thereby
inhibiting factors Xa and IIa (thrombin). The smaller fragments bind to factor Xa, while the
larger fragments inhibit factor IIa. UFH acts directly on coagulation cascade, is taken up by
the reticuloendothelial system, metabolized by the liver and is excreted by the kidneys. Plasma
half-life varies between 30 minutes and 3 hours.3
Conventional strategies involve priming the ECC with heparinized saline (5-10IU/ml). A
bolus of 2000-5000 IU is injected into the circuit at commencement of the CRRT procedure.
Continuous infusion is started at the arterial side of the ECC at 3-15IU/kg/hr to maintain an
activated partial thromboplastin time (aPTT) 1.5-2 times the upper limit of normal 20,26
(Fig. 10.1).
The efficacy and anticoagulant effect of UFH is measured by activated partial thromboplastin
time (aPTT). The advantages of UFH are:
• Widely available.
• Low cost.
• Large clinical experience.
• Effective anticoagulant.
• aPTT measurement generally available.
• Effects can be neutralized with protamine.
Current data comparing heparin to alternate forms of anticoagulants reveal that the risk benefit
ratio of heparin in critically ill patients is unfavorable. One needs to be aware of the following
factors when UFH is used in critically ill patients:
Heparin is dependent on AT III for its action.27 Critically ill patients, particularly those with
sepsis have an acquired deficiency of AT III which reduces the substrate available for heparin
action.28 AT III levels have been found low in patients undergoing CRRT.29 Moreover, heparin
may not provide ideal anticoagulation in patients that have intrinsic clotting system activation,
AT III deficiency or evidence of intravascular coagulation.30
• Half-life of UFH is prolonged in critical illness and may contribute to bleeding risk. Heparin
is associated with serious bleeding complications in 10-50% of patients. Mortality is reported
to be as high as 15%.31 Transfusion requirements go up in patients receiving heparin as
an anticoagulant. This, in turn, leads to decrease filter survival.32
• Heparin is immunogenic and causes Heparin-induced thrombocytopenia (HIT) by antibody
against platelets. It is not only a common but also a potentially serious drug adverse effect.
Unlike other drug induced thrombocytopenias, HIT does not usually cause bleeding, but
instead causes thrombosis. Thrombosis in HIT can lead to limb gangrene or even death.33
The incidence is reported to be between 1 and 4% of patients treated with Heparin.34,35
Also, the incidence of HIT type II is reported to be higher in critically ill patients when
compared to patients receiving long-term intermittent haemodialysis.36
• Unfractionated heparin binds non-specifically to macrophages, endothelial cells and plasma
proteins, which causes it to have complex dose-dependent pharmacokinetics and an
unpredictable anticoagulant effect.37 Also, this non-specific binding of unfractionated heparin
to plasma proteins is the most common cause of “heparin resistance”.38
• aPTT is not a reliable predictor of bleeding. It is poorly standardized and is affected by
numerous factors that are unrelated to the heparin effect.39 The larger UFH fragments
(anti-IIa) are cleared more rapidly than the smaller fragments (anti-Xa), resulting in the
prolongation of anti-Xa effect in vivo. As the aPTT is much more sensitive to anti-IIa, it
may be normal, while UFH still has anticoagulant effect.40-42
REGIONAL CITRATE ANTICOAGULATION (RCA)

Free calcium is required for all components of the coagulation cascade. Consequently, binding
calcium with citrate offers a mechanism to prevent coagulation. Regional citrate anticoagulation
involves the infusion of citrate ions into the extracorporeal circuit. On entry into the circuit,
ionized calcium combines with citrate to form calcium-citrate complexes of non-ionized calcium.
As blood travels along the circuit, the level of ionized calcium is reduced as the level of non-
ionized calcium rises and prevents the coagulation of blood.27 Citrate is converted to citric
acid, yielding bicarbonate, and is further metabolized in the citric acid (Krebs) cycle in liver,
skeletal muscle and renal cortex or is metabolized to glucose (gluconeogenesis).
Fig. 10.2: Circuit Diagram for UCSD CVVHDF regional citrate anticoagulation. The system is primed
with saline only. Blood is pumped from a venous catheter into a dialysis filter at a rate of 100-150ml/min. If
dialysate is used it is typically run countercurrent at approximately 1/6 the blood flow rate. At our institution,
500ml of Normal saline is given as a dilutional agent pre-filter. If replacement fluid is given it is typically done
post-filter at our institution. The composition of the replacement fluid is varied by acid base status of the patient.
Heparin is infused pre-filter at a rate to maintain the post-filter aPTT at 1.5-2x upper limit of normal and the
Protamine is administered at a dose of 100:1 (units of heparin: mg protamine) to maintain the patient’s aPTT
<45. TSC, tri sodium citrate; CVC, central venous catheter
Systemic anticoagulation is avoided by limiting the effect of citrate to the extracorporeal

circuit. Citrate is infused at the entry port of the circuit. On return to the patient, the total
body calcium plus other divalent cations inundate the circulating citrate thereby normalizing
the serum calcium. This is further supplemented by calcium administration at a non-CRRT
related venous line in the patient (typically a separate central venous catheter).
At our institute, Citrate solution (tri-sodium citrate or ACD) is administered at the site of
blood entry into the CRRT circuit; the rate is adjusted to a target of post-filter ionized calcium
level of 0.2-0.3. Calcium is supplemented via a separate central venous catheter (Fig. 10.2).
The use of citrate for anticoagulation in dialysis was first described by Morita in 1961.43
It has been used as an effective anticoagulant in CRRT circuits since the 1990s.44,45 A
variety of methods have been described in published literature for regional citrate
anticoagulation.31,32,44,46,47
Potential complications are hypernatremia, hypo-/hypercalcaemia, hypomagnesemia,
alkalosis, and citrate toxicity leading to metabolic acidosis.48 This is due to the following reasons:
• Citrate is a buffer, and is used as such, in addition to being an anticoagulant.3
• The loss of calcium, bound to citrate is more than with haemodialysis or Haemofiltration
with heparin.
• Tri-sodium citrate (TSC) contains a substantial amount of sodium.49
• If liver and skeletal muscle fail to metabolize citric acid, accumulation of citrate occurs.50
A number of studies have looked at the effectiveness of RCA on circuit survival. These
studies are summarized in Table 10.3.
Advantages of RCA
• Circuit and filter survival are better than with UFH as outlined in Table 10.3.
• As there is no systemic anticoagulation, there is less risk of bleeding.51,52
• Monitoring is fast and cost effective as ionized measurement of calcium is available with
most blood gases.53
• Life-threatening complication rate is lower when compared to heparin.22
• Less activation of leucocytes and coagulation than heparin.54,55
• Citrate mitigates the inflammatory pathway,56,57 and lowers oxidative stress.57
Commercially Available Citrate Solutions

The evolution of CRRT techniques has been iterative over the past three decades. It has
evolved from a renal supportive technique for haemodynamically unstable patients to a
metabolic support device in critical care medicine. One of the most important developments
has been the introduction of RCA to avoid effects of systemic heparinization.44,45 Although
several other methods of anticoagulation have been described, each has its own unique
problems. For example, saline flushes alone have a very high rate of filter clotting.58 Regional
anticoagulation using heparin and protamine is time consuming and expensive. Also, protamine
can itself cause hemodynamic instability or coagulopathies, and does not alter the risk for
HIT. Hirudin and Argatroban have no specific antidotes. In this scenario, RCA is an attractive
alternative to UFH. There is great interest in RCA for CRRT, but many units are prevented
from adopting published protocols because of the need of extemporaneous solution mixing
on a large scale.53 One of the limitations to the wide-spread use of Citrate is the paucity of
commercially available solutions.
For use in critically ill patients, all CRRT solutions must be sterile and pyrogen free.59
Physicians have to rely on local hospital pharmacies to prepare prescribed solutions. However,
despite the best intentions, most hospital pharmacies cannot match the quality control of
regulated, commercial pharmaceutical manufacturers.60,61
Three solutions have labeling approved by the Food and Drug Administration for use as
dialysate solutions in CRRT. They are also approved for use as CRRT replacement solutions.
These solutions are Normocarb (Dialysis Solutions, Inc., Richmond, OH) and two formulations
of PrismaSate, BK0/3.5 and BGK2/0 (Gambro Renal Products, Lakewood, CO). PrismaSate
is approved and marketed in Europe and Canada under the brand name Hemosol (Hospal,
Lyon, France) for use as both dialysate and replacement solutions in CRRT.
There have been few papers on the effect of commercially available solutions on patient
survival. However, two studies have shown improvement in cardiovascular stability and control
of metabolic acidosis with bicarbonate buffered solutions.62,63 These commercially available
solutions have been found to be more cost-effective than custom made solutions.64
106
Table 10.3: Comparison of regional citrate anticoagulation on circuit survival
Study design CRRT Citrate dose UFH dose Circuit survival- Circuit survival- Reference
modality Citrate UFH
Retrospective, controlled CAVHD 24 mmol/h 3-12 U/kg/hr 63.1 + 5.9 44.3 + 6.1 (44, 58)
Combined retrospective and CVVHDF 13-20 mmol/h 830-1000 IU/hr Median 24.2 Median 42.5 (83)
prospective crossover adjusted to pH (IQR 17.4-42.3) (IQR 20.6-69.1)
Observational Cohort CVVHD 28 mmol/h ACT>200 s 26 + 1.6 18.3 + 11 (84)
Citrate-prospective cohort; CVVHD 40 mmol/h 500 IU/h 57.6 + 26.3 + 24 (85)

UFH-retrospective
Retrospective historical controls CVVH 18 mmol/h 500-775 IU/h 1 filter/day 1.1 filter/day (86)
predilution
Randomized controlled CVVHDF 25 mmol/h PTT between 124.5 38.3 (87)
45-65 secs (CI 95.3-157.4) (CI 24.8-61.9)
Randomized cross-over CVVH 39 mmol/h 500-1000 IU/h Median 70 Median 40 (88)

postdilution (IQR 44-140) (IQR 17-48)
Prospective observational CVVHDF 51.4 mmol/hr 25 IU/kg bolus; Median 40 Median 20 (89)
cohort citrate; CVVH 5 U/kg/hr to (IQR 14-72) (IQR 5-44)
heparin maintain PTT
45-55 secs
Several studies have compared citrate to other forms of anticoagulation in CRRT (Table
10.3). In general, citrate has superior filter longevity, increased efficacy and minimal to no
bleeding risk with some side effects that can be easily predicted and managed. This has
prompted a greater utilization of citrate for AC, particularly in high risk individuals.
Special Issues in Developing Countries

Limited literature exists on the use of CRRT in developing countries. In a retrospective study
from China, blood pumps were used to perform CRRT procedures from 1978 to 1996. These
patients were dialyzed using home grown lactate as dialysate. Specialized machines for CRRT
were used only after 1997.65 Some of the issues are:66
• Non-availability of CRRT machines. CRRT is done using an ordinary blood pump.
• Use of “home grown” PD fluid as dialysate.
• Use of ordinary F8 dialyzer as CRRT filter.
• Lack of facilities of continuous biochemical parameters and blood gases.
• Limited availability of tri-sodium citrate as an anticoagulant.
• Increased cost of special calcium free solutions.
• Balancing features of prolonged filter life and circuit efficiency versus cost of citrate,
monitoring and custom solutions.
• Need for training and expertise in recognizing complications of citrate therapy.
CONCLUSION
In conclusion, we feel that individualized prescription is needed for each patient.67 The
reduction of life-threatening complications with RCA makes it an attractive alternative. The
availability of simplified RCA protocols is a relatively new phenomenon and as the collective
experience of various centers with RCA increases, we may see a further improvement of filter
life. As with any anticoagulant procedure, appropriate monitoring is required. Barriers to
adoption of this technique in emerging countries remain availability and cost of citrate. We
anticipate that as these procedures are utilized further, costs will come down and the key
components will be more easily available.
REFERENCES
1. White RR, Sullenger BA, Rusconi CP. Developing aptamers into therapeutics. J Clin Invest 2000;106(8):929-
34.
2. Bouman CS, de Pont AC, Meijers JC, Bakhtiari K, Roem D, Zeerleder S, et al. The effects of continuous
venovenous hemofiltration on coagulation activation. Crit Care 2006;10(5):R150.
3. Oudemans-van Straaten HM, Wester JP, de Pont AC, Schetz MR. Anticoagulation strategies in continuous
renal replacement therapy: can the choice be evidence based? Intensive Care Med 2006;32(2):188-202.
4. Fealy N, Baldwin I, Bellomo R. The effect of circuit “down-time” on uraemic control during continuous
veno-venous haemofiltration. Crit Care Resusc 2002;4(4):266-70.
5. Uchino S, Fealy N, Baldwin I, Morimatsu H, Bellomo R. Continuous is not continuous: the incidence and
impact of circuit “down-time” on uraemic control during continuous veno-venous haemofiltration. Intensive
Care Med 2003;29(4):575-8.
6. Mehta RL, McDonald B, Gabbai FB, Pahl M, Pascual MT, Farkas A, et al. A randomized clinical trial of
continuous versus intermittent dialysis for acute renal failure. Kidney Int 2001;60(3):1154-63.
7. Manns B, Doig CJ, Lee H, Dean S, Tonelli M, Johnson D, et al. Cost of acute renal failure requiring
dialysis in the intensive care unit: clinical and resource implications of renal recovery. Crit Care Med
2003;31(2):449-55.
8. Levi M, Opal SM. Coagulation abnormalities in critically ill patients. Crit Care 2006;10(4):222.
9. Vervloet MG, Thijs LG, Hack CE. Derangements of coagulation and fibrinolysis in critically ill patients
with sepsis and septic shock. Semin Thromb Hemost 1998;24(1):33-44.
10. Gando S, Nanzaki S, Sasaki S, Aoi K, Kemmotsu O. Activation of the extrinsic coagulation pathway in
patients with severe sepsis and septic shock. Crit Care Med 1998;26(12):2005-9.
11. Webb AR, Mythen MG, Jacobson D, Mackie IJ. Maintaining blood flow in the extracorporeal circuit:
haemostasis and anticoagulation. Intensive Care Med 1995;21(1):84-93.
12. Stefanidis I, Hagel J, Maurin N. Influence of coagulation parameters on filter running time during continuous
venovenous hemofiltration. Contrib Nephrol 1995;116:145-9.
13. de Pont AC, Oudemans-van Straaten HM, Roozendaal KJ, Zandstra DF. Nadroparin versus dalteparin
anticoagulation in high-volume, continuous venovenous hemofiltration: a double-blind, randomized,
crossover study. Crit Care Med 2000;28(2):421-5.
14. Ward D. Anticoagulation in Patients on Hemodialysis. Fourth ed. Nissenson A, Fine, R, editor: McGraw-
Hill Professional; 2005.
15. Voss R, Matthias FR, Borkowski G, Reitz D. Activation and inhibition of fibrinolysis in septic patients in
an internal intensive care unit. Br J Haematol 1990;75(1):99-105.
16. Schmitt GW, Moake JL, Rudy CK, Vicks SL, Hamburger RJ. Alterations in hemostatic parameters during
hemodialysis with dialyzers of different membrane composition and flow design. Platelet activation and
factor VIII-related von Willebrand factor during hemodialysis. Am J Med 1987;83(3):411-8.
17. Opatrny K, Jr., Zemanova P, Opatrna S, Vit L. Fibrinolysis in chronic renal failure, dialysis and renal
transplantation. Ann Transplant 2002;7(1):34-43.
18. Cardigan RA, McGloin H, Mackie IJ, Machin SJ, Singer M. Activation of the tissue factor pathway occurs
during continuous venovenous hemofiltration. Kidney Int 1999 55(4):1568-74.
19. Bellomo R, Teede H, Boyce N. Anticoagulant regimens in acute continuous hemodiafiltration: a comparative
study. Intensive Care Med 1993;19(6):329-32.
20. Morabito S, Guzzo I, Solazzo A, Muzi L, Luciani R, Pierucci A. Continuous renal replacement therapies:
anticoagulation in the critically ill at high risk of bleeding. J Nephrol 2003;16(4):566-71.
21. Tan HK, Baldwin I, Bellomo R. Continuous veno-venous hemofiltration without anticoagulation in high-
risk patients. Intensive Care Med 2000;26(11):1652-7.
22. Brophy PD, Somers MJ, Baum MA, Symons JM, McAfee N, Fortenberry JD, et al. Multi-centre evaluation
of anticoagulation in patients receiving continuous renal replacement therapy (CRRT). Nephrol Dial
Transplant 2005;20(7):1416-21
23. Uchino S, Fealy N, Baldwin I, Morimatsu H, Bellomo R. Continuous venovenous hemofiltration without
anticoagulation. ASAIO J 2004;50(1):76-80.
24. Kellum JA, Mehta RL, Angus DC, Palevsky P, Ronco C. The first international consensus conference on
continuous renal replacement therapy. Kidney Int 2002;62(5):1853-63.
25. Alonso A, Lau J, Jaber BL. Biocompatible hemodialysis membranes for acute renal failure. Cochrane
Database Syst Rev 2008(1):CD005283.
26. van de Wetering J, Westendorp RG, van der Hoeven JG, Stolk B, Feuth JD, Chang PC. Heparin use in
continuous renal replacement procedures: the struggle between filter coagulation and patient hemorrhage.
J Am Soc Nephrol 1996;7(1):145-50.
27. Abramson S, Niles JL. Anticoagulation in continuous renal replacement therapy. Curr Opin Nephrol
Hypertens 1999;8(6):701-7.
28. Seitz R, Wolf M, Egbring R, Havemann K. The disturbance of hemostasis in septic shock: role of neutrophil
elastase and thrombin, effects of antithrombin III and plasma substitution. Eur J Haematol 1989;43(1):22-
8.
29. Davenport A, Bouman C, Kirpalani A, Skippen P, Tolwani A, Mehta RL, et al. Delivery of renal replacement
therapy in acute kidney injury: what are the key issues? Clin J Am Soc Nephrol 2008;3(3):869-75.
30. Davenport A. The coagulation system in the critically ill patient with acute renal failure and the effect of
an extracorporeal circuit. Am J Kidney Dis 1997;30(5 Suppl 4):S20-7.
31. Cointault O, Kamar N, Bories P, Lavayssiere L, Angles O, Rostaing L, et al. Regional citrate anticoagulation
in continuous venovenous haemodiafiltration using commercial solutions. Nephrol Dial Transplant
2004;19(1):171-8.
32. Palsson R, Niles JL. Regional citrate anticoagulation in continuous venovenous hemofiltration in critically
ill patients with a high risk of bleeding. Kidney Int 1999;55(5):1991-7.
33. Chong BH. Heparin-induced thrombocytopenia. J Thromb Haemost 2003;1(7):1471-8.
34. Kelton JG. The pathophysiology of heparin-induced thrombocytopenia: biological basis for treatment.
Chest 2005;127(2 Suppl):9S-20S.
35. Yamamoto S, Koide M, Matsuo M, Suzuki S, Ohtaka M, Saika S, et al. Heparin-induced thrombocytopenia
in hemodialysis patients. Am J Kidney Dis 1996;28(1):82-5.
36. Davenport A. Heparin-induced thrombocytopenia during renal replacement therapy. Hemodialysis
International 2004;8(3):295-303.
37. Hirsh J, Anand SS, Halperin JL, Fuster V. Guide to anticoagulant therapy: Heparin : a statement for
healthcare professionals from the American Heart Association. Circulation 2001;103(24):2994-3018.
38. Levine MN, Hirsh J, Gent M, Turpie AG, Cruickshank M, Weitz J, et al. A randomized trial comparing
activated thromboplastin time with heparin assay in patients with acute venous thromboembolism requiring
large daily doses of heparin. Arch Intern Med 1994;154(1):49-56.
39. Francis JL, Groce JB, 3rd. Challenges in variation and responsiveness of unfractionated heparin.
Pharmacotherapy 2004;24(8 Pt 2):108S-19S.
40. Hirsh J, Warkentin TE, Shaughnessy SG, Anand SS, Halperin JL, Raschke R, et al. Heparin and low-
molecular-weight heparin: mechanisms of action, pharmacokinetics, dosing, monitoring, efficacy, and safety.
Chest 2001;119(1 Suppl):64S-94S.
41. Greaves M. Limitations of the laboratory monitoring of heparin therapy. Scientific and Standardization
Committee Communications: on behalf of the Control of Anticoagulation Subcommittee of the Scientific
and Standardization Committee of the International Society of Thrombosis and Haemostasis. Thromb
Haemost 2002;87(1):163-4.
42. Baker BA, Adelman MD, Smith PA, Osborn JC. Inability of the activated partial thromboplastin time to
predict heparin levels. Time to reassess guidelines for heparin assays. Arch Intern Med 1997;157(21):2475-
9.
43. Morita Y, Johnson RW, Dorn RE, Hall DS. Regional anticoagulation during hemodialysis using citrate. Am
J Med Sci 1961;242:32-43.
44. Mehta RL, McDonald BR, Aguilar MM, Ward DM. Regional citrate anticoagulation for continuous
arteriovenous hemodialysis in critically ill patients. Kidney Int 1990;38(5):976-81.
45. Mehta RL, McDonald BR, Ward DM. Regional citrate anticoagulation for continuous arteriovenous
hemodialysis. An update after 12 months. Contrib Nephrol 1991;93:210-4.
46. Tolwani AJ, Prendergast MB, Speer RR, Stofan BS, Wille KM. A practical citrate anticoagulation continuous
venovenous hemodiafiltration protocol for metabolic control and high solute clearance. Clin J Am Soc
Nephrol 2006;1(1):79-87.
47. Hofmann RM, Maloney C, Ward DM, Becker BN. A novel method for regional citrate anticoagulation in
continuous venovenous hemofiltration (CVVHF). Ren Fail 2002;24(3):325-35.
48. Morgera S, Scholle C, Voss G, Haase M, Vargas-Hein O, Krausch D, et al. Metabolic complications during
regional citrate anticoagulation in continuous venovenous hemodialysis: single-center experience. Nephron
Clin Pract 2004;97(4):c131-6.
49. Davenport A. Anticoagulation for continuous renal replacement therapy. Contrib Nephrol 2004;144:228-
38.
50. Kramer L, Bauer E, Joukhadar C, Strobl W, Gendo A, Madl C, et al. Citrate pharmacokinetics and
metabolism in cirrhotic and noncirrhotic critically ill patients. Crit Care Med 2003;31(10): 2450-5.
51. Betjes MG, van Oosterom D, van Agteren M, van de Wetering J. Regional citrate versus heparin
anticoagulation during venovenous hemofiltration in patients at low risk for bleeding: similar hemofilter
survival but significantly less bleeding. J Nephrol 2007;20(5):602-8.
52. Cubattoli L, Teruzzi M, Cormio M, Lampati L, Pesenti A. Citrate anticoagulation during CVVH in high risk
bleeding patients. Int J Artif Organs 2007;30(3):244-52.
53. Tobe SW, Aujla P, Walele AA, Oliver MJ, Naimark DM, Perkins NJ, et al. A novel regional citrate
anticoagulation protocol for CRRT using only commercially available solutions. J Crit Care 2003;18(2):121-
9.
54. Bohler J, Schollmeyer P, Dressel B, Dobos G, Horl WH. Reduction of granulocyte activation during
hemodialysis with regional citrate anticoagulation: dissociation of complement activation and neutropenia
from neutrophil degranulation. J Am Soc Nephrol 1996;7(2):234-41.
55. Bos JC, Grooteman MP, van Houte AJ, Schoorl M, van Limbeek J, Nube MJ. Low polymorphonuclear
cell degranulation during citrate anticoagulation: a comparison between citrate and heparin dialysis. Nephrol
Dial Transplant 1997;12(7):1387-93.
56. Gabutti L, Ferrari N, Mombelli G, Keller F, Marone C. The favorable effect of regional citrate anticoagulation
on interleukin-1beta release is dissociated from both coagulation and complement activation. J
Nephrol2004;17(6):819-25.
57. Gritters M, Grooteman MP, Schoorl M, Bartels PC, Scheffer PG, Teerlink T, et al. Citrate anticoagulation
abolishes degranulation of polymorphonuclear cells and platelets and reduces oxidative stress during
haemodialysis. Nephrol Dial Transplant 2006;21(1):153-9.
58. Ward DM, Mehta RL. Extracorporeal management of acute renal failure patients at high risk of bleeding.
Kidney Int 1993;41(Suppl):S237-44.
59. Sundaram S, Barrett TW, Meyer KB, Perrella C, Neto MC, King AJ, et al. Transmembrane passage of
cytokine-inducing bacterial products across new and reprocessed polysulfone dialyzers. J Am Soc Nephrol
1996;7(10):2183-91.
60. Johnston RV, Boiteau P, Charlebois K, Long S, U D. Responding to tragic error: lessons from Foothills
Medical Centre. CMA J2004;170(11):1659-60.
61. Romagnoni M, Beccari M, Sorgato G. Life-threatening hyperkalaemia during a haemodialysis session: an
avoidable risk. Nephrol Dial Transplant 1998;13(10):2480-1.
62. McLean AG, Davenport A, Cox D, Sweny P. Effects of lactate-buffered and lactate-free dialysate in CAVHD
patients with and without liver dysfunction. Kidney Int 2000;58(4):1765-72.
63. Barenbrock M, Hausberg M, Matzkies F, de la Motte S, Schaefer RM. Effects of bicarbonate- and lactate-
buffered replacement fluids on cardiovascular outcome in CVVH patients. Kidney Int 2000;58(4):1751-
7.
64. Brahmbhatt Y, Kuo S, Chaparala H, Suh H, Wadhwa NK. Cost-effectiveness of commercially available
solutions in continuous veno-venous hemo-diafiltration. In: Mehta R, editor. The 12th International
Conference on Continuous Renal Replacement Therapies;March 7-10, 2007;San Diego, California:
Karger;2007. p. 183-208.
65. Ji D, Gong D, Xie H, Xu B, Liu Y, Li L. A retrospective study of continuous renal replacement therapy
versus intermittent hemodialysis in severe acute renal failure. Chin Med J (Engl) 2001;114(11):1157-61.
66. Avadaiammal V, Leelakumari M, George J, Upendran B, Varma S, Mohandas M. CRRT in Developing
Countries. In: Mehta R, editor. The 11th International Conference on Continuous Renal Replacement
Therapies;March 2-4, 2006;San Diego, California: Blood Purification; 2006. p. 247-73.
67. Aucella F, Di Paolo S, Gesualdo L. Dialysate and replacement fluid composition for CRRT. Contrib Nephrol
2007;156:287-96.
68. Cheung AK, Chenoweth DE, Otsuka D, Henderson LW. Compartmental distribution of complement
activation products in artificial kidneys. Kidney Int 1986;30(1):74-80.
69. Lavaud S, Canivet E, Wuillai A, Maheut H, Randoux C, Bonnet JM, et al. Optimal anticoagulation strategy
in haemodialysis with heparin-coated polyacr ylonitrile membrane. Nephrol Dial Transplant
2003;18(10):2097-104.
70. Baldwin I, Bellomo R, Koch B. Blood flow reductions during continuous renal replacement therapy and
circuit life. Intensive Care Med 2004;30(11):2074-9.
71. Pawlak K, Borawski J, Naumnik B, Mysliwiec M. Relationship between oxidative stress and extrinsic
coagulation pathway in haemodialyzed patients. Thromb Res 2003;109(5-6):247-51.
72. Frank RD, Muller U, Lanzmich R, Groeger C, Floege J. Anticoagulant-free Genius haemodialysis using
low molecular weight heparin-coated circuits. Nephrol Dial Transplant 2006;21(4):1013-8.
73. Davies H, Leslie G. Maintaining the CRRT circuit: non-anticoagulant alternatives. Aust Crit Care
2006;19(4):133-8.
74. du Cheyron D, Bouchet B, Bruel C, Daubin C, Ramakers M, Charbonneau P. Antithrombin supplementation
for anticoagulation during continuous hemofiltration in critically ill patients with septic shock: a case-
control study. Crit Care 2006;10(2):R45.
75. Baldwin I, Tan HK, Bridge N, Bellomo R. Possible strategies to prolong circuit life during hemofiltration:
three controlled studies. Ren Fail 2002;24(6):839-48.
76. de Pont AC, Bouman CS, Bakhtiari K, Schaap MC, Nieuwland R, Sturk A, et al. Predilution versus
postdilution during continuous venovenous hemofiltration: a comparison of circuit thrombogenesis. ASAIO
J 2006;52(4):416-22.
77. van der Voort PH, Gerritsen RT, Kuiper MA, Egbers PH, Kingma WP, Boerma EC. Filter run time in
CVVH: pre- versus post-dilution and nadroparin versus regional heparin-protamine anticoagulation. Blood
Purif 2005;23(3):175-80.
78. Shulman RI, Singer M, Rock J. Continuous renal replacement therapy. Keeping the circuit open: lessons
from the lab. Blood Purif 2002;20(3):275-81.
79. Martin RK. Who should manage continuous renal replacement in the intensive care setting? A nursing
viewpoint. EDTNA ERCA J 2002;Suppl 2:43-5, 53.
80. Canaud B, Desmeules S, Klouche K, Leray-Moragues H, Beraud JJ. Vascular access for dialysis in the
intensive care unit. Best Pract Res Clin Anaesthesiol 2004;18(1):159-74.
81. Baldwin I. Is there a need for a nurse emergency team for continuous renal replacement therapy? Contrib
Nephrol 2007;156:191-6.
82. Cutts MW, Thomas AN, Kishen R. Transfusion requirements during continuous veno-venous haemofiltration:
-the importance of filter life. Intensive Care Med 2000;26(11):1694-7.
83. Gabutti L, Marone C, Colucci G, Duchini F, Schonholzer C. Citrate anticoagulation in continuous
venovenous hemodiafiltration: a metabolic challenge. Intensive Care Med 2002;28(10):1419-25.
84. Mitchell A, Daul AE, Beiderlinden M, Schafers RF, Heemann U, Kribben A, et al. A new system for regional
citrate anticoagulation in continuous venovenous hemodialysis (CVVHD). Clin Nephrol 2003;59(2):106-
14.
85. Swartz R, Pasko D, O’Toole J, Starmann B. Improving the delivery of continuous renal replacement therapy
using regional citrate anticoagulation. Clin Nephrol 2004;61(2):134-43
86. Thoenen M, Schmid ER, Binswanger U, Schuepbach R, Aerne D, Schmidlin D. Regional citrate
anticoagulation using a citrate-based substitution solution for continuous venovenous hemofiltration in
cardiac surgery patients. Wien Klin Wochenschr 2002;114(3):108-14.
87. Kutsogiannis DJ, Gibney RT, Stollery D, Gao J. Regional citrate versus systemic heparin anticoagulation
for continuous renal replacement in critically ill patients. Kidney Int 2005;67(6):2361-7.
88. Monchi M, Berghmans D, Ledoux D, Canivet JL, Dubois B, Damas P. Citrate vs. heparin for anticoagulation
in continuous venovenous hemofiltration: a prospective randomized study. Intensive Care Med
2004;30(2):260-5.
89. Bagshaw SM, Laupland KB, Boiteau PJ, Godinez-Luna T. Is regional citrate superior to systemic heparin
anticoagulation for continuous renal replacement therapy? A prospective observational study in an adult
regional critical care system. J Crit Care 2005;20(2):155-61.
Eleven
Can the Choice of Resuscitation
Fluid Adversely Affect Renal
Function in the Critically Ill?
Roop Kishen, Sara Blakeley
INTRODUCTION
It is a matter of common observation that renal dysfunction accompanies critical illness;
commonest causes being sepsis, trauma, poor organ perfusion and nephrotoxins. Applying
RIFLE criteria for acute kidney injury (AKI),1 it is suggested that AKI occurs in as many as two
thirds of the critically ill2 and renal failure requiring renal replacement therapy (RRT) in as
many as 6%.3 As mortality of this condition is very high,2 it is logical to prevent AKI rather
than to let renal failure supervene.
It is now well recognized that fluid therapy is the only definitive intervention that prevents
progression of AKI in the critically ill. Time honored management of emerging AKI and
prevention of acute renal failure (ARF) involves fluid resuscitation, optimizing haemodynamic
parameters, treating any underlying cause (e.g., treating sepsis, stabilizing and treating fractures
etc.) and avoiding nephrotoxin administration. Although the concept of ‘pre-renal’ renal failure
is being challenged, especially in sepsis associated AKI (SAAKI),4 perturbations of circulating
volume and fluid shifts due to capillary leak make fluid therapy a logical first step in preventing
AKI. It has also been suggested that renal perfusion in sepsis is cardiac output dependent,5
again making fluid resuscitation a logical initial step. Furthermore, in a single centre randomized
controlled trial (RCT) of early goal directed therapy in sepsis and septic shock, it has been
shown that early aggressive fluid resuscitation reduces organ failure (and by implication AKI,
although data on renal function are not available) and improves outcome.6 Concomitant
hypovolaemia and dehydration increase the risk of AKI and renal failure in the ‘at risk’
population (those with diabetes, hypertension, pre-existing renal dysfunction and the elderly).7,8
Fluid resuscitation corrects hypovolaemia, improves preload, cardiac output and blood pressure,
renal perfusion and urine output, and in many situations may improve internal renal
haemodynamics and glomerular filtration rate (GFR). Thus, most clinicians agree (and studies
show) that appropriate fluid challenges/resuscitation is the first step in prevention and treatment
of emerging AKI and early (non-oliguric) stage of acute renal failure.8,9 However, while fluid
administration reduces the incidence of AKI, there are concerns that some components of
Can the Choice of Resuscitation Fluid Adversely Affect Renal Function? 113
commonly administered fluids and some types of fluids may cause or worsen AKI in the
critically ill.
AVAILABLE FLUIDS IN CLINICAL USE

Just after O’Shaughnessy’s description of the biochemical changes in severe cholera,10
electrolyte solutions injected into the veins were used to ‘good effect’ in these patients for the
first time in 1832 by an Edinburgh doctor, Thomas Latta.11 Over the years as ‘fluid injections’
became an accepted therapy, many modifications of the electrolyte composition of these
fluids were suggested. However, it was not till a century later that Alexis Hartmann12 suggested
modifications to the Ringer’s solution for a better therapeutic profile in patients given such
fluids. Today, there are a variety of fluids available for intravenous infusion.
Fluids in routine use are broadly classified into crystalloids and colloids. Crystalloids are
electrolyte solutions, and the two commonly used in clinical practice are 0.9% saline (NS)
and ringer’s lactate (RL), a balanced salt solution, known as Hartmann’s solution in the UK,
Europe and Australasia. Crystalloids are routinely used in resuscitation in almost all clinical
scenarios, to replace lost circulating volume in trauma, burns and other hypovolaemic situations.
Colloids are thought to be better plasma volume expanders, stay in circulation longer than
crystalloids, are generally better at supporting cardiac output, have beneficial effects on oxygen
delivery and microcirculation and may have other important effects on inflammatory
mediators.13 Common colloids in use are dextrans, gelatins and starches. Until now most
colloids have been suspended in NS but colloid suspensions in balanced salt solutions are
becoming available now. The crystalloid/colloid controversy has been going on for the better
part of the last half century with a shift now to a colloid/colloid debate;13 however, these
debates are outside the scope of this chapter.
CRYSTALLOIDS AND RENAL FUNCTION
Effect of Chemical Composition

The commonest crystalloid solution used in resuscitation, fluid loading and general fluid
replacement is NS.14 NS is a hyperchloraemic solution compared to plasma (154 mmol.l-1 of
Cl- in NS compared to 100 mmol.l-1 in plasma). The next commonly used solution is RL,
which contains less chloride compared to NS (111 mmol.l-1). Infusions of large volumes of
NS can cause significant alterations in acid-base balance and electrolytes in the recipients
whereas this is not seen after RL infusions. It is suggested that such alterations in acid-base
after NS infusion are of little significance in the healthy adults, however, in the critically ill and
those with multiple trauma these effects may lead to serious abnormalities which may be life
threatening in some15 and may lead to inappropriate clinical decisions.16,17
Serious acidosis due to hyperchloraemia, usually called hyperchloraemic metabolic acidosis
(HCMA), has been demonstrated in many studies in patients receiving large volumes of NS
against those receiving RL. In a randomized controlled trial of NS vs. RL in patients undergoing
elective gynaecological surgery, a significant decrease in pH (7.41 to 7.28), serum bicarbonate
concentration (23.5 to 18.4 mmol.l-1) and increase in serum chloride concentration (104 to
115 mmol.l-1) was observed in patients receiving NS compared to those receiving RL.18
Unfortunately, effects on renal function were not recorded in this study; although the fluid
infusions were comparable in both groups (about 6000ml), there was a tendency to better
urine output (though not significant) in patients receiving RL (1100 ml) than those receiving
NS (700 ml). Another study comparing NS and RL in patients undergoing abdominal aortic
aneurysm repair demonstrated HCMA more often associated with NS than with RL.19 However,
in this study no differences in renal function were observed in the two groups. In a double-
blind randomized study of intraoperative fluid administration,20 patients undergoing renal
transplantation received either NS (n = 26) or RL (n = 25). Unexpectedly, hyperkalaemia
was observed in 5 (19%) patients in NS group and none in RL group; eight patients (31%)
in NS group developed metabolic acidosis serious enough to warrant treatment. The study
was terminated early because of these concerns, although there were no deleterious long-
term effects on renal function.
As stated earlier, HCMA in healthy individuals as a consequence of NS as against RL
infusion is thought to be inconsequential; however where this has been studied, several
interesting observations have been made. In a randomized study of NS infusion versus RL,
healthy volunteers (n = 18) received 50 ml.kg-1 of either solution over 1 hour.21 HCMA
developed in the NS group with increased time to urination. In another double blind crossover
study in healthy volunteers (n = 9), 2000 ml of either NS or RL was infused over 1 hour on
separate occasions in random order22. The authors observed that subjects receiving RL retained
less volume at 6 hours (30% vs. 56%; p = 0.049), voided more urine (median urine output
1000 ml vs. 450 ml; p = 0.049) with less time to first micturition (median 70 vs. 185 min;
p =0.008).22
Beneficial effects of sodium bicarbonate as against NS in preventing contrast induced
nephropathy (CIN) have been shown in various studies.23,24 It is however unclear if the
protective effect of bicarbonate based volume expansion is due to prevention of HCMA at
the same time as contrast administration15 or reduced chloride in infused fluid. A recent single-
blind, randomized trial of sodium bicarbonate versus NS did not find that the former solution
was superior to the latter in preventing CIN.25 The study however was a single centre study
with 40% diabetic patients in each study arm. Thus pre-procedure hydration with crystalloids
still remains the best option for preventing CIN; the nature of such fluid does not seem to
matter. Aggressive hydration in rhabdomyolysis has been shown to prevent development of
AKI;26,27 although mechanisms of kidney injury in this condition are still a matter of debate.
Traditionally, sodium bicarbonate is included in such fluid resuscitation; however, there is
currently little evidence that bicarbonate therapy along with crystalloids is better than crystalloid
infusion alone.28
In animal experiments, it has been shown that hyperchloraemia causes a progressive renal
vasoconstriction and a fall in GFR that is independent of the renal innervation.29 This
vasoconstrictive effect is potentiated by prior salt depletion and is specific to the kidney.
Metabolic acidosis irrespective of cause potentiates these effects.30
The Effect of Tonicity

Hypotonic solutions are not in routine use for resuscitation from hypovolaemia and shock.
Over the past few years there has been a keen interest in small volume resuscitation with
hypertonic crystalloid solutions especially hypertonic saline (HTS). Besides requiring smaller
volumes for resuscitation, HTS solutions are thought to have additional beneficial effects
especially in sepsis.31 HTS causes intravascular volume expansion by drawing fluid from tissues,
increases cardiac output and oxygen delivery mediated by improved cardiac contractility,
thus resulting in better haemodynamics in haemorrhagic shock, sepsis or hypovolaemic shock.31
Reduction of endothelial cell oedema decreased vasoactive mediator generation in sepsis
and beneficial immunomodulation (e.g., decrease in proinflammotory and increase in anti-
inflammatory cytokines) are other benefits of HTS resuscitation.31,32 HTS solution are also
thought to have a therapeutic advantage where they are felt to be useful in optimizing cerebral
perfusion pressure in hypotensive head injured patient with raised intra-cranial pressure. 33
Despite these advantages, no durable benefits in outcome have been identified by plasma
volume expansion with HTS in a large multi-centre trial of pre-hospital fluid resuscitation in
head injuries.34 A Cochrane review in 2000 and updated in 2007 found insufficient data to
be able to decide if HTS in trauma patients, burns and postoperative patients is better than
isotonic or near isotonic crystalloids.35 However, the confidence intervals were very wide and
a benefit could not be excluded. HTS is often combined with dextrans and hydroxyethyl
starches (HES) to enhance their oncotic and plasma expander effect. Almost all studies in this
regard do not report renal function changes. One such study comparing 250 ml of 7.5% HTS
with 6% dextran-70 against hydroxyethyl starch (HES) in NS in patients undergoing
abdominal aortic aneurysm repair found no difference in acid-base status or renal function
up to 24 hours postoperatively.36 However, another before/after study in burns patients found
a 4-fold increase in renal failure and doubling of mortality with HTS as compared to RL.37
Despite this study,37 there does not seem to be enough evidence that HTS causes significant
renal dysfunction in the critically ill. This is probably because only small amounts of the
hypertonic fluids are used in resuscitation.
Concerns have been expressed in the literature about other observed adverse effects of
HTS, and hyperosmolar/hypertonic sodium lactate solutions (HLS) have been developed
and studied as an alternative to HTS. In a recent study involving cardiac surgery patients
receiving HLS or RL for postoperative fluid resuscitation, HLS resulted in a better cardiac
index but a smaller positive fluid balance as less fluid was infused in this group. However,
there was a non-significant reduction in urine output in the HLS group despite a higher cardiac
index (and similar heart rate, mean arterial, central venous and pulmonary artery occlusion
pressures).38
Mannitol, for a long time the ‘sheet anchor’ of AKI preventative therapy is now known to
cause or increase AKI in various situations. Mannitol causes an osmotic diuresis and may
cause rapid increase in plasma volume leading to pulmonary oedema and hyperoncotic kidney
injury in those without a plasma volume deficit.39,40 Moreover excessive osmotic diuresis can
lead to kidney injury in those with significant volume deficit, especially in patients at risk.
Mannitol does not offer any additional benefit over aggressive fluid therapy in rhabdomyolysis28
nor does it ameliorate or prevent CIN as compared to fluid therapy alone. On the contrary,
it has been found to aggravate nephrotoxicity induced by contrast media,41 especially in
patients with diabetes and/or chronic renal dysfunction.42
COLLOIDS AND RENAL FUNCTION
Albumin
Human albumin has widely been used as a resuscitation fluid in the critically ill.
Hypoalbuminaemia is associated with poor clinical outcome and most critically ill patients
become hypoalbuminaemic at some stage during their illness. Many of these patients also
exhibit widespread tissue oedema giving rise to the belief that increasing albumin level will
raise oncotic pressure, ‘draw fluid out of the tissue’ and improve clinical outcome. A Cochrane
review (including 30 randomized trials; n = 1419) on this subject found increased mortality
with albumin administration.43 Subsequently, the largest RCT comparing 4% albumin versus
saline in the critically ill (SAFE study; n = 6997) did not find any difference in renal outcomes,
need for or duration of RRT, ICU length of stay or 28 day mortality.44 It must be emphasized
that the SAFE study used albumin in saline and compared it to NS in controls. Whether
specific subgroups of critically ill patients will benefit from albumin administration is not known
from this study. This issue has been addressed in a small (n = 126) but well designed RCT
of cefotaxime versus cefotaxime and albumin in cirrhotic patients with spontaneous bacterial
peritonitis.45 Contrary to the findings of the Cochrane Review, albumin was found to
significantly decrease renal failure (OR, 4.6; 95% CI 1.3 - 16.1; p = 0.02) and mortality (OR
4.5; 95% CI 1.9 - 20.9; p = 0.05), this benefit was sustained at 3 months. However, the
available evidence does not support albumin therapy for volume expansion or prevention of
AKI in an unselected population of the critically ill.
Synthetic Colloids
These are classified as being hypooncotic, such as gelatins (e.g. gelofusine) or hyperoncotic,
such as starches (e.g. HES) and dextrans. Like albumin, they stay in the circulation longer
than crystalloids, and therefore, improve cardiac output, blood pressure and microvascular
flow and may well have other beneficial effects on the immune system. They are also cheaper
than albumin. Although advocated for plasma expansion, there have been concerns about
their safety and undesirable side effects; AKI being one of them.46 Colloids are also associated
with increased anaphylactic reactions, pruritus, coagulopathy and renal failure.47 Although
there is no paucity of studies, the literature is controversial at best.
Studies have shown that infusion with either high-molecular weight HES solutions or HES
with a high molar substitution can cause kidney injury in patients with sepsis. In one study,48
HES 6% (molecular weight/substitution – 200/0.60-0.66) caused a higher rate of acute renal
failure, oliguria and higher creatinine levels in patients with severe sepsis or septic shock
when compared to 3% gelatin. This kidney injury manifests as reversible tubular cell oedema,
tubular obstruction and medullary ischaemia.49 HES is associated with AKI in post cadaveric
renal transplantation.50 Additionally, HES has been shown to accumulate in renal tubular
cells and this thought to contribute to onset of AKI.51 Although HES resuscitation of trauma
victims has been shown to improve blood pressure with less infused volume than crystalloids,
this may well be the drawback of colloid resuscitation as less free water is infused in these
situations. This may lead to dehydration, oligurea, increased creatinine levels and AKI with
colloids.
In a large RCT of 537 septic patients, 10% pentastarch (HES 200/0.5) was compared to
modified RL for fluid resuscitation.52 This study also compared conventional to intensive insulin
therapy. The study was stopped after the first safety analysis of 488 patients because of higher
incidence of hypoglycaemia in the intensive insulin therapy group (12.1% vs. 2.1%; p =
<0.001). Planned interim analysis after enrolment of 600 patients who continued to receive
the study fluids, showed a significantly higher incidence of renal failure and significantly higher
90-day mortality in patients who received HES than those who were infused with RL. Although
some patients received higher than recommended fluid doses (20 ml.kg-1 day-1), the study
showed an increased need for RRT even in patients where HES had been administered at the
recommended daily dose. In a prospective evaluation of a cohort of 822 patients receiving
fluids, increased incidence of AKI was found in groups receiving HES (130/0.4 and older
starches) and hyperoncotic albumin than in those receiving crystalloids, gelatins or 4%
albumin.53 Although the mechanisms of AKI with HES are far from clear, there is a strong
suggestion in the literature that HES causes direct kidney injury, at least in septic patients.54
On the other hand, colloid fluid therapy has been shown to be safe in a variety of post-
operative situations. Kumle at al55 studied the effect of low molecular weight HES, medium
molecular weight HES and gelatin in patients undergoing major abdominal surgery (n = 60).
They found no impairment of renal function as measured by α1-microglobulin, N-acetyl-β-
glucosaminidase, fractional sodium clearance and creatinine clearance. In another randomised
controlled trial of HES versus gelatin in aortic aneurysm surgery, Mahmood et al56 found that
renal function was better with two types of 6% HES (200/0.62 kDA and 130/0.4 kDa) than
with 4% gelatin, although gelatin group had a higher urine output. Boldt et al57 studied
60 elderly patients (age >80 years) undergoing cardiac surgery and assigned them to receive
6% HES (130/0.4) or 4% gelatin for volume replacement intraoperatively and up to the second
postoperative day. Serum creatinine levels were significantly lower in the HES group on first
(126 μmol.l-1 vs. 151 μmol.l-1; p =0.004) and second (133 μmol.l-1 vs. 161 μmol.l-1; p =
0.004) postoperative days. Creatinine clearance was significantly lower in the gelatin group.
These results have been confirmed in another study in patients undergoing abdominal aortic
surgery but with a preoperative creatinine clearance of <80 ml.min58. In this study HES
(130/0.4) was compared with 3% gelatin in a random fashion. No differences were seen in
the mean serum creatinine level or urine output in either group.
The differential results in these trials on HES seem to suggest that although the starches
may be safe in many situations, they may exert deleterious effect in septic patients.
CONCLUSION
There is evidence to show that the choice of resuscitation fluid can adversely, as well as
positively affect renal function in the critically ill. It should be remembered that fluids are still
the mainstay of treatment in the prevention of AKI in the critically ill, but in certain clinical
situations, certain fluids may be more advantageous compared to others. Renal effects may
be related to chemical composition of the fluid or its tonicity, but often the mechanisms are
not fully understood. These effects may not be seen when used in patients outside the intensive
care unit, but when dealing with critically ill patients the following practical points (Table
11.1) should be taken into consideration when choosing a fluid.
Table 11.1: Practical points
• Timely and adequate fluid therapy is vital in the prevention of AKI. For most situations, crystalloids or
colloids are ADEQUATE as resuscitation fluids.
• Ongoing and excessive NS infusions or saline-based colloid solutions cause HCMA, which may be deleterious
in the critically ill. Although there is no convincing evidence that excess Na or Cl promote kidney injury in
humans, experimental data from animal studies suggests CAUTION.
• Hyperoncotic crystalloids have NOT been shown to be superior to iso-oncotic or near iso-oncotic fluids
although they may have immunomodulating benefits.
• Sufficient CARE and diligence should be exercised while using large quantities of colloids for resuscitating
septic patients.
• HES seems to be detrimental in septic but not in postoperative patients. Colloids in balanced salt solutions
may have less deleterious side effects,59 but their safety in sepsis has yet to be studied.
• Albumin does not cause or prevent AKI, although it may have a place in SELECTED ICU patients.
REFERENCES
1. Bellomo R, Ronco C, Kellum JA, et al. Acute renal failure – definition, outcome measures, animal models,
fluid therapy and information technology needs: the second international consensus conference of the
Acute Dialysis Quality Initiative (ADQI) group. Crit Care 2004;8:R204-12.
2. Hoste EA, Clermont G, Kersten A, et al. RIFLE criteria for acute kidney injury are associated with hospital
mortality in critically ill patients: a cohort analysis. Crit Care 2006;10:R73.
3. Uchino S, Kellum JA, Bellomo R, et al. Beginning and Ending Supportive Therapy for the Kidney (BEST
Kidney) Investigators. Acute renal failure in the critically ill patients: a multinational, multicentre study.
JAMA 2005;294:813-18.
4. Bellomo R, Bagshaw S, Langenberg C, et al. Pre-renal azotemia: a flawed paradigm in critically ill septic
patient. Contrib Neprol 2007;156:1-9.
5. Langenberg C, Bellomo R, May C, et al. Renal blood flow in sepsis. Crit Care 2005;9:R363-74.
6. Rivers E, Nguyen B, Havastad S, et al. Early goal-directed therapy in the treatment of severe sepsis and
septic shock. New Eng J Med 2001;345:1368-77.
7. Leblanc M, Kellum JA, Gilbney M, et al. Risk factors for acute renal failure: inherent and modifiable risks.
Curr Opin Crit Care 2005;11:533-6.
8. Venkataraman R. Can we prevent acute kidney injury? Crit Care Med 2008;36(Suppl):S166-71.
9. Venkataraman R, Kellum JA. Prevention of acute renal failure. Chest 2007;131:300-08.
10. O’Shaughnessy WB. Chemical pathology of cholera: remarks on Dr Thomson’s analysis of the blood of
individuals affected with malignant cholera. Lancet 1832-ii;18(456): 225-56.
11. Latta T. Malignant cholera. Documents communicated by the Central Board of Health, London, relative
to the treatment of cholera by copious injection of aqueous and saline fluids into veins. Lancet 1832-
ii;18(457):274-7.
12. Hartmann AF. Theory and practice of parenteral fluid administration. JAMA 1934;103:1349-54.
13. Boldt J. Do plasma substitutes have additional properties beyond correcting volume deficit? Shock
2006;25:103-16.
14. Lobo DN, Dube MG, Neal KR, et al. Problems with solutions: drowning in the brine of an inadequate
knowledge base. Clin Nutr 2001;20:125-30.
15. Kellum JA, Creda J, Kaplan LJ, et al. Fluids for prevention and management of acute kidney injury. Int
J Artif Organs 2008;31:96-110.
16. Kishen R. Personal communication.
17. Boldt J. Saline versus balanced hydroxyethyl starch: Does it matter? Curr Opin Anaesthesiol 2008;21:679-
83.
18. Scheingraber S, Rehm M, Sehmisch C, et al. Rapid saline infusion produces hyperchloraemic acidosis in
patients undergoing gynaecologic surgery. Anaesthesiology 1999;90:1265-70.
19. Walters JH, Gottlieb A, Schoenwald P, et al. Normal saline versus lactated Ringer’s solution for intraoperative
fluid management in patients undergoing abdominal aortic aneurysm repair: an outcome study. Anesth
Analg 2001;93:817-22.
20. O’Malley CM, Frumento RJ, Hardy Ma, et al. A randomized double-blind comparison of lactated Ringer’s
solution and 0.9% NaCl during renal transplantation. Anesth Analg 2005;100:1518-24.
21. Williams EL, Hildebrand KL, McCormick SA, et al. The effect of intravenous lactated ringer’s solution
versus 0.9% sodium chloride solution on serum osmolarity in human volunteers. Anesth Analg 1999;88:999-
1003.
22. Reid F, Lobo DN, Williams RN, et al. (Ab)normal saline and physiological Hartmann’s solution: a randomized
double-blind crossover study. Clin Scien 2003;104:17-24.
23. Merten GJ, Burgess WP, Gary LV, et al. Prevention of contrast-induced nephropathy with sodium
bicarbonate: a randomized controlled trial. J A M A 2004;291:2328-34.
24. Briguori C, Airoldi F, D’Andrea D, et al. Renal insufficiency following contrast media administration trial
(REMEDIAL): a randomized comparison of 3 preventive strategies. Circulation 2007;115: 1211-7.
25. Brar SS, Shen AY-J, Jorgensen MB, et al. Sodium bicarbonate vs. sodium chloride for the prevention of
contrast medium-induced nephropathy in patients undergoing coronary angiography. JAMA 2008;300:1038-
46.
26. Knottenbelt JD. Traumatic rhabdomyolysis from severe beating – experience of volume dieresis in 200
patients. J Trauma 1994;37:214-9.
27. Better OS, Stein JH. Early management of shock and prophylaxis of acute renal failure in traumatic
rhabdomyolysis. N Eng J Med 1990;322:825-9.
28. Brown CV, Rhee P, Chan L, et al. Preventing renal failure in patients with rhabdomyolysis: do bicarbonate
and mannitol make a difference? J Trauma 2004: 56:1191-6.
29. Wilcox CS. Regulation of renal blood flow by plasma chloride. J Clin Invest 1983;71:726-35.
30. Gordon D, Nashat FS, Wilcox CS. An analysis of the regulation of sodium excretion during induced changes
in plasma sodium concentrations in anaesthetized dog. J Physiol 1981;314:531-45.
31. Oliveira RP, Velasco I, Soriano FG, et al. Clinical review: Hypertonic saline resuscitation in sepsis. Crit
Care 2002;6:418-23.
32. Rhee P, Wang D, Ruff A, et al. Human neutrophil activation and increased adhesion by various resuscitation
fluids. Crit Care Med 2000;28:74-8.
33. Chiara O, Bucci L, Sara A, et al. Quality and quantity of volume replacement in trauma patients. Minerva
Anestesiol 2008;74:303-6.
34. Cooper DA, Myles PS, McDermott FT, et al. Prehospital hypertonic saline resuscitation of patients with
hypotension and severe traumatic brain injury: a randomized controlled trial. JAMA 2004; 291:1350-57.
35. Bunn F, Roberts IG, Tasker R, et al. Hypertonic versus near isotonic crystalloid for fluid resuscitation in
critically ill patients. Cochrane Database of Sys Rev 2000;updated 2007. CD002045 pub 2.
36. Bruegger D, Bauer A, Rehm M, et al. Effect of hypertonic saline dextran on acid-base balance in patients
undergoing surgery of abdominal aortic aneurysm. Crit Care Med 2005;33:556-63.
37. Huang PP, Stucky FS, Dimick AR, et al. Hypertonic sodium resuscitation is associated with renal failure
and death. Ann Surg 1995;221:543-54.
38. Levere XM, Boon C, Hakim T, et al. Half-molar sodium lactate solution has a beneficial effect on patients
after coronary artery bypass grafting. Intensive Care Med 2008;34:1796-803.
39. Perez-Perez AJ, Pazos B, Sobrado B, et al. Acute renal failure following massive mannitol infusion. Am J
Nephrol 2002;22:573-5
40. Visweswaran P, Massin EK, Dubose TD Jr. Mannitol-induced acute renal failure. J Am Soc Nephrol
1997;8:1028-33.
41. Solomon R, Werner C, Mann D, et al. Effect of saline, mannitol and furosemide to prevent acute decrease
in renal function induced by radiocontrast agents. N Eng J Med 1994;331:1416-20.
42. Weisberg LS, Kurnik PB, Kurnik BR. Risk of radiocontrast nephropathy in patients with and without diabetes
mellitus. Kidney Int 1994;45:259-65.
43. The Albumin Reviewers (Alderson P, Bunn F, Li WPA, et al). Human albumin solutions for resuscitation
and volume expansion in critically ill patients. Cochrane Database of Systm Rev 2004, Issue 4, Art No:
CD001208.pub 2.
44. Finfer S, Bellomo R, Boyce N, et al. A comparison of albumin and saline for fluid resuscitation in the
intensive care unit. New Eng J Med 2004;350:2247-56.
45. Sort P, Navasa M, Arroyo V, et al. Effect of intravenous albumin on renal impairment and mortality in
patients with cirrhosis and spontaneous bacterial peritonitis. New Eng J Med 1999;341:403-9.
46. Barron ME, Wilkes MM, Navickis RJ. A systematic review of the comparative safety of colloids. Arch Surg
2004;139:552-63.
47. Bagshaw S, Bellomo R. The influence of volume therapy on outcome. Curr Opin Crit Care 2007;13:541-
8.
48. Schortgen F, Lacherada JC, Bruneel F, et al. Effects of hydroxyethylstarch and gelatin on renal
function in severe sepsis: a multicentre randomized study. Lancet 2001;357:911-6.
49. Boldt J, Suttner S. Plasma substitutes. Minerva Anestesiol 2005;71:741-58.
50. Legendre C, Thervet E, Page B, et al. Hydroxyethylstarch and osmotic nephrosis-like lesions in kidney
transplantation. Lancet 1993;342:248-9.
51. Boldt J, Priebe HJ. Intravascular volume replacement therapy with synthetic colloids: Is there an influence
on renal function. Anaesth Analg 2003;96:376-82.
52. Bunkhorst FM, Engel C, Bloos F, et al (VISEP study). Intensive insulin therapy and pentastarch resuscitation
in sever sepsis. New Eng J Med 2008;358:125-39.
53. Schortgen F, Girou E, Deye N, et al (CRYCO study group). The risk associated with hyperoncotic colloids
in patients with shock. Intensive Care Med 2008;34:2157-68.
54. Wiedermann CJ. Systematic review of randomized controlled trials on the use of hydroxyethyl starch for
fluid management in sepsis. BMC Emerg Med 2008;8: doi: 10.1186/1471-227X-8-1.
55. Kumle B, Boldt J, Piper S, et al. The influence of different intravascular volume replacement regimens on
renal function in the elderly. Anaesth Analg 1999;89:1124-30.
56. Mahmood A, Gosling P, Vohra RK. Randomized clinical trial comparing the effects on renal function of
hydroxyethyl starch or gelatin during aortic aneurysm surgery. Br J Surg 2007;94:427-33.
57. Boldt J, Brosch Ch, Röhm K, et al. Comparison of the effects of gelatin and a modern hydroxyethyl starch
solution on renal function and inflammatory response in elderly cardiac surgery patients. Br J Anaesth
2008;100:457-64.
58. Godet G, Lehot J-J, Janvier G, et al. Safety of HES 130/0.4 (Voluvenâ) in patients with preoperative renal
dysfunction undergoing abdominal aortic surgery: a prospective, randomized, controlled, parallel-group
multicentre trial. Eur J Anaesthesiol 2008;25:1042;Epub 2008 June 25.
59. Boldt J. Saline versus balanced hydroxyethyl starch: does it matter? Curr Opin Anaesthesiol 2008;21:679-
83.
Section D: Trauma/Neuro-critical Care
Twelve
Sodium-Water Disturbances
in a Neuro ICU
Mathew Joseph, Shalini Nair
INTRODUCTION
Hyponatremia defined as a serum sodium level below 135 mmol/L is the commonest electrolyte
abnormality in patients hospitalized for any cause.1 In a prospective study on 196 patients
admitted in a general surgical – medical hospital Anderson et al2 found that the daily incidence
of clinically significant hyponatremia (defined as a serum Na+ < 130 mmol/L) was 1%.
Hyponatremia was associated with a significant worsening of clinical outcomes in this study,
with a reported mortality of 11% for those with hyponatremia against 0.2% for patients with
normal serum sodium (Na) values. This dramatic difference cannot be directly attributed to
the abnormality itself, as several of the underlying diseases known to cause hyponatremia
have an intrinsically poor prognosis. In addition, severe stress can cause hyponatremia, and
it is possible that the low sodium was an indicator of the sickest patients.
Since water homeostasis is controlled by the central nervous system, the incidence of Na-
water abnormalities is likely to be even higher in a population of neuro ICU patients, though
there are no published data to back this assumption. In this review, we begin by briefly
explaining normal sodium-water control mechanisms and describe the effects of sodium
abnormalities and their correction on the brain with clinical correlates. This is followed by a
discussion of the common causes of hyponatremia and hypernatremia, a description of the
diagnostic algorithms used to make a diagnosis and a discussion on treatment options. A few
specific problems such as surgery in the region of the pituitary and hypothalamus and the use
of mannitol are also discussed.
NORMAL WATER METABOLISM

Sodium metabolism is primarily controlled by the renin-angiotensin-aldosterone mechanism,
while water is controlled by arginine vasopressin (AVP), also called the anti-diuretic hormone
(ADH).3 AVP is a nine amino acid peptide secreted by magnocellular neurons in the supraoptic
and paraventricular nuclei of the hypothalamus and transported down the pituitary stalk to
Sodium-Water Disturbances in a Neuro ICU 123
be released in the posterior pituitary gland. Normal levels are <4 pg/ml; it is metabolized by
vasopressinases in both the liver and kidney and has a half life of about 35 minutes.4
The primary input into these nuclei is from hypothalamic osmoreceptors located in the
subfornicial organ (outside the blood brain barrier), and secondarily from medullary
cardiovascular centres responding to baroreceptors located in the cardiac atria, aorta and
carotid arteries, as well as the carotid bodies and area postrema.5,6 The osmoreceptors maintain
serum osmolality in a tight range between 280 and 295 mOsm/ kg; an increase of as little as
2% causing a significant increase in urine osmolality due to water reabsorption, and a decrease
of 2% causing maximal dilution of urine due to water excretion.7 An increase of osmolality
also stimulates the thirst mechanism. The non-osmotic stimulus from the cardiovascular system
requires a change of 10 to 20% in circulating volume or blood pressure to influence AVP
secretion. AVP production is also stimulated to some degree by nausea, hypoxia, hypercapnea,
stress, hypoglycaemia and IPPV, and can be diminished by opioids.4
AVP acts on the V2 receptors located on the cells of the renal collecting ducts, stimulating
the movement of aquaporin-2 water channels from intracellular vesicles to the apical plasma
membrane to promote reabsorption of water.7 AVP also has a long-term effect on the expression
of the aquaporin-2 gene to increase water channels.8
Though primary control of AVP secretion is in response to serum osmolality, the numbers
of different factors that can affect secretion make dysregulation of water metabolism and the
resulting hyponatremia a frequent problem in the ICU.
EFFECTS OF ABNORMAL SODIUM LEVELS AND THEIR CORRECTION

The effects of hyponatremia on tissues and their clinical correlates depend on the severity of
the hyponatremia and the rate of decline in sodium levels. Though hyponatremia affects all
tissues of the body, the major consequences are due to the effect on the brain. When serum
osmolality drops, water enters the brain cells from the extracellular space, causing neuronal
dysfunction and brain edema with potentially lethal consequences. Decreased level of
consciousness, seizures and raised pressure are common manifestations. If hyponatremia
develops slowly the brain cells adapt by releasing various solutes from the brain.9 These include
K+, Na+, Cl- and organic osmolytes including amino acids, myoinositol and creatine.10
The effect of an acute drop in serum sodium levels has much more significant clinical
consequences than a slower decline, such that patients with chronic hyponatremia often remain
asymptomatic at sodium levels that would cause devastating effects in an acute setting. The
levels at which symptoms of hyponatremia develop vary with the rate of decline, and therefore
it is not possible to define a particular sodium concentration that results in clinical consequences.
Early symptoms are headache, irritability, nausea and confusion, which progress to
hyporeflexia, drowsiness, seizures, coma and eventually brainstem compression due to
herniation and death. In a study of 100 patients with hyponatremia,11 all (100%) patients
with serum Na of 112 mmol/ L that developed over a period of less than 12 hours were
comatose, while only 6% of patients with 115 mmol/ L that developed over more than 3 days
were in coma. Seizures were present in 29% of patients with an acute fall but only in 4% of
the remaining patients, and mortality was 50% and 6% in the respective groups.
Correction of symptomatic hyponatremia is a medical emergency and will be discussed
later in this chapter. A devastating consequence of rapid correction of chronic hyponatremia
is central pontine myelinolysis (now called osmotic demyelination syndrome since
demyelination occurs in the cerebellum and basal ganglia as well) first attributed to correction
of hyponatremia by Kleinschmidt-DeMasters and Norenberg in 1981.12 There is marked
destruction of myelin with relative preservation of axon cylinders. The pathogenesis of this
phenomenon is not fully understood. One theory is that in chronic hyponatremia a sudden
rise in serum sodium leads to brain cell shrinkage, and in order to maintain volume the cell
takes up K+, Na+, Cl– and organic osmolytes.13 These organic osmolytes are thought to be
protective for intracellular proteins and DNA, and since it takes time to accumulate these
compounds, cell damage occurs. It is also possible that the process is related to blood brain
barrier damage in these areas with immune mediated destruction of myelin, as the concen-
tration of IgG is high in the areas of demyelination and dexamethasone has been used
successfully to prevent these lesions in rats.14 Therefore, correction of hyponatremia requires
close monitoring. Osmotic demyelination is unlikely in any situation of acute hyponatremia
above 120 mEq/ L.
Acute hypernatremia can theoretically cause brain cell shrinkage, but the phenomenon
has not been well studied experimentally. Symptoms are initially similar to those of hypo-
natremia with thirst, irritability and lethargy proceeding to hyperreflexia, seizures, coma and
death. These clinical correlates are not as well defined, as the sodium levels often increase
only in patients in whom the thirst mechanism is not functional, and these patients are usually
already unconscious. There is a report of extrapontine myelinolysis in basal ganglia, cerebellum
and hippocampus in three children with sudden onset hypernatremia from previously normal
sodium values.15
HYPONATREMIA
In order to accurately diagnose and treat hyponatremia it is essential to have a rational
diagnostic algorithm for the various etiologies. The standard classification is to divide the
patients on the basis of serum tonicity and then subdivide the cases of hypotonic hyponatremia
based on the volume status. It is not the purpose of this review to discuss all causes of
hyponatremia, and therefore the focus will be on the commonest causes seen in neurological
patients.
Classification of Hyponatremia Based on Serum Tonicity

Hypertonic hyponatremia: This is due to the presence of osmotically active compounds in the
serum that, in turn cause an osmotic movement of water from the intracellular to extracellular
fluid, diluting the sodium though the serum remains hyperosmolar. Common particles
responsible are glucose, mannitol and contrast agents. A correction factor of 2.4 mmol/ L has
been suggested for every 100 mg/ dL of glucose, especially at values over 400 mg/ dL.16
Isotonic hyponatremia: Also known as pseudohyponatremia, this is primarily seen in marked
hyperlipidemia or hyperproteinemia. Normally these compounds constitute about 7% of serum,
with the remainder being aqueous. When this non-aqueous component increases, the amount
of water and therefore the quantity of sodium per unit volume of serum decreases. The sodium
concentration per unit of water is not low, and this is primarily a measurement artifact. The
low sodium values in these patients have no clinical significance, but the clinician must be
aware of this entity.
Hypotonic hyponatremia: Sodium is the main osmotically active particle in normal serum,
and therefore this is syndrome most frequently seen in patients. Patients in this group are
further divided based on their volume status into hypervolemic, euvolemic and hypovolemic.
The common causes of hypotonic hyponatremia are listed in Table 12.1. In a retrospective
study of hypotonic hyponatremia, 26% of patients were found to be hypovolaemic, 24%
hypervolemic and 50% euvolemic.18 The syndrome of inappropriate ADH secretion (SIADH)
and cerebral salt wasting syndrome (CSW) are discussed below.
Table 12.1: Causes of hypotonic hyponatremia17
Hypovolaemic Euvolaemic Hypervolaemic
Cerebral salt wasting SIADH SIADH

Diuretics Thiazide diuretics Cardiac failure
Diarrhoea / vomiting Adrenal insufficiency Cirrhosis
Sweating Hypothyroidism Acute renal failure
Blood loss Iatrogenic Iatrogenic
Adrenal insufficiency
Syndrome of Inappropriate Antidiuretic Hormone Secretion (SIADH)

Pathophysiology: After ruling out causes listed in Table 12.1 in patients with euvolaemic or
hypervolaemic hyponatremia, a significant number of patients will still have developed
hyponatremia due to non-osmotic secretion of AVP. The secretion of AVP without an osmotic
stimulus results in dilutional hyponatremia, and the cause will have to be established from
among a long list of potential diagnoses (Table 12.2).
Diagnosis: The criteria for diagnosis of SIADH laid down in 1967 by Bartter and Schwartz19
are still in use today. In summary the diagnostic criteria are
• Hypotonic (osmolality <280 mOsm/L) hyponatremia (Na <135 mmol/ L).
• Urine sodium >18 mmol/ L.
Table 12.2: Common causes of SIADH3,6
Malignancies Bronchogenic carcinoma, mesothelioma

Carcinomas of the gut or urinary system
Thymoma
Hodgkin’s disease, acute leukaemia, lymphosarcoma
Uterine carcinoma
Nasopharyngeal carcinoma
CNS Meningitis, encephalitis
Trauma, subarachnoid haemorrhage
Stroke, hypoxia, sinus thrombosis
Vasculitis, multiple sclerosis
Tumours, abscess
Chest disorders Tuberculosis, mycoplasma, acute pneumonia, fungal infections
Respiratory failure, COPD, IPPV
Asthma, cystic fibrosis
Drug induced ACE inhibitors
Cyclphosphamide, vincristine
Phenothiazines, tricyclics
Carbamazepine
Morphine, barbiturates
Serotonin reuptake inhibitors
Omeprazole
Ecstasy
Others Prolonged severe exercise
Old age
Idiopathic
• Normal thyroid, renal and adrenal function.

• Clinical euvolaemia.
As mentioned above, practically three quarters of patients with hypotonic hyponatremia
are either euvolemic or hypervolemic.18
Treatment: The underlying cause of the SIADH should be treated, if identified. Hyponatremia
is managed with the following modalities of treatment:
• Fluid restriction to 1000 ml a day is all that is needed in asymptomatic or mildly symptomatic
patients for a gradual correction of serum sodium. However, this is often not possible due
to patient discomfort or because the underlying medical condition does not permit a
decrease in administered fluid (hypotension, subarachnoid hemorrhage, risk of stroke or
renal impairment).
• Hypertonic saline may be used to correct sodium in more severely symptomatic patients
and in those for whom fluid restriction is not possible.
• Loop diuretics may be useful in promoting water excretion.
• Demeclocycline and lithium have been used to inhibit renal response to AVP, but both
have significant side effects.
• AVP receptor antagonists promise to be an effective means of treatment for SIADH. They
block action at the V2 receptor, promoting excretion of free water. The compounds under
trial, collectively known as the vaptans, include tolvaptan, lixivaptan and conivaptan (has
been approved for use in hospital patients). In a double blinded, multicenter, placebo
controlled trial for patients with euvolemic or hypervolemic hyponatremia, conivaptan
was shown to markedly increase serum sodium.20
Cerebral Salt Wasting Syndrome (CSW)

Pathophysiology: CSW is under diagnosed in practice today; several patients with this syndrome
are probably misclassified and treated as SIADH. It is commonly seen in trauma patients and
those with subarachnoid hemorrhage, though it can develop in patients with any CNS
pathology. The precise mechanism of the natriuresis that causes hyponatremia is not clear,
but is believed to be the effect of atrial natriuretic peptide (ANP) and brain natriuretic peptide
(BNP) on sodium absorption at the proximal tubule.21
Diagnosis: In the absence of fluid loss and with normal adrenal function, hypovolemic
hyponatremia in patients with CNS pathology is likely to be due to CSW. Criteria for diagnosis
require
• Hyponatremia (serum sodium < 135 mmol/L).
• Natriuresis (urine sodium > 40 mmol/L).22,23
• Hypovolaemia .
The major differentiation between SIADH and CSW is the volume status. SIADH is a
condition associated with water retention and therefore, cannot have volume depletion.
Treatment: CSW is usually a transient phenomenon, and the goal of treatment is to replace
lost sodium. A treatment protocol for neurosurgery patients with hyponatremia and natriuresis
developed at the authors’ institution22,23 is to administer normal saline to correct the volume
deficit, and to transfuse blood if the hematocrit is less than 27%. Most patients correct their
serum sodium within 72 hours on this regimen. Hypertonic saline may be added in cases of
severely symptomatic hyponatremia. In occasional patients it may be necessary to administer
fludrocortisone to correct and maintain serum sodium.
DIAGNOSTIC ALGORITHM FOR HYPONATREMIA IN THE NEURO ICU

It must be emphasized again that SIADH and CSW are diagnoses of exclusion, and systemic
causes and laboratory error must be ruled out first. For example, in patients with adrenal
insufficiency, steroids have to be administered for sodium correction.
If no underlying causes have been identified in a patient with hyponatremia, the first
investigation should be a urine sodium level. If this is low, the diagnosis is insufficient intake
and the treatment is to provide enough salt either orally or intravenously. If the urine sodium
is high, a central line is inserted and confirmed radiologically and simultaneously a hematocrit
Fig. 12.1: Algorithm for diagnosis of hypotonic hyponatremia in a Neuro ICU22,23
is checked. If the central venous pressure is low, a diagnosis of CSW is made and the volume
deficit is corrected with normal saline and blood transfusion, if necessary. If the central venous
pressure is high then a diagnosis of SIADH is made and fluid is restricted as much as possible.
We tend to treat patients with near normal CVP also as CSW, and decide on further treatment
based on response to treatment. This algorithm is represented diagrammatically in Figure
12.1. Hypertonic saline is used in all cases of severe symptomatic hyponatremia and in SIADH
when fluid restriction is not possible due to other medical reasons.
The possibility of a mixed disorder also exists—volume loss associated with natriuresis can
cause a secondary rise in AVP secretion. It is also possible that the rise in AVP secretion is
independent of the natriuresis and hypovolaemia. In a study on 19 patients with subarachnoid
haemorrhage, one third of them developed hyponatremia in spite of infusion of sufficient
normal saline to maintain intravascular volume at all times.24
Rate of Correction of Hyponatremia

When treating a patient the risks of hyponatremia have to be balanced against the risks of
correction. The severity of symptoms and the acuity of the hyponatremia both influence the
rate at which the sodium can be corrected.
A patient who has developed severe hyponatremia in <48 hours (the generally accepted
definition of acute hyponatremia) and is comatose or having seizures is a medical emergency.
In these patients, treatment with 3% saline can even begin to achieve an initial correction rate
of 3-5 mmol/L per hour to avoid irreversible brain damage.25 In less severe acute hyponatremia,
the initial rate of increase should be no more than 1-2 mmol/ L per hour. The correction rate
can be slowed down once a level of 125 mmol / L has been reached.
In mildly symptomatic patients and in patients with chronic hyponatremia an hourly increase
of 0.5 mmol/ L is recommended to minimize the risks of complications. Young women are
particularly vulnerable to complications of rapid correction,9 and special care must be taken.
Serum sodium must be measured frequently (every 2-4 hours) during the acute phase of
correction.
HYPERNATREMIA
Sodium is the main osmotically active substance in the body, and therefore hypernatremia is
always associated with a hyperosmolar state. For logical diagnosis it is useful to consider
hypernatremia in 3 different scenarios:26
Hypervolaemic hypernatremia: It is an uncommon state caused usually caused by
administration of hypertonic fluids or enteral feeds with insufficient free water. It is occasionally
reported due to an increased intake of oral salt.27
Inadequate fluid intake: This is usually seen in postoperative patients or those who are unable
to ingest water in response to thirst. A rare cause of inadequate fluid intake is damage to the
thirst centres in the hypothalamus, usually seen after surgery in the region.
Increased water loss: This is the commonest cause of hypernatremia—the loss of free water
from the system. High blood sugar with glycosuria, gastrointestinal losses, hypokalaemia and
hypercalcemia can all cause this phenomenon. In the absence of any of these, water loss
could be due to insufficient AVP secretion with resultant water loss.
Diabetes Insipidus (DI)

Pathophysiology: DI can be divided into central (insufficient secretion of AVP) or nephrogenic
(an abnormal response of the kidneys to available AVP).
Central DI can be congenital or due to any lesions that affect the hypothalamo-pituitary
axis. Common causes are trauma, subarachnoid haemorrhage, tumours and surgery in the
region, granulomatous disease and some autoimmune conditions. DI does not manifest until
at least 90% of the magnocellular neurons that produce AVP are destroyed. DI is also seen
as a preterminal event in brain death.
Nephrogenic DI is due to end organ resistance of the kidney and can be congenital due
to mutations of either the V2 receptor or aquaporin water channels. It may also be seen in
severe hypercalcaemia and hypokalaemia.
Diagnosis: Once other causes of polyuria such as high sugars have been ruled out, DI can be
diagnosed in the presence of:
• Serum sodium >145 mmol/L.
• Polyuria with output > 3 ml/kg/hour.
• Dilute urine with a specific gravity < 1.005.
• Poydipsia in awake patients with intact thirst mechanisms OR hypovolaemia in patients

unable to drink water.
Treatment: Once central DI has been confirmed the treatment consists of replacement of the
water deficit and replacement of the AVP deficiency. In awake patients with access to water,
severe hypernatremia rarely develops, as the thirst mechanism will ensure that the patient
drinks enough water. In patients unable to drink adequately either due to nausea or altered
sensorium, hypotonic fluids may be administered either intravenously or through a nasogastric
tube to correct the water deficit. Care must be taken during this process not to cause fluid
overload.
Definitive treatment consists of administration of desmopressin (1-deamino-8-D-arginine
vasopressin) either intranasally or parenterally28. As in the correction of hyponatremia, rapid
reduction of serum sodium can result in cerebral oedema. However, the risk is less in these
patients, since DI is usually diagnosed early and correction of acute hypernatremia is unlikely
to have complications.
DI Following Surgery in the Suprasellar Region

Postoperative DI can be transient, permanent or triphasic.29 Transient DI is thought to be due
to damage to the pituitary stalk, and resolves in a few days as full function of the AVP producing
neurons resumes. If the damage extends to the magnocellular neurons then the DI might be
permanent. A major determinant of the duration of DI is the level of section of the pituitary
stalk—if it is near the median eminence then almost 100% of patients develop permanent
DI.30 The classic triphasic response begins with the transient DI. This is followed 5-7 days
later by a period of SIADH as AVP in the disintegrating cells is released, lasting up to two
weeks.29 After these stores are consumed, the third phase of prolonged DI begins if enough
of the magnocellular neurons have been destroyed.
Mannitol in the Neuro ICU

Mannitol is one of the most common drugs used in the Neuro ICU. Osmotic diuresis induced
by mannitol helps reduce intracranial pressure. The effect of intravenous infusion of mannitol
on serum sodium is variable. Although traditionally thought to raise sodium levels due to
dehydration caused by osmotic diuresis, with modern standards of care, it is more likely to
cause hypertonic hyponatremia.31 Frequent measurement of electrolytes (at least once a day
in the absence of fluid balance abnormalities) is mandatory.
CONCLUSION
Alterations of sodium and water metabolism are common in critically ill patients. A logical
approach to detected abnormalities in most cases enables an accurate diagnosis. Rigorous
monitoring of fluid status and electrolytes is an indispensable part of management and
knowledge of complications associated with the abnormality and correction of tonicity promotes
patient safety.
REFERENCES
1. Wong LL, Verbalis JG. Vasopressin [V.sub.2] receptor antagonists. Cardiovasc Res 2001;51:391-402.
2. Anderson RJ, Chung H-M, Kluge R, Schrier RW. Hyponatremia: a prospective analysis of its epidemiology
and the pathogenetic role of vasopressin. Ann Intern Med 1985;102:164-8.
3. Adler SM, Verbalis JG. Disorders of body water homeostasis in critical illness. Endocrinol Metab Clin N
Am 2006;35:873-94.
4. Sharman A, Low J. Vasopressin and its role in critical care. Cont Education in Anaesth, Crit Care and
Pain 2008;4:134-7.
5. Robinson AG, Verbalis JG. The posterior pituitary. In: Larsen PR, Kronenberg HM, Melmed S, et al,
editors. Williams Textbook of Endocrinology, 10th edition. Philadelphia:WB Saunders; 2003;281-329.
6. Multz AS. Vasopressin dysregulation and hyponatremia in hospitalized patients. Intensive Care Med
2007;22(4):216-23.
7. Wong LL, Verbalis JG. Systemic diseases associated with disorders of water homeostasis. Endocrinol
Metab Clin North Am 2002;31:121-40.
8. Nielsen S, Kwon T-H, Christensen BM, et al. Physiology and pathophysiology of renal aquaporins. J Am
Soc Nephrol 1999;10:647-63.
9. Lien YH, Shapiro JI. Hyponatremia: clinical diagnosis and management. Am J Med 2007;120(8): 653-
8.
10. Lien YH, Shapiro JI, Chan L. Study of brain electrolytes and organic osmolytes during correction of
chronic hyponatremia: implications for the pathogenesis of central pontine myelinolysis. J Clin Invest
1991;88(1):303-9.
11. Arieff AI, Llach F, Massry SG. Neurological manifestations and morbidity of hyponatremia: correlation
with brain water and electrolytes. Medicine 1976;55:121-9.
12. Kleinschmidt-DeMasters BK, Norenberg MD. Rapid correction of hyponatremia causes demyelination:
relation to central pontine myelinolysis. Science 1981;211:1068-70.
13. Yancey PH, Clark ME, Hand SC, et al. Living with water stress: evolution of osmolyte systems, Science
1982;217:1214-22.
14. Sugimura Y, Murase T, Takefuji S, et al. Protective effect of dexamethasone on osmotic-induced
demyelination in rats. Exp Neurol 2005;192(1):178-83.
15. Brown WD, Caruso, JM. Extrapontine myelinolysis with involvement of the hippocampus in three children
with severe hypernatremia. J Child Neurol 1999;14(7):428-33.
16. Hillier Abbott RD, Barrett EJ. Hyponatremia: evaluating the correction factor for hyperglycemia. Am J
Med 1999;106:399-403.
17. Bradshaw K, Smith M. Disorders of sodium balance after head injury. Cont Education in Anaesth, Crit
Care and Pain 2008;8(4):129-33.
18. Bennani SL, Abouqual R, Zeggwagh AA, et al. Incidence, causes and prognostic factors of hyponatremia
in intensive care. Rev Med Interne 2003;24:224-9.
19. Bartter FC, Schwartz WB. The syndrome of inappropriate secretion of antidiuretic hormone. Am J Med
1967;42:790-806
20. Ghali JK, Koren MJ, Taylor JR, et al. Efficacy and safety of oral conivaptan: a V1A / V2 receptor antagonist,
assessed in a randomised placebo controlled trial in patients with euvolemic or hypervolemic hyponatremia.
J Clin Endocrinol Metabol 2006;91:2145-52.
21. Betjes MG. Hyponatremia in acute brain disease: the cerebral salt wasting syndrome. Eur J Intern Med
2002;13:9-14.
22. Sivakumar V, Rajshekhar V, Chandy MJ. Management of neurosurgical patients with hyponatremia and
natriuresis. Neurosurgery 1994;34(2):269-74.

23. Damaraju SC, Rajshekhar V, Chandy MJ. Validation study of a central venous pressure based protocol for
the management of neurosurgical patients with hyponatremia and natriuresis. Neurosurgery 1997;40(2):312-
6.
24. Diringer MN, Wu KC, Verbalis JG, et al. Hypervolemic therapy prevents volume contraction but not
hyponatremia following subarachnoid hemorrhage. Ann Neurol 1992;31(5):543-50.
25. Gross P, Reimann D, Neidel J, et al. The treatment of severe hyponatremia. Kidney Int Suppl 1998;64:S6-
11.
26. Loh JA, Verbalis JG. Disorders of water and salt metabolism associated with pituitary disease. Endocrinol
Metab Clin N Am 2008;37:213-34.
27. Moder KG, Hurley DL. Fatal hypernatremia from exogenous salt intake: report of a case and review of
literature. Mayo Clin Proc 1990;65(12):1587-94.
28. Richardson DW, Robinson AG. Desmopressin. Ann Intern Med 1985;103(2):228-33.
29. Hollinshead WH. The interphase of diabetes insipidus. Mayo Clin Proc 1964;39:92-100.
30. Lipsett MB, Maclean JP, West CD, et al. An analysis of the polyuria induced by hypophysectomy in man.
J Clin Endocrinol Metab 1956;16(2):183-95.
31. Yun JJ, Cheong I. Mannitol induced hyperosmolal hyponatremia. Intern Med J 2008;38(1):73.
Delirium in the ICU and its Consequences 133
Thirteen
Delirium in the ICU
and its Consequences
Gareth L Thomas
INTRODUCTION
The American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders
(DSM-IV) defines delirium as a disturbance of consciousness with inattention and cognitive
disturbance that develops over a short period of time (hours to days) and fluctuates over
time.1 Within the ICU, various terms have been applied to this syndrome, and delirium has
been considered to be a component of critical illness2 occurring secondary to medication
effects. Although associated with potential adverse outcome,3-5 delirium amongst the critically
ill has received little attention until recent years.
Recognition of ICU delirium as a clinical entity is now increasing; techniques used for
diagnosis and management, however, are still suboptimal. In a survey of 912 healthcare
professionals conducted during 2001-02, delirium was perceived to be a common problem
amongst critically ill patients. Respondents indicated that delirium is under diagnosed and
acknowledged that it had implications for outcome.2 Despite this, only 32% believed that
routine monitoring for ICU delirium represented evidence based practice. Of those respondents
who screened for delirium, only 16% reported using specific assessment tools; the majority
used tools that were not validated for this purpose. The perception that delirium screening
will not change management or outcome, is likely to impair the introduction of these tools
into routine clinical practice.6,7 Although acknowledged to be delirogenic, benzodiazepines,
particularly lorazepam, were not infrequently prescribed for the management of delirium.2
ICU delirium may go unrecognised amongst the critically ill; some patients with delirium
may appear withdrawn and lethargic. Of all delirious ICU patients, 23-94% may present with
“hypoactive delirium”.3,4,8,9 This data is clearly confounded by the presence of ongoing sedative
and analgesic medication, but it is likely that a spectrum exists between the “hypoactive”,
“mixed” and “hyperactive” subtypes of delirium. Such information reinforces the need for
routine screening for the presence of delirium and excessive sedation.
There is scope, therefore, to raise awareness of ICU delirium and improve the care of this
patient group. This review will discuss the current literature and related controversies as
applicable to incidence, risk factors, diagnosis, outcome effects and patient management.
RISK FACTORS FOR THE Table 13.1: Risk factors for ICU delirium
DEVELOPMENT OF DELIRIUM Underlying patient factors
IN THE CRITICALLY ILL • Cerebrovascular disease
• Baseline dementia
There are numerous risk factors for the develop- • Age
ment of delirium in the critically ill. These can • Genetic predisposition
be classified as follows3,10 (Table 13.1): • Hypertension
• Sensory impairment
• Underlying patient factors. • Smoking and alcohol history
• Acute illness related factors.
Acute illness related factors
• Iatrogenic or environmental factors. • Illness severity
These risk factors can be further sub- • Hypoxia
• Hypotension
classified as either modifiable or unmodifiable.
• Metabolic derangement
Ely3 observed an 81% incidence of delirium in • Sepsis
a cohort of 48 medical ICU patients. Each • Sleep-wake cycle interruption
patient within the cohort had a mean of 11 ± • Drug withdrawal
• Central nervous system ischaemia/infection
4 risk factors for delirium. • Seizure activity
Sedative and analgesic medication is asso-
Iatrogenic or environmental factors
ciated with an increased risk of ICU delirium. • Sedative medications
In Ely’s study,3 the most common risk factor was • Anticholinergic drug effects
the use of benzodiazepines or opioids - 98% of • Drug interactions
patients with delirium received these drugs. • Epidural analgesia
5 • Unfamiliar, noisy environment
Similarly, Ouimet observed an increased risk • Urinary catheterisation
of delirium (OR 3.22 95% CI 1.5–6.8) when
sedative and analgesic medication was used to
induce iatrogenic coma. Pandharipande11 reported that the use of lorazepam was an
independent risk factor for delirium. All patients receiving a total daily dose greater than 20mg
lorazepam became delirious. Other sedatives and opioids were associated with a trend towards
the development of delirium. Dubois12 reported that morphine was associated with delirium,
with an odds ratio of 6.0–9.2 depending upon the dose used. When appraising these studies,
it is important to consider that these medications may have been administered to manage
delirium or rather could have been a contributory cause of the delirium. This will be considered
in greater detail later in this review.
The severity of physiological derangement as assessed by increased APACHE II score,
PaO 2 : FiO 2 ratio < 200 mmHg, and decreased arterial pH is associated with ICU
delirium.5,11,13,14 For every additional point on the APACHE II scale, the risk of delirium
increases by 5-6%.5,11 Chronic health issues associated with the development of delirium
include hypertension (OR 1.9-2.6), alcohol excess (OR 2.0), smoking (OR 2.2) and
hyperbilirubinaemia (OR 1.2).5,12 The effect of age is controversial. Ouimet5 reports no effect
of age upon the development of delirium in a cohort of 820 medical and surgical ICU patients.
By contrast, Pandharipande11 observed that for each additional year above age 65 there was
a 2% increase in the risk of delirium.
Genetic predisposition to ICU delirium may be determined by an individual’s apolipoprotein
genotype. Apolipoprotein E exists in three isoforms – ApoE2, ApoE3, ApoE4 – and likely
modifies inflammatory response within the central nervous system. The ApoE4 genotype has
been associated with poor outcome following brain injury and with late onset Alzheimer’s
disease. In a cohort of 53 medical ICU patients with an 89% incidence of delirium, Ely15
observed that 23% of patients were either homozygous or heterozygous for the ApoE4 allele.
These patients were younger than those patients without an ApoE4 allele (53.2 ± 21.9 yrs
vs 65.4 ± 13.4 yrs), and had a longer duration of delirium (median 4 days vs 2 days).The
presence of an ApoE4 allele was the strongest predictor of delirium duration when compared
to other covariates (age, presence of sepsis, illness severity, use of lorazepam). The authors
postulate that ApoE4 may be a relatively less effective immunomodulator which may allow
uncontrolled inflammation of the central nervous system and hence delirium.
INCIDENCE OF ICU DELIRIUM

These multiple risk factors are reflected in the variable reported incidence of ICU delirium of
between 22% and 82%. This variability is due to differing patient case-mix, sedation practice
and the occurrence of medication-induced coma. Delirium may be diagnosed using different
scoring tools – for example the Confusion Assessment Method for the ICU (CAM-ICU),16 the
Intensive Care Delirium Screening Checklist (ICDSC)17 or the DSM-IV criteria of the American
Psychiatric Association.1 These tools will be discussed in detail later: it is important to consider
here that the use of different scoring tools and the degree to which they are confounded by
patient sedation will have an effect upon the reported incidence of delirium.
Lin4 reports a 22% incidence of delirium over 5 days in 102 mechanically ventilated medical
ICU patients. 45% of delirious patients cases manifested on day 2 of the ICU stay. Patients
receiving morphine >50 mg/24hours or midazolam > 0.09 mg/kg/hour were excluded a priori,
which may account for the relatively low incidence of delirium in this study.
Using the ICDSC in 820 medical and surgical ICU patients, Ouimet5 reports a 32% incidence
of delirium over a mean 5.7 ± 7 days per ICU patient. The mean daily dose of opioid was
also low in this study (8.9 ± 24.3 mg morphine equivalents). Of note, the dose of sedative
and analgesic medication did not differ between delirious and non-delirious patients. The
authors state that they do not consider that drug induced depression of consciousness represents
delirium. Furthermore, they observe that studies using the CAM-ICU scoring system may
report an elevated incidence of delirium, as sedation related fluctuations in conscious level
may “trigger” a diagnosis of delirium.
Using either DSM-IV criteria or CAM-ICU to diagnose delirium, Ely reports an 81% incidence
of delirium.3,18 In the first of these studies,3 sedation and analgesic practices were not controlled
and remained at the discretion of the attending physicians; these drugs were not interrupted
or modified during delirium assessments. The administration of benzodiazepines or opioids

was identified as a risk factor for delirium in 98% of cases, although the cumulative doses
were not recorded as part of the study. On simple linear regression, the duration of delirium
was correlated with the duration of benzodiazepine and opioid use. In the second of these
studies,18 Ely does not state whether sedation and analgesia were withheld to permit delirium
scoring. Delirium developed in 81.7% of patients for a median of 2 days (IQR 1-3). Of these
delirious patients, 67% were also recorded as being comatose for a median of 2 days (IQR
1-4). It is unclear whether coma occurred due to the titration of sedation. The mean doses
of lorazepam, propofol and opioids were greater in the delirium group, although this was
statistically significant only for lorazepam. Of note, delirium developed in only 54.5% of patients
who were easily arousable. The variable sedation practices in these studies, the unclear role
of sedation during delirium assessment, and the lower incidence of delirium in the patients
who were arousable imply that the relative contribution of sedation and true delirium can be
difficult to determine in critically ill patients.
When assessing all aspects of ICU delirium—etiology, diagnostic scoring systems,
management strategies, prognostic implications—it is important to consider that the underlying
differences described above have a marked confounding influence upon the diagnosis of
delirium and hence upon the interpretation and significance of the current literature.
SCORING FOR SEDATION AND DELIRIUM IN THE ICU

Consciousness can be defined as the combination of a person’s level of arousal plus the
content of consciousness (e.g. the presence or absence of delirium).19 Hence, when assessing
the conscious state of a critically ill patient, the observer should consider 3 interrelated
component features: sedation, analgesia and delirium.20
Excessive sedation is associated with prolonged duration of mechanical ventilation and
increased length of ICU and hospital stay.21 Conversely, appropriate titration of sedation may
reduce the number of days of mechanical ventilation and also the length of ICU and hospital
stay.22,23 Despite this evidence, sedation scoring scales are not applied appropriately in all
ICUs. European data suggests that scales are used in 18-72% of ICUs; in many instances,
sedation may not be reduced over time and analgesia is not titrated to requirements.24,25
The Society of Critical Care Medicine (SCCM) recommends both the use of validated
sedation assessment scales and the routine assessment for the presence of delirium in critically
ill patients.20 A sedation goal should be prescribed for each patient as part of a sedation
protocol. This goal should be subject to regular reassessment, taking into consideration the
patient’s clinical status and their response to sedative therapy.
Sedation Scoring
The Richmond Agitation Sedation Scale (RASS) is a 10 point scale which describes 4 levels
of anxiety (+4 to +1) and 5 levels of sedation (-1 to -5) (Table 13.2).26 It can be completed
within 60 seconds, and is based upon patient observation and response to verbal and physical
Table 13.2: The Richmond agitation-sedation scale (RASS)
Score Term Observer action
+4 Combative Observation only

+3 Very agitated Observation only
+2 Agitated Observation only
+1 Restless Observation only
0 Alert and calm Observation only
–1 Drowsy Verbal stimulus
–2 Light sedation Verbal stimulus
–3 Moderate sedation Verbal stimulus
–4 Deep sedation Physical stimulus
–5 Unrousable Physical stimulus
stimulation. Scores between -1 and -3 are differentiated by the degree of patient eye contact
following verbal stimulation—this aspect of the scoring system therefore allows titration of
sedation. The RASS has been demonstrated to have good inter-observer reliability and
validity26 when used by a multidisciplinary team across a diverse population of critically ill
patients. For these reasons, this scale has been widely applied in much of the published literature
investigating ICU delirium.
The RASS can be used serially to monitor a patient’s sedation status over consecutive
days in ICU. Ely19 validated the RASS over time in a cohort of 275 patients with a mean
APACHE II score of 25.0; all patients were receiving mechanical ventilation. RASS was perfor-
med by Critical Care nurses, an Intensivist and a Psychiatrist. Validity was assessed in 3 ways:
• Criterion validity – comparison with externally derived criteria. In this study the opinion
of the neuropsychiatrist was taken as a reference standard.
• Construct validity – comparison with other supporting measures, e.g. Glasgow Coma Score,
quantities of sedative drugs administered, electroencephalography.
• Face validity – does the RASS appear to measure what it was designed to measure? This
was assessed through direct questioning of the critical care nurses.
As in previous investigations,26 the inter-relater reliability was very high across the different
professional groups. The RASS scores discriminated between different levels of consciousness
when compared to the opinion of the neuropsychiatrist. This correlation was maintained over
successive days of ICU care, and correctly identified changes in the patients’ levels of
consciousness. RASS scores correlated with patients’ GCS over time, sedative doses and also
EEG activity. Face validity was demonstrated in that 77% of nurses felt that the RASS score
was relevant and easy to use.
Delirium Scoring
Concurrent with sedation scoring, ICU patients should also be scored for the presence of
delirium. In patients with verbal communication, the diagnosis of delirium requires assessment
by formally trained psychiatrists using the DSM-IV criteria of the American Psychiatric
Association.1 This time consuming process is not appropriate for serial delirium scoring in
ventilated ICU patients who are unable to talk. ICU patients may also have impaired vision
or hearing, and variable levels of consciousness due to the use of sedative and analgesic
medications. The use of specific ICU delirium scoring systems will be considered below.
The CAM-ICU
The Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) is used in
combination with the RASS, and assesses four features:16
1. Acute onset of mental status changes or fluctuating course within preceding 24 hrs.
2. Inattention (difficulties focussing, maintaining or shifting attention).
3. Disorganised thinking (incoherent ideas, inability to follow questions and commands).
4. Altered level of consciousness (any level other than alert).
Fluctuation in conscious level is defined as a change in a patient’s GCS or RASS scores
within the preceding 24 hours. Features 2, 3 and 4 are assessed using a standard pro-forma
and a series of simple pictures published by Ely’s group.16,27 These are available on the internet
(www.icudelirium.org). Delirium is diagnosed when both features 1 and 2 are present on the
CAM-ICU, along with either features 3 or 4. It is important to note that patients should be
scored using the RASS prior to CAM-ICU assessment. Patients with RASS scores of -4 (deeply
sedated) or -5 (unarousable) cannot be assessed using the CAM-ICU, as they are unable to
respond to verbal commands (Table 13.2).
The CAM-ICU has been shown to have high inter-observer reliability and high sensitivity
and specificity when compared to formal assessments using DSM-IV criteria. This has been
demonstrated in cohorts of both ventilated and non-ventilated patients.16,27,28 The CAM-ICU
also performed well in subgroups of patients who could be expected to be difficult to assess
(age > 65 years; suspected dementia; higher severity of illness; intubated patients). The multi-
disciplinary team was able to use the CAM-ICU after a brief period of training, with patient
assessments typically being completed within 2-3 minutes.
The CAM-ICU has not been validated in
trauma patients, either with or without traumatic Table 13.3: The Intensive Care Delirium
brain injury (TBI). 30% of patients with TBI may Screening Checklist (ICDSC)29
not be suitable for assessment using CAM-ICU
• Altered level of consciousness**
due to RASS scores of –4 or –5.The utility of • Inattention
CAM-ICU scoring in this group of patients has • Disorientation
therefore been questioned, with 48% of nurses • Hallucination-Delusion-Psychosis
stating that the CAM-ICU did not enhance • Inappropriate speech or mood
patient management.5 • Sleep-wake cycle disturbance
• Symptom fluctuation
Intensive Care Delirium Screening
Checklist (ICDSC) **Altered consciousness scores 1 point for either a
response to mild/moderate stimulation or an
The Intensive Care Delirium Screening Checklist exaggerated response to normal stimulation. Normal
(ICDSC)29 is a screening checklist of 8 items wakefulness scores 0.
based upon DSM-IV criteria (Table 13.3). One point is scored for each item manifested by the
patient. As with the CAM-ICU, patients with markedly impaired levels of consciousness
(unarousable or response only upon repeated intense stimulation) cannot be assessed using
the ICDSC. When compared with a formal psychiatric assessment to diagnose delirium in an
ICU population, a score of 4 or more points on the ICDSC correctly identified delirium in
93% of patients.29 Inter-observer reliability between critical care nurses and physicians was
>94%. ROC curve analysis demonstrated a sensitivity of 99% but a specificity of just 64%.
Of those 36% of patients incorrectly identified by the ICDSC as having delirium, 93% had a
confounding diagnosis of a psychiatric illness, dementia, structural neurological abnormalities
or hepatic encephalopathy. It is notable that such diagnoses have been pre-defined as exclusion
criteria in studies evaluating the CAM-ICU.16,27 An evaluation of CAM-ICU and ICDSC using
374 paired scores in 174 patients, however, demonstrated good agreement between the two
scoring systems (kappa coefficient 0.8).30 Further, comparison between CAM-ICU and ICDSC
is given in Table 13.4.
Table 13.4: Comparison of Delirium Scoring Systems: the CAM-ICU and ICDSC
CAM-ICU ICDSC
Easy to use; permits rapid assessment Easy to use; permits rapid assessment
High inter-observer reliability High inter-observer reliability
High sensitivity; high specificity High sensitivity; low specificity
Requires additional testing materials Much information collected as part of routine daily ICU
assessment
Does not require verbal response from patient Inappropriate speech is an item within the score
Utilises the opinion of a single assessor at a Can summarise the opinion of multiple assessors over a
single point in time 24 hour period
Simple objective tests Relatively subjective assessment
Used as a diagnostic tool Used as a screening tool
These scoring processes can be successfully introduced into daily practice as part of a
quality improvement process. Pun et al7 describe implementation of sedation/ delirium monitor-
ing on two separate ICUs over a 21 month period, targeting > 4,000 ICU patient days. RASS
was documented every 4 hours, with CAM-ICU scoring at least once per 12 hour shift.
Education was provided using multi-modal teaching aids. Nursing staff demonstrated a high
degree of agreement with the reference standard, and reported an improved assessment of
patients’ conscious status, an improved understanding of delirium and a greater ability to
achieve a target sedation level. The reported obstacles to change included time and (untrained)
medical staff.
IMPACT OF DELIRIUM ON ICU OUTCOME

Onset of delirium during an ICU stay has implications for both short term and long-term
outcome. Patients who become agitated (i.e., develop hyperactive delirium) are at risk of self-
harm and are commonly seen to remove medical devices. This alone has been estimated to
cost US$181 per removal event (1997 data).31 Agitated patients are at greater risk of inadvertent
catheter removal and self extubation;12,13 the majority require emergent re-intubation. Carrion32
documented the accidental removal rate of medical devices in ICU patients. The authors do
not state the incidence of delirium in this patient cohort. Expressed per 1000 days in situ, 16
endotracheal tubes, 49 nasogastric tubes, 6 venous catheters and 15 arterial catheters were
removed by patients. Of those who accidentally self-extubated, 88% required re-intubation
within 8 hours. Re-intubation was associated with increased mortality; 25% of those patients
requiring re-intubation after unplanned extubation died, whilst there were no deaths amongst
those patients who did not require emergent re-intubation.
It has been documented that patients who develop delirium during their ICU admission
have increased length of ICU and hospital stay and increased mortality. This data is, of course,
subject the caveats expressed elsewhere in this review with regard to the incidence of delirium
and its relation to the level of ongoing sedation. In a cohort of medical and surgical ICU
patients, Ouimet5 reported that the occurrence of delirium increased mean ICU length of stay
from 4.4 to 11.5 days and hospital length of stay from 13.2 to 18.2 days. In patients who
developed delirium, ICU mortality increased from 10.3 to 19.8% and hospital mortality
increased from 21.4 to 26.7% after adjustment for age and APACHE II scores. Lin28 reported
that delirious patients had higher ICU mortality (63.6% vs 32.5%) with a hazard ratio for
death of 2.57 (95% CI 1.56 – 8.15). This relationship between delirium and increased mortality
was reportedly independent of age, illness severity, co-morbidities or any laboratory
derangements.
Not only the occurrence of delirium but also its duration may be of prognostic significance.
Ely3 reported that both ICU and hospital length of stay were correlated with the duration of
delirium. ICU delirium was the strongest predictor of hospital length of stay after adjusting for
illness severity and age. In a subsequent publication, Ely’s group18 reported that the risk of
death increased by 10% for each additional day of ICU delirium (hazard ratio 1.1; 95% CI
1.0-1.3). Delirious patients were twice as likely as non-delirious patients to remain in the
hospital at any given time during their total hospital stay. Adjusting for covariates, delirious
patients had a 3-fold increase in mortality at 180 days.
Thomason also describes a relationship between delirium and adverse outcome in his
study of ICU patients who did not receive invasive mechanical ventilation.14 The presence of
delirium increased both ICU and hospital length of stay. Total hospital mortality was noted to
increase from 6 to 19%, although delirium was not an independent predictor for death in this
group. Similar to the evidence described above for ventilated patients, delirious non-ventilated
patients were 41% more likely than non-delirious patients to remain in hospital at any given
time after adjusting for other covariates.
Long-term neurocognitive impairment has also been reported more frequently in association
with occurrence of delirium in the ICU. Neurocognitive impairment is estimated to occur in
25-78% of ICU survivors. This typically affects memory, executive function, attention,
processing speed, intellectual function and visual spatial awareness.33 Ely3 reported marked
cognitive decline in a cohort of medical ICU patients having a very high (81%) incidence of
delirium. Although just 22% of the overall cohort (both delirious and non-delirious) had
evidence of possible dementia at ICU admission, 58% had cognitive impairment at the time
of ICU discharge. Patients developing delirium have been documented to have a 9-fold risk
of cognitive impairment.18 However, this data is confounded by the fact that these patients
were not screened for pre-existing cognitive impairment at ICU admission. These associations
have lead to speculation that ICU delirium and subsequent cognitive decline should be
considered as a continuum that is dependent upon individual patients susceptibility to insults
like hypotension, hypoxia or metabolic derangements, which occur during critical illness.33
ICU delirium increases hospital costs. Adjusting for multiple covariates, the presence of
delirium has been reported to increase ICU costs by a factor of 1.39 (95% CI 1.12–1.72) and
total hospital costs by a factor of 1.31 (95% CI 1.01–1.70).34 Studies go so far as describing
a “dose-response” relationship between the severity of delirium (as assessed by RASS scores)
and total costs. These increased costs are invariably related to an increased length of stay, as
daily costs were unaffected by the presence of delirium. At present, it is unknown whether
minimising the incidence of ICU delirium will reduce the overall costs of hospital stay.
MANAGEMENT OF DELIRIUM
The principles of management of the at risk patient are as follows:
• Screen all patients for the presence of risk factors.
• Minimise the impact of modifiable clinical risk factors, wherever possible.
• Minimise the use of sedation and analgesia according to published guidelines.
• Screen all critically ill patients for delirium using a validated tool.
• Consider multi-component environmental modifications to reinforce patient orientation.
• Consider the use of appropriate medications to treat delirium, if needed.
Management of Patients at Risk

Modifiable risk factors should be identified and controlled. Clearly, as part of routine ICU
management, hypoxia, hypotension, electrolyte disorders and derangements of glucose control
will be corrected, wherever possible. Drug and alcohol withdrawal should be treated, whilst
the use of medication with anti-cholinergic side effects should be minimised. Benzodiazepines
should be avoided as these impair slow wave sleep and may cause increased delirium via
activation of GABA receptors. As recommended by The Society for Critical Care Medicine,
sedation should be titrated to a sedation assessment scale, and patients should be routinely
assessed for the presence of delirium.20
A recent Cochrane review35 found a paucity of evidence to support specific interventions

for the prevention and treatment of delirium in either general hospital patients or in the critically
ill. Inouye36 described a multi-component protocol targeting six risk factors for delirium in
general medical (non-ICU) patients. A trained multidisciplinary team addressed cognitive
impairment, sleep deprivation, immobility, visual and hearing impairment, and dehydration.
Patients who received this intervention had a lower incidence of delirium (9.9% versus 15.0%)
and fewer overall days of delirium (105 versus 161 days). Patients also demonstrated improved
orientation and required fewer sedative drugs to promote sleep. Unfortunately, this did not
result in an improved outcome at 6 months post hospital discharge as assessed by self-rated
quality of health, activities of daily living, nursing home stay, depression, cognitive impairment
or presence of delirium.37
Other studies have assessed non-ICU patients presenting for elective or emergent hip surgery.
In a randomised trial, Marcantonio38 studied the effect of a proactive geriatrics consultation
in 126 patients aged > 65 undergoing emergent hip surgery. The consultations considered
the following multi-component clinical areas: oxygen delivery, fluid and electrolyte balance,
pain management, drug interactions, bowel and bladder function, nutritional intake, early
mobilisation, cardiorespiratory complications, environmental stimulation and management
of agitation. The intervention group had a lower incidence of delirium (number needed to
treat = 5.6). Patients without pre-existing cognitive impairment derived the greatest benefit
from intervention. Prophylactic haloperidol may be of benefit in this patient group. Kalisvaart39
randomised 430 hip surgery patients to perioperative haloperidol 1.5 mg/24hours or placebo.
Although there was no difference in the incidence of delirium, the haloperidol group had less
severe delirium of shorter duration and also a shorter hospital stay. No side effects of haloperidol
were reported with this regimen.
Treatment
Therapeutic haloperidol is the preferred agent for the management of ICU delirium (grade C
recommendation)20 and as such is used by around two thirds of practitioners.2 The beneficial
effect of haloperidol is thought to be mediated through its dopamine antagonist action. An
intravenous loading dose of 2 mg is given, with repeated double doses given every
15-20 minutes until agitation has subsided. Typical median dose has been reported as 30 mg/
24 hours.2 A haloperidol infusion (3-25 mg/hr) may be required. Possible adverse effects
include QT prolongation with ventricular arrhythmias, extrapyramidal symptoms, and
neuroleptic malignant syndrome.
Newer sedative agents may be of benefit in the critically ill patient. Olanzapine has fewer
extrapyramidal side effects and no active metabolites. It has been associated with an increased
risk of stroke in elderly patients with dementia and so is contraindicated in this group. Skrobik40
randomised patients on a mixed medical and surgical ICU to either enteral olanzapine or
haloperidol. The mean daily dose of olanzapine was 4.5 mg (range 2.5–13.5 mg) and of
haloperidol was 6.5mg (range 1-28 mg). Both drugs were equally effective at controlling
delirium; 14% of the haloperidol group developed mild extrapyramidal side effects.
Dexmedetomidine is a potent, selective α-2 adrenoceptor agonist, with sedative, anxiolytic,
sympatholytic and analgesic effects. It may cause cardiovascular instability; dosage should be
reduced in the presence of hepatic impairment. Dexmedetomidine is currently licensed by
the FDA for use in the ICU for maximum 24 hours at infusion rates < 0.7 microg/kg/hour.
In a double blind, randomised controlled trial, Pandharipande41 compared dexmedetomidine
with lorazepam for sedation of ventilated ICU patients. Patients received a median
dexmedetomidine infusion rate of 0.74 microg/kg/hour (IQR 0.39-1.05 microg/kg/hour). In
this off-licence study the maximum duration of infusion was 120 hours. The dexmedetomidine
group received more fentanyl than the lorazepam group (575 microg/24hours versus
150 microg/24hours). Patients who received dexmedetomidine remained at their target RASS
for longer periods of time and had more days alive without delirium or coma.
CONCLUSION
Delirium is increasingly recognised as a common problem amongst the critically ill. This group
of patients have multiple risk factors for the development of delirium—these are related to
baseline clinical status, the acute illness itself and iatrogenic influences. Delirium is associated
with increased length of stay, increased mortality and increased costs after adjustment for
confounding variables. All critically ill patients should therefore be screened for the presence
of delirium using a validated tool. This process can be readily incorporated into daily practice.
Modifiable risk factors should be identified and controlled; benzodiazepines should be avoided
whenever possible. Current management recommendations include multi-component
environmental modifications and the cautious prescription of anti-psychotic medications.
Internet Resources for Further Reading

Further information, links and teaching resources can be found on the following websites:
www.icudelirium.org
www.icudelirium.co.uk
REFERENCES
1. American Psychiatric Association: Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, text
revision. Washington DC. 2000.
2. Ely EW, Stephens RK, Jackson JC, Thomason JWW, Truman B, Gordon S, et al. Current opinions regarding
the importance, diagnosis and management of delirium in the intensive care unit: A survey of 912 healthcare
professionals Crit Care Med 2004;32:106-12.
3. Ely EW, Gautam S, Margolin R, Francis J, May L, Speroff T, et al. The impact of delirium in the intensive
care unit on hospital length of stay. Intensive Care Med 2001;27:1892-1900.
4. Lin S-M, Liu C-Y, Wang C-H, Lin H-C, Huang C-D, Huang P-Y, et al. The impact of delirium on the
survival of mechanically ventilated patients. Crit Care Med 2004;32:2254-9.
5. Ouimet S, Kavanagh BP, Gottfried SB, Skrobik Y. Incidence, risk factors and consequences of ICU delirium.
Intensive Care Med 2007;33:66-73.
6. Soja SL, Pandharipande PP, Fleming SB, Cotton BA, Miller LR, Weaver SG, et al. Implementation, reliability
testing and compliance monitoring of the Confusion Assessment Method for the Intensive Care Unit in
trauma patients. Intensive Care Med 2008;34:1263-8.
7. Pun BT, Gordon SM, Peterson JF, Shintani AK, Jackson JC, Foss J, Harding SD, et al. Large scale
implementation of sedation and delirium monitoring in the intensive care unit: a report from two medical
centres. Crit Care Med 2005;33:1199-1205.
8. Ouimet S, Riker R, Bergeon N, Cossette M, Kavanagh B, Skrobik Y. Subsyndromal delirium in the ICU:
evidence for a disease spectrum. Intensive Care Med 2007;33:1007-13.
9. Pandharipande P, Cotton BA, Shintani A, Thompson J, Cosabile S, Pun BT, et al. Motoric subtypes of
delirium in mechanically ventilated surgical and trauma intensive care unit patients. Intensive Care Med
2007;33:1726-31.
10. Pun BT, Ely WE. The importance of diagnosing and managing ICU delirium. Chest 2007;132:624-36.
11. Pandharipande P, Shintani A, Peterson J, Pun BT, Wilkinson GR, Dittus RS, et al. Lorazepam is an
independent risk factor for transitioning to delirium in intensive care unit patients. Anesthesiology
2006;104:21-6.
12. Dubois M-J, Bergeron N, Dumont M, Dial S, Skrobik Y. Delirium in an intensive care unit: a study of risk
factors. Intensive Care Med 2001;27:1297-1304.
13. Woods JC, Mion LC, Connor JT, Viray F, Jahan L, Huber C, et al. Severe agitation among ventilated
medical intensive care unit patients: frequency, characteristics and outcomes. Intensive Care Med
2004;20:1066-72.
14. Thomason JWW, Shintani A, Peterson JF, Pun BT, Jackson JC, Ely EW. Intensive care unit delirium is an
independent predictor of longer hospital stay: a prospective analysis of 261 non-ventilated patients. Crit
Care 2005;9:R375-R381.
15. Ely EW, Girard TM, Shintani AK, Jackson JC, Gordon SM, Thomason JWW, et al. Apolipoprotein E4
polymorphism as a genetic predisposition to delirium in critically ill patients. Crit Care Med 2007;35:112-
7.
16. Ely EW, Margolin R, Francis J, May L, Truman B, Dittus R, et al. Evaluation of delirium in critically ill
patients: validation of the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU). Crit
Care Med 2001;29:1370-79.
17. Bergeron N, Dubois M-J, Dumont M, Dial S, Skrobik Y. Intensive Care delirium Screening Checklist:
evaluation of a new screening tool. Intensive Care Med 2001;27:859-64.
18. Ely EW, Shintani A, Truman B, Speroff T, Gordon SM, Harrell FE, et al. Delirium as a predictor of mortality
in mechanically ventilated patients in the intensive care unit. JAMA 2004;291(14): 1753-62.
19. Ely WE, Truman B, Shintani A, Thomason JWW, Wheeler AP, Gordon S, et al. Monitoring sedation status
over time in ICU patients. Reliability and Validity of the Richmond Agitation-Sedation Scale (RASS).
JAMA 2003;289:22 2983-91.
20. Jacobi J, Fraser GL, Coursin DB, Riker RR, Fontaine D, Wittbrodt ET, et al. Clinical Practice Guidelines
for the sustained use of sedatives and analgesics in the critically ill adult. Crit Care Med 2002;30:1 119-
41.
21. Kollef MH, Levy NT, Ahrens TS, Schaiff R, Prentice D, Sherman G. The use of continuous i.v. sedation
is associated with prolongation of mechanical ventilation. Chest 1998;114(2):541-8.
22. Kress JP, Pohlman AS, O’Connor MF, Hall JB. Daily interruption of sedative infusions in critically ill patients
undergoing mechanical ventilation. N Engl J Med 2000;342:1471-7.
23. Girard TD, Kress JP, Fuchs BD, Thomason JWW, Schweickert WD, Pun BT, et al. Efficacy and safety of
a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care
(Awakening and Breathing Controlled trial): a randomised controlled trial. Lancet 2008;371:126-34.
24. Payen JF, Chanques G, Mantz J, Hercule C, Auriant I, Leguillou J-L, et al. Current practices in sedation
and analgesia for mechanically ventilated critically ill patients: a prospective multicenter patient-based
study. Anaesthesiology 2007;106:687-95.
25. Soliman HM, Melot C, Vincent J-L. Sedative and analgesic practice in the intensive care unit: the results
of a European survey. BJA 2001;87(2):186-92.
26. Sessler CN, Gosnell MS, Grap MJ, Brophy GM, O’Neal PV, Keane KA, et al. The Richmond Agitation-
Sedation Scale. Validity and Reliability in Adult Intensive Care Unit Patients. Am J Resp Crit Care Med
2002;166:1338-44.
27. Ely EW, Inouye SK, Bernard GR, Gordon S, Francis J, May L, et al. Delirium in Mechanically Ventilated
Patients. Validity and reliability of the Confusion Assessment Method for the Intensive Care Unit (CAM-
ICU). JAMA 2001;286:2703-10.
28. Lin SM, Liu CY, Wang CH, Lin HC, Huang CD, Huang PY, et al. The impact of delirium on the survival
of mechanically ventilated patients. Crit Care Med 2004;32:2254-9.
29. Bergeron N, Dubois MJ, Dumont M, Dial S, Skrobik Y. Intensive Care delirium Screening Checklist:
evaluation of a new screening tool. Intensive Care Med 2001;27:859-64.
30. Plaschke K, von Haken R, Scholz M, Engelhardt R, Brobeil A, Martin E, et al. Comparison of the confusion
assessment method for the intensive care unit (CAM-ICU) with the Intensive Care Delirium Screening
Checklist (ICDSC) for delirium in critical care patients gives high agreement rate(s). Intensive Care Med
2008;34:431-6.
31. Fraser GL, Riker RR, Prato BS, Wilkins ML. The frequency and cost of patient-initiated device removal
in the ICU. Pharmacotherapy 2001;21(1):1-6.
32. Carrion MI, Ayuao D, Marcos M, Robles M. Accidental removal of endotracheal and nasogastric tubes
and intravascular catheters. Crit Care Med 2000;28(1):63-6.
33. Hopkins RO, Jackson JC. Long-term neurocognitive function after critical illness. Chest 2006;130:869-78.
34. Milbrandt EB, Deppen S, Harrison PL, Shintani AK, Speroff T, Stiles RA, et al. Costs associated with
delirium in mechanically ventilated patients. Crit Care Med 2004;32:955-62.
35. Siddiqi N, Stockdale R, Britton AM, Holmes J. Interventions for preventing delirium in hospitalised patients.
Cochrane database of Systematic Reviews 2007, Issue 2. Art No.: CD005563. DOI: 10.1002/
14651858.CD005563.pub2.
36. Inouye SK, Bogardus ST, Charpentier PA, Leo-Summers L, Acampora D, Holford TR, et al. A multicomponent
intervention to prevent delirium in hospitalised older patients. N Engl J Med 1999;340:669-76.
37. Bogardus ST Jr, Desai MM, Williams CS, Leo-Summers L, Acampora D, Inouye SK. The effects of a
targeted multicomponent delirium intervention on postdischarge outcomes for hospitalised older adults.
Am J Med 2003;114:383-90.
38. Marcantonio ER, Flacker JM, Wright RJ, Resnick NM. Reducing delirium after hip fracture: a randomised
trial. J Am Geriatr Soc 2001;49:516-22.
39. Kalisvaart KJ, de Jonghe JFM, Bogaards MJ, Vreeswijk R, Egberts TCG, Burger BJ, et al. Haloperidol
prophylaxis for elderly hip-surgery patients at risk for delirium: a randomised placebo-controlled study. J
Am Geriatr Soc 2005;53:1658-66.
40. Skrobik YK, Bergeron N, Dumont M, Gottfried SB. Olanzapine vs haloperidol: treating delirium in a critical
care setting. Intensive Care Med 2004;30:444-9.
41. Pandharipande PP, Pun BT, Herr DL, Maze M, Girard TD, Miller RR, et al. Effect of sedation with
dexmedetomidine vs lorazepam on acute brain dysfunction in mechanically ventilated patients. The MENDS
randomised controlled trial. JAMA 2007;298(22):2644-53.
Fourteen
Cervical Spine Clearance in

Obtunded Multi-trauma Patients
Peter T Clark
INTRODUCTION
The standard pre-hospital practice for any patient who has suffered injury or trauma, however
minor, is to have full spinal immobilisation at the scene of the incident. Properly applied, this
requires a rigid collar and a spinal board with sandbags and tape to keep the head completely
still and the spinal column ‘in-line.’ In other words, all patients are assumed to have a neck
injury until proven otherwise. In the vast majority, no spinal injury will be identified by clinical
examination and standard imaging in those who are awake and able to comply with instructions
or report pain. However, among those who are unconscious, spinal immobilisation will be
continued until the spine is ‘cleared’. Unfortunately, an optimal protocol for clearing the spine
in these patients remains controversial and poorly understood even now.
EPIDEMIOLOGY
Injuries to the cervical spine (C-spine) occur in 2.0-6.6% of blunt trauma patients.1-5 With the
co-existence of head injury, the incidence of cervical spine injury increases to about 10%.6
Among those with concurrent head injury, factors making injury more likely are focal
neurological deficit and a GCS of < 8.
Cervical spinal cord injury is a devastating event for both the patient and the family.
Financial costs of quadriplegic individuals runs into millions for a lifetime of medical care and
lost opportunity. Not all patients with a bony injury have a cord injury. On the other hand,
injury to the spinal cord in the absence of bony fractures has been reported to occur in 0.07-
0.7% of all trauma admissions.5
Missed or delayed diagnosis of cervical spine injury increases the incidence of secondary
neurological deficit by 10 times compared to those in whom a correct diagnosis is made
initially. Looked at another way, nearly 70% of those who have a missed diagnosis have an
associated alteration in sensorium.
Cervical Spine Clearance in Obtunded Multi-trauma Patients 147
ASSESSMENT OF C-SPINE IN Table 14.1: Complications associated with

prolonged spinal immobilisation
UNCONSCIOUS PATIENTS
• Increased intracranial pressure
Since a missed diagnosis has serious physical, • Difficult laryngoscopy and intubation
social and economic consequences, the primary • Difficult CVC insertion
objective of any cervical spine protocol is to • Increased risk of pulmonary embolism
• Poor oral care
promptly recognize any significant injury. A delay
• Pressure necrosis
in diagnosis increases the risk of neurological • Infections
deficit by 10 fold and is as serious as a missed • Increased nursing time and inputs
diagnosis of cervical spine injury. As a corollary
to this, a secondary but equally important aim is to “clear” the cervical spine in the remainder
of patients because prolonged and unnecessary spinal immobilisation is associated with
significant risks (Table 14.1).
Since the vast majority of trauma victims do not have a cervical spine injury, a spine
clearance protocol requires a high degree of specificity in order to make it practically applicable.
Unfortunately, most screening programmes have a high sensitivity that results in a large number
of false positives in whom, spinal immobilisation is invariably continued. Prolonged
immobilisation has significant attributable morbidity that becomes apparent after 48 hours of
use of a cervical collar. Extended use of a cervical collar can result in decubitus ulceration in
nearly one-third of unconscious trauma patients.7 There is evidence, in addition that rigid
collars do not work effectively especially in patients with unstable cervical injuries.8,9 Thus,
cervical clearance protocols must balance the small, but significant risk of a missed injury
against the potential for harm from ongoing C-spine immobilisation. A missed cervical injury
is easily recorded and tends to be remembered and publicised. As a result, a conservative
approach is followed more often than is needed.
Evaluation and management of an unconscious trauma patient with a potential cervical
injury is a contentious and controversial issue in large volume trauma centres. Randomised
controlled trials have not been conducted and are unlikely to be ethically appropriate; hence
observational studies have guided clinical practice so far. This chapter will review all the
variables involved in decision making to allow for evidence-based practice of cervical clearance
in obtunded or unconscious trauma patients. For radiological investigation of a conscious,
trauma patient, the reader is referred to the Table 14.2: EAST clinical criteria
NEXUS10 and Canadian C-spine rule.11
1. GCS 15 with patient alert and oriented
2. No intoxicants or drugs consumed
CLEARING THE CERVICAL SPINE 3. No significant distracting injuries
4. No signs or symptoms on clinical examination
The exclusion of spinal injury can be done No midline tenderness of pain
by clinical examination in combination with Full range of active movement
focussed radiological investigation. Clinical No referable neurological deficit
evaluation of the cervical spine assesses The reliability and performance of these criteria requires
four parameters (Table 14.2). These criteria judgement and strict application by the clinician
have been incorporated in the ATLS and Eastern Association for Surgery of Trauma (EAST)
guidelines12 and are supported by more than 20 studies.13 If these criteria cannot be met,
radiological investigation is indicated in addition to an appraisal of the mechanism of trauma
and associated injuries. In a prospective evaluation of 1000 blunt trauma victims, all patients
with cervical spine injury failed at least one of the preconditions set out by the EAST
guidelines.14
In unconscious patients or those who are on heavy sedation for their head injury, the
clinical signs are obscured; hence clinicians have to rely on imaging procedures to exclude
spinal injury. The safety of cessation of spinal precautions instituted without definitive
knowledge of disco-ligamentous injury is hotly debated.
RADIOLOGICAL IMAGING
Plain X ray
An anatomically and technically adequate film will visualize the cervical spine from the cranio-
cervical junction to the cervico-thoracic junction. The sensitivity of a lateral cross table view,
when interpreted by an expert is as high as 74-90%.15,16 Unfortunately, the rate of adequate
lateral views in a multi-trauma patient can be as low as 50%.13 Tracheal intubation makes
adequate visualization even more difficult. Thus, it is recognized that although under ideal
circumstances, a lateral view will miss only 15% of injuries, an additional 20% of missed
injuries result from misinterpretation of suboptimal radiographs. It has been shown that in up
to 50% of cases, the cervico-thoracic junction where up to 60% of injuries occur, could not
be seen even with arm traction.17
The lateral cervical spine x-ray occupies a prestigious position within the ATLS guidelines,18
however, this film alone does not provide a reliable enough result for the exclusion of cervical
injury and may, thus promote unsafe practice in the emergency department. As a result, the
three view cervical trauma series has been incorporated into the ATLS and EAST guidelines
with the anticipation that an additional open mouth view (for the odontoid) and an AP neck
x ray will be able to decrease missed injury rate below 15%. Even with an improvement in
sensitivity to over 90% the recommended ATLS guidelines will miss 1% of injuries when
applied in a population of unconscious trauma victims (expected injury rate ~10%). The
addition of two extra oblique views (or swimmer’s views) to the three-view trauma series
improves visualization of the cervico-thoracic junction and appears to confer some cost
advantage compared to CT scanning, but it does not improve the sensitivity to the extent
required to rule out a cervical injury.
CT Scanning
The introduction of helical CT, multi-detector row scanners and 3-D reconstruction has led to
greatly improved quality and accurate visualization of injuries. While concerns over excess
exposure to radiation have been raised, the actual amount of radiation during a neck CT
scan is only a small fraction of what is necessary for a CT brain.
Helical CT screening is now carried out using 1 mm slices with 3D reconstruction instead
of the standard 3 or 5 mm cuts. Such detailed scanning allows for the diagnosis of an unstable
injury and even detects a ligamentous injury with image quality approaching the standard of
MRI scanning.19 Even so, CT scanning and plain films should be considered complementary
diagnostic techniques.
MRI Scanning
MRI is widely recognized as the gold standard for imaging soft tissues and is a valuable tool
for assessment of disco-ligamentous instability. The EAST guidelines12 suggest an evaluation
with MRI in all patients with a neurological deficit. Some have argued for the inclusion of MRI
as the sole investigation for assessment of spinal injuries in all obtunded trauma patients20.
However, this approach is problematic for many reasons. MRI is expensive, has limited
availability and is not suitable for patients with metallic objects like traction equipment, pelvic
fixateurs and monitoring equipment.7 These impediments can cause significant delays in spinal
clearance. In addition, moving critically ill trauma patients to the MRI suite and monitoring
them during the procedure is associated with potential risk of adverse events.21 Moreover, by
detecting minor injuries in addition to major ones, MRI has the potential to distract the clinician
towards inappropriate management of injuries that are clinically insignificant.22
Osseous and soft-tissue structures contribute to cervical spine stability.23,24 MR imaging
routinely demonstrates nonalignment injuries such as epidural haematoma, disc herniation,
non-ligament injuries and spinal cord or spinal root injuries. By applying Denis’ three-column
theory of spinal stability,25 MR imaging can be used to identify unstable injuries that require
intervention.26 The data on usefulness of MRI in obtunded, blunt trauma victims is limited27
although much more has been published on MR imaging in a more general population of
trauma patients.27 The recurring suggestion from these studies in a non-select population is
that soft tissue abnormalities are seen in MRI scan of up to 25% of all trauma patients.27 Not
all are unstable and not all require intervention.
Far from clearing the spine, routine use of MRI would run a very real risk of finding an
abnormality in nearly 25% of patients, who will then have a cervical collar left in place for
weeks. In the study by Benzel et al28 only 2 out of 35 patients having an abnormality on
MRI required surgical fusion and both injuries were detected on the CT images as well. One
interpretation is that both MRI and CT perform equally well in detecting unstable ligamentous
injuries that require surgery. Horn et al22 found in predominantly conscious patients that
MRI identified no instability which was not evident on CT scanning or dynamic flexion /
extension studies. Conclusions of this study are difficult to apply to a population of uncon-
scious patients. Unfortunately, no direct comparison of different modalities of imaging has
been published so far.
In actual practice, purely ligamentous injuries without fracture are uncommon relative to
the overall number of blunt trauma injuries. Demetriades et al2 reported that only 0.2% of
14,755 patients admitted to their facility had subluxation without fracture. Chiu and colleagues29
reported a slightly higher rate of pure ligamentous injury – 0.54% within a large population
of blunt trauma victims. On the basis of a low rate of unstable soft tissue injuries and notwith-
standing the obvious advantages of MRI scanning, Davis et al30 have suggested that a normal
three view radiographs along with thin slice CT is adequate to clear the cervical spine in
obtunded multi-trauma patients.
Dynamic Fluoroscopy
In 1998, the Eastern Association for the Surgery of Trauma put forth a level II recommendation
for cervical spine clearance in the obtunded patient.12 Specifically, patients not fit for clinical
clearance of cervical spine were recommended to undergo a three-view series of plain radio-
graphs, including antero-posterior, lateral and open mouth view and a thin slice CT imaging
of C1-C2. After conducting a survey of adherence to its published guidelines31 and following
a meta-analysis, the EAST group revised their recommendations in 200012 to include flexion/
extension lateral cervical spine fluoroscopy in order to clear the c-spine. This has remained
a contentious issue ever since, as the evidence supporting routine use of dynamic fluoroscopy
is very limited.
Dynamic fluoroscopy involves passively manipulating the neck under real time (lateral
view) imaging to demonstrate instability. It is a pragmatic approach – if the spine is unstable,
then it will be demonstrated upon stressing it during movement. Dynamic fluoroscopy must
be distinguished from static maximal flexion /extension views that are inherently unsafe in
unconscious patients. There are also legitimate concerns about manipulation of the cervical
spine in obtunded patients who may be also on sedation and muscle paralysis while undergoing
dynamic studies. In one study, a patient reportedly developed tetraplegia after undergoing
dynamic fluoroscopy.30 Fluoroscopy is therefore potentially unsafe despite strict adherence to
protocol.
However, fluoroscopy is less expensive and more widely available than MRI. Fluoroscopy
is expected to recognize a small subset of patients with purely ligamentous injury, although
the overall yield in an unselected population of trauma patients may be very low (<0.9% of
all evaluations). It is likely that true positive rate is close to the false positive rate, making the
positive predictive value of an abnormal test rather low. In the largest published trial so far,
only 2 of 301 patients (0.66%) were found to have partial ligament injuries.30 Contrast CT
had already revealed one injury, while the other was outside the range of a directed CT scan.
Both were ‘relatively stable’ and were managed conservatively. The number of examinations
required to detect an abnormality not identified on CT or plain films (NNT) has been estimated
to range from 300 to 500. In addition to this low yield, there is still a problem of inadequate
visualization of cervico-thoracic and cranio-cervical junction. This is not dissimilar to problems
with lateral x-ray of the neck for trauma patients. False negative rates are unknown but non-
visualisation rates are estimated to be as high as 30%.32
Dynamic fluoroscopy thus lacks sufficient evidence of sensitivity, specificity, safety and
simplicity to recommend it for routine use in unconscious trauma patients. Unfortunately, the
2000 update of EAST guidelines, continue to endorse it as a level II-III recommendation.
WHEN TO REMOVE THE COLLAR

This is really the crux of the topic under review. In recent studies, the majority of “critical
errors” have occurred as a result of protocol violations and not because of protocol failures.33
Therefore, adopting and adhering to locally endorsed guidelines, which are acceptable to all
relevant specialties is the best option. A protocol thus developed, needs review on the basis
of more current information. At the core of it, such a protocol should contain definite and
unambiguous directions on how to proceed, what imaging procedures to include and what
to order under what circumstances.
Based on the imaging report from an experienced radiologist and clinical history, the ICU
team in conjunction with the trauma services can agree on the likely risk of cervical spine
injury. If this is low or negligible, the collar can be removed within 24 hours, which should
allow enough time for ICU staff to also assess the likelihood of recovery from the associated
brain injury. A flow chart for clearing the c-spine in an unconscious patient such as the one
in Figure 14.1 can be developed.13
Overall, a consensus approach between responsible consultants is important and, if any
doubt exists, spinal immobilisation should continue. In addition, there needs to be widespread
acceptance of the fact that no single investigation or screening procedure will have 100%
sensitivity and 100% specificity. It is a tragic reality that cervical spine injuries will be missed
in the future. We hope that with objective quantification of risks and through a rigorous process
of clinical evaluation, the false negative rate will diminish to a level as close to 0% as possible.
CONCLUSION
Clinicians remain at the forefront of making decisions about clearing the c-spine in a trauma
victim. While making a decision, they must be aware of the devastating implications of a
missed injury but also of the fact, that they alone bear the responsibility for any consequences,
which follow. The most effective and expedient way of detecting spinal injuries is a matter of
some debate. What is clear-cut is that no single investigation will either detect all c-spine
injuries or completely exclude them. Therefore, a decision based on a combination of tests,
imaging procedures and other investigations is needed. In the last 10 years, CT technology
has improved to a point that renders other modalities of imaging either superfluous (plain x-
ray or dynamic fluoroscopy) or initially unwarranted (MRI). Dynamic flexion-extension
fluoroscopy has poor sensitivity, high rate of image inadequacy and a low cost to benefit
ratio. It should no longer be part of any protocol.
152
Fig. 14.1: Algorithm for clearance of cervical spine in obtunded blunt trauma patients
On the basis of the evidence reviewed, we suggest that all obtunded trauma patients should
have a helical CT of the c-spine with 1 mm cuts and 3-D reconstruction along with plain
x-rays. If these images are reported as normal by an experienced radiologist, then the risk of
a significant missed injury is exceedingly small. MRI images are superior to CT in the detection
of disco-ligamentous injuries. However, use of MRI can only be suggested for selected trauma
patients, who are either considered high risk for spine injury or those, who have a definitive
abnormality requiring further assessment. There are no prospective studies comparing multi-
slice CT and MRI for the evaluation of occult cervical spine injuries in unconscious patients.
Even so, an evidence-based protocol for clearance of cervical spine must exists in all centers
that receive and treat trauma patients.
Internet Resources for Further Reading

Further information, links and resources can be found on the following websites
www.east.org
www.trauma.org
REFERENCES
1. Ajani AE, Cooper DJ, Scheinkenstel CD, et al. Optimal assessment of cervical trauma in critically ill patients:
a prospective evaluation. Anaesth Intensive Care 1998;26:487-91.
2. Demetriades D, Chalambides BS, Chahwan S, et al. Non-skeletal cervical spine injuries: Epdiemiology
and diagnostic pitfalls. J Trauma 2000;48:724-7.
3. Schenarts PJ, Diaz J, Kaiser C, et al. Prospective comparison of admission computed tomographic scan
and plain films of upper cervical spine in trauma patients with altered mental state. J Trauma 2001;51:663-
9.
4. Pasquale M, Fabian TC. Practice management guidelines for trauma: EAST ad hoc committee on guidelines
development – identifying cervical spine instability after trauma. J Trauma 1998;44:941-56.
5. Hendy GW, Wolfson AB, Mowere WR, et al. Spinal cord injury without radiographic abnormality; Results
of the National Emergency X-Radiography utilisation study in blunt trauma. J Trauma 2002;53:1-4.
6. Harrison P, Cairns C. Clearing the cervical spine in the unconscious patient. Cont Ed Anaesth Crit Care
Pain 2008;8:117-20.
7. Cooper DJ, Ackland HM. Clearing the cervical spine in unconscious head injured patients – the evidence.
Crit Care Resuscit 2005;7:181-4.
8. Rosen PB, McSwain NE, Arata M, Stahl S, Mercer D. Comparison of 2 new immobilisation collars. Ann
Emerg Med 1992;21:1189-95.
9. Hughes SJ. How effective is the Newport / Aspen collar? A prospective radiographic evaluation in healthy
adults. J Trauma 1998;45:374-8.
10. www.acutecare.com/nexuscspine.htm Accessed 3 Jan 2009.
11. Steill IG, Clement CM, McKnight RD, et al. The Canadian C-spine rule versus the NEXUS low-risk criteria
in patients with trauma. N Engl J Med 2003;349:2510-8.
12. www.east.org. Practice guidelines for identifying cervical spine injury following trauma. Accessed 3 Jan
2009.
13. Morris CGT, McCoy E. Clearing the cervical spine in unconscious polytrauma victims; balancing risks and
effective screening. Anaesthesia 2004;59:464-82.
14. Zabel DD, Tinkoff G, Witenborn W, Ballard K, Fulda G. Adequacy and efficicay of lateral cervical spine
radiography in alert, high-risk blunt trauma patients. J Trauma 1997;43:952-6.
15. Streitweiser DR, Knopp R, Wales LR, Williams JL, Tonnemaker K. Accuracy of standard radiographic
views in detrmining cervical spine fractures. Ann Emerg Med 1983;12:538-42.
16. Blahd WH, Iserson KV, Bjelland JC. Efficacy of the post-traumatic cross table lateral view of the cervical
spine. J Emerg Med 1984;2:243-9.
17. Nichols C, Young D, Schiller W. Evaluation of cervico-thoracic junction injury. Ann Emerg Med 1987;16:640-
2.
18. www.facs.org/trauma/atls/information.htm. Accessed 3 Jan 2009.
19. Hogan GJ, Mirvis SE, Shanmuganathan K, Scalea TM. Exclusion of unstable cervical spine injury in
obtunded patients with blunt trauma: is MR imaging needed when multi-detector row CT findings are
normal. Radiology 2005;237:106-13.
20. Katzberg RW, Benedetti PF, Drake CM, et al. Acute cervical spine injuries: Prospective MR imaging
assessment at a level I trauma centre. Radiology 1999;213:203-12.
21. Ackland HM, Cooper DJ, Malham GM, Stuckey SL. Magnetic resonance imaging for clearing the cervical
spine in unconscious intensive care trauma patients. J Trauma Injury Crit Care 2006;60:668-73.
22. Horn EM, Lekovic GP, Feiz-Erfan I, et al. Cervical magnetic resonance imaging abnormalities not predictive
of cervical sine instability in traumatically injured patients. J Neurosurg Spine 2004;1:39-42.
23. Warner J, Shamuganathan K, Mirvis SE, Cerva D. Magnetic resonance imaging of ligamentous injury of
the cervical spine. Emerg Radiol 1996;3:9-15.
24. Saifuddin A. MRI of acute spinal trauma. Skeletal Radiol 2001;30:237-46.
25. Denis F. Spinal instability as defined by the three-column spine concept in acute spinal trauma. Clin Orthop
1984;189:65-76.
26. Keiper MD, Zimmerman RA, Bilanuik LT. MRI in the assessment of the supportive soft tissues of the
cervical spine in acute trauma in children. Neuroradiology 1998;40:359-63.
27. Sliker CW, Mirvis SE, Shamuganatahn K. Assessing cervical spine stability in obtunded blunt trauma patients:
Review of medical literature. Radiology 2005;234:733-9.
28. Benzel EC, Hart BL, Ball PA, Baldwin NG, Orrison WW, Espinosa MC. Magnetci resonance imaging for
the evaluation of patients with occult cervical spine injury. J Neurosurg 1996;85:824-9.
29. Chiu WC, Haan JM, Cushing bm, Kramer ME, Scalea TM. Ligamentous injuries of the cervical spine in
unreliable blunt trauma patients: incidence, evaluation and outcome. J Trauma 2001;50:457-64.
30. Davis JW, Kaups KL, Cunningham MA, et al. Routine evaluation of the cervical spine in head injured
patients with dynamic fluoroscopy: a reappraisal. J Trauma 2001;50:1044-7.
31. Marion D, Domeier R, Dunham CM, et al. Determination of cervical spine stability in trauma patients.
EAST 2000;1-6.
32. Sees DW, Rodriguez Cruz LR, Flaherty SF, et al. The use of bedside fluoroscopy to evaluate the cervical
spine in obtunded trauma patients. J Trauma 1998;45:768-71.
33. Stassen NA, Williams VA, Gestring ML, Cheng JD, Bankey PE. Magnetic resonance imaging in combination
with helical computer tomography provides a safe and efficient method of cervical spine clearance in the
obtunded patients. J Trauma 2006;60:171-7.
Dysautonomia Following Traumatic Brain Injury 155
Fifteen
Dysautonomia following
Traumatic Brain Injury
Ian Baguley
INTRODUCTION
Following trauma, the autonomic nervous system (ANS) normally coordinates a predictable
response with features such as tachycardia, hypertension and increased respiratory rate.
Following traumatic brain injury (TBI), elevated autonomic parameters are extremely common,
occurring in 621 to 92%2 of patients admitted to the intensive care unit (ICU). In most situations,
the autonomic response is proportional to the originating stimulus, however, in certain situations
the ANS response becomes excessive with the potential to produce additional morbidity. Such
a condition, termed Dysautonomia, occurs in a subgroup of survivors of severe acquired
brain injuries. In this context, Dysautonomia identifies a syndrome of simultaneous and
paroxysmal overactivity of the sympathetic and motor nervous systems.3 The paroxysmal
sympathetic changes occur over a range of severities but include documented heart rates up
to 190 beats per minute, respiratory rates of 60 breaths per minute, core temperatures to
42°C, arterial blood pressures of 170/120 mmHg and sweating.3 These ANS signs are
accompanied by assorted forms of motor overactivity such as decerebrate or decorticate
posturing, dystonias, rigidity and spasticity.
EPIDEMIOLOGY OF DYSAUTONOMIA
Dysautonomia following moderate and severe TBI has been estimated to occur in between
8 and 33% of patients.4 5 studies have estimates around the lower end of this range (i.e.,
8–14.1%),2,5-8 while 2 studies suggest a much higher (32-33%) incidence.9,10 Of these studies,
three in particular suggest a reason for the marked variability in incidence estimates. In two
of these studies, short duration sympathetic overactivity (ie, of less than 2 weeks) was found
in 24%2 and 33%10 of TBI subjects. Conversely, prolonged Dysautonomia was seen in 32%
of TBI subjects who were in a vegetative state.9 Combining these findings suggests that
Dysautonomia is a spectral disorder, with the observed incidence in any given sample being
related to the relative injury severity of the sample and the time post injury that diagnostic
criteria were applied.
While TBI accounts for most published cases, many acute neurological disorders that have
been reported to precipitate Dysautonomia. These include severe hypoxic brain injury,11-19
stroke (haemorrhagic stroke in particular),16,18-22 tumours,14,21,23-27 and acute hydroce-
phalus.20,21,28,29 There are also isolated case reports of uncommon aetiologies, such as
Dysautonomia following extracranial ascending aortic artery dissection 30 and hypoglycaemic
coma.31
In addition, it is worth noting that Dysautonomia bears a striking resemblance to a range
of other conditions, some of which are considered to be medical emergencies. These “overlap”
syndromes32 include Neuroleptic Malignant Syndrome, Malignant Hyperthermia, Parkinsonian-
Hyperpyrexia Syndrome, 33,34 Serotonin Syndrome, 35 acute intrathecal baclofen
withdrawal.36,37 Finally, similar clinical features can result from exposure to certain animal
venoms (such as the jellyfish related Irukandji Syndrome38), chemicals (such as gamma
butyrate39) and infectious agents such as tetanus and rabies.
PATHOPHYSIOLOGY
Recent data would suggest that Dysautonomia is the result of changes that occur at the level
of both the brain and spinal cord. The earliest theories proposed an epileptogenic aetiology,40
however, multiple attempts to either identify or treat epilepsy in Dysautonomic patients have
produced negative results.13,14,21,29,41-43 The limited autopsy and pathophysiological data
suggest the syndrome results from a relative disconnection of pathways at or around the level
of the midbrain (reviewed in44). Most disconnection theories suggest that paroxysms are driven
by upper brainstem and diencephalon lesions, with clinical features arising from either excitatory
or inhibitory centres in these areas.
However, it is becoming evident that an important component of Dysautonomia is an
abnormality of afferent stimulus processing. Since early reports,45 clinicians have observed
that Dysautonomic paroxysms may be provoked by stimuli such as pain, endotracheal
suctioning, passive movements (e.g., turning, bathing and muscle stretching),3,17,22,29,46-52
constipation or urinary retention46,53 and emotional29 or environmental54 stimuli. Support for
these anecdotal observations has recently been provided in empirical research. In one study,
the sympathetic response to endotracheal suctioning was examined in subjects on Day 7
post-TBI.55 On Day 14, subjects were grouped according to the presence or absence of
continuing paroxysmal sympathetic activity. Reviewing Day 7 heart rate data revealed a
differentiated response pattern between the Dysautonomic and non-dysautonomic groups,
suggesting that Dysautonomic subjects had developed excessive sympathetic responsivity to
afferent stimuli by Day 7 post injury.
This new evidence highlights the previously noted physiological similarities of Dysautonomia
and Autonomic Dysreflexia46,50 (that is, sympathetic over-responsivity to stimuli resulting from
the loss of supraspinal inputs). In an attempt to produce a theoretical framework, the Excitatory/
Inhibitory Ratio (EIR) Model,32 has been put forward to combine these syndromes under a
common pathophysiological process. This integrative model suggests that the causative
brainstem centres are inhibitory in nature, and that damage to these structures predisposes
the spinal cord to overreact to afferent stimuli in a manner akin to that which produces allodynia
(where non-painful stimuli become perceived as nociceptive).56
CLINICAL RELEVANCE OF DYSAUTONOMIA

Early identification of Dysautonomia has been suggested to be minimize unnecessary diagnostic
testing, 47 promote timely and appropriate intervention, 52 and minimise secondary
morbidity.3,57,58 The major causes of potentially preventable morbidity have been suggested
to include prolonged hyperthermia59 (a recognised negative prognostic indicator60-62), cardiac
damage,63,64 and high catecholamine levels.65 In addition, posturing subjects show increased
metabolic rates.66,67 Combined with the prolonged abnormalities of gastrointestinal tract
function that follow severe TBI,68 a highly catabolic state results,69 causing an estimated 25%
decrease in body weight in the first weeks post injury.2 Affected individuals have also been
found to be significantly more likely to develop heterotopic ossification.6 These factors combine
to place the individual at risk of developing critical illness or compression neuropathies.
Dysautonomic subjects commonly show reduced capacity for voluntary movement,
particularly during paroxysms, and may have under-managed spasticity and dystonia.3,51,70
To this end, ‘locked-in’ syndromes have been identified in Dysautonomic subjects,2,15,27 a
situation that can result in under-managed pain or a misdiagnosis of persistent vegetative
state.
In injury severity matched cohorts, long duration Dysautonomia has been shown to be an
independent predictor of worse outcome following severe TBI.3 This is not the case with the
more common short duration sympathetic hyperactivity seen in 24-33% of subjects. For this
group, limited data suggests that the sympathetic features do not produce an adverse effect
on outcome.55 In contrast, one third of this group (that is, 8-14.1% of all severe TBI survivors)
develop prolonged Dysautonomia and have significantly worse outcomes,3,8,9 prolonged
swallowing abnormalities, longer coma8 or post-traumatic amnesia3 duration, longer hospital
admissions2,3 and greater overall health care costs.2
This latter group forms approximately 5% of rehabilitation admissions2 and their clinical
presentation follows a common three phase pattern. In the first phase, Dysautonomic and
non-dysautonomic patients are not identifiable on physiological variables for around the first
7 days post injury.2,3,71 While it has been argued that this is due to the effects of regular
sedation, this has yet to be confirmed in prospective research. Thereafter, frequent, prolonged
and intense episodes of sympathetic overactivity become evident, usually coupled with
increasing regional muscle tone. At this time, Dysautonomic patients usually have consistently
raised HR and temperature, with paroxysmal posturing and autonomic overactivity
superimposed on these elevated baseline levels. The pattern of posturing is most often
asymmetrical and may not fit into classical decorticate or decerebrate postures.72
Fig. 15.1: Heart rate in an untreated dysautonomic subject.

Note: Heart rates are averaged minutely values derived from a 24 hour Holter monitor recording
With increasing time post injury, paroxysms decrease in duration, frequency and magnitude.
The reduction in severity of paroxysms usually coincides with observable improvement in
neurological status,3,14 although whether this is causative or an epiphenomena has not been
identified. As subjects improve, resting blood pressure, respiratory rate, heart rate and
temperature trend towards normal values. The impact that Dysautonomia has on heart rate
can be seen in Figure 15.1. This 24 hour heart rate data is generally representative of
Dysautonomic subjects and comes from an individual 28 days post TBI. Mean heart rate for
the entire day was 129 beats per minute (bpm), ranging from 67 to a maximum of 182. As
can be seen, the subject had multiple paroxysmal episodes over the 24 hours, each lasting
between 2 and 6 hours. While the subject had several episodes with heart rates below 80bpm
(amounting to 11% of the day), heart rates above 140bpm totalled 39% of the day.
As recovery continues, posturing patterns may change, revealing an underlying tetraplegia
or other focal neurological deficit. Most subjects are left with some permanent degree of dystonia
and spasticity.2 Sweating patterns change from whole body to upper trunk, head and neck,
before ceasing entirely.12 Three studies have reported this feature longitudinally. In two of
these, paroxysmal diaphoresis resolved a mean of 2.5 months post injury in 35 subjects3 and
5.9 months post-injury in 87 subjects.9 In another study, 22 of 31 subject’s features had
settled by 6 months post injury.8 While the overt features of sympathetic overactivity settle
over time, the tendency to display paroxysmal tachycardias remains evident for at least 14
months post injury.58
MANAGEMENT
Despite the impressive nature of the symptomatology and the largely circumstantial evidence
suggesting that early intervention is important, there is surprisingly little literature available to
guide management.4,50 As the diagnosis of Dysautonomia is one of exclusion, other primary
disease processes such as acute hydrocephalus, sepsis and the autonomic emergency
syndromes need to be considered and treated. A careful examination for evidence of triggering55
and treatment of potential noxious stimuli (such as undiagnosed fractures, heterotopic
ossification, pressure areas, painful spasticity or dystonias, infection, etc) has the potential to
decrease the frequency or severity of paroxysms.50,54 In this context, one author has suggested
pre-treatment of patients prior to nociceptive procedures being undertaken to reduce ‘triggering’
of paroxysms.54
In spite of these possible interventions, conservative measures alone are rarely sufficient
to control the paroxysmal sympathetic overactivity of severe Dysautonomia. Although a large
number of medications have been used to treat Dysautonomia, information regarding their
efficacy is largely restricted to anecdotal reports. These individual studies have been reviewed
elsewhere.4,50,71,73 Accepting the limitations of current literature, a range of medications from
many classes of neurologically active agents have been proposed to be beneficial. In particular,
these medications include alpha and beta antagonists, dopamine agonists (and antagonists),
GABA A and B agonists, opioids and gabapentin.
The intervention that appears to be the most effective is intrathecal baclofen (ITB). Although
anecdotal in nature, all reported cases have shown a cessation or large reduction in paroxysmal
sympathetic overactivity.16,18,20,49,52,74,75 However, deciding the threshold for implementing
ITB can be difficult. Many subjects will exhibit paroxysmal overactivity for only a few
months,3,6,8 the procedure is invasive and has relatively high complication rates (20-50%76).
On the other hand, ITB is also effective for managing spasticity77,78 and this may be
advantageous in Dysautonomic patients in the long term. Intravenous morphine and
midazolam are also effective but their sedative effects limit their usefulness in the latter stages
of an ICU admission or during rehabilitation.50
Gabapentin has recently been reported to reduce the number and severity of dysautonomic
paroxysms.31,52 Furthermore, gabapentin allowed an overall reduction in other medications,
including ITB, without a recurrence of symptoms.52 However, it has been suggested that the
drug should be trialled at a low starting dose (100 mg) to minimise the risk of marked
bradypnoea or sedation, presumed to result from a sudden loss of sympathetic drive.43
Drugs with sympathetic activity have also been commonly used in Dysautonomia. Of the
beta-blockers, propanolol has the largest literature base, with mixed reports between those
where it has been useful14,26,46,75,79-82 or been unable to control paroxysms.21,27,54,74 Recent
research suggests that beta-blockers improve survival post TBI83,84 and propanolol has long
been known to decrease post-TBI catecholamine levels.85,86 Labetalol has little literature
available, evenly split between beneficial42 and unhelpful. 22 Clonidine also reduces
catecholamine levels,87,88 but has been reported to be useful in less than half of the cases in
the literature.4 It has been suggested that clonidine may control Dysautonomic crises but
requires high doses (for example, 800 mcg 2nd hourly).89
Several other pharmacological agents show a mixed picture in terms of efficacy. The number
of cases where dopamine agonists (predominantly bromocriptine) has been at least partially
effective12,21,30,47,48,53,75 is balanced against those where no effect was observed.14,27,52,54

Benzodiazepines also present a mixed picture, with anecdotal reports favouring diazepam
and midazolam.14,53 While chlorpromazine has been reported to be effective in reducing some
components of Dysautonomia,27,50 dopamine antagonists are generally considered to be contra-
indicated in TBI.27,50,90
Drugs that have been reported to be ineffective include methadone,14,21 metoprolol,16,42,52,54
oral baclofen14,16,18,74 and phenytoin.21,28,47 Oral dantrolene has been reported to be unhelpful
in all but one patient.14,16-18,21
It is readily apparent that there is insufficient available literature to provide a definitive
protocol to guide management. However, in terms of minimising the possibility of adverse
drug reactions, it would appear appropriate to utilise morphine/midazolam, gabapentin and/
or propanolol in the first instance. There are occasional reports of excellent efficacy with
bromocriptine, suggesting it may have a secondary role. ITB is most useful in cases that are
resistant to oral medication, are of greater severity or longer duration.
CONTROVERSIES IN DYSAUTONOMIA
There has been increasing research interest in Dysautonomia over the last 5 to 10 years, with
a number of case control series being published. However, there remains considerable disparity
between the estimated 8-33% incidence of Dysautonomia following severe TBI and the paucity
of scientific literature. One possible explanation for this discrepancy could be a lack of clinical
awareness of the condition. This would appear to explain the large and ever increasing number
of names used for the condition. In a recent review, 16 different names were used to define
the syndrome including autonomic dysfunction syndrome, acute midbrain syndrome,
paroxysmal sympathetic storms, hypothalamic-midbrain dysregulation syndrome, paroxysmal
autonomic instability with dystonia and so on.4 Another explanation for the lack of literature
is that Dysautonomia is a diagnosis of exclusion and can be mistaken for, or complicated by,
problems such as opiate withdrawal, epileptic seizures and sepsis, as well as rarer conditions
such as the aforementioned neuroleptic malignant syndrome, malignant hyperthermia and
others.44
Another controversy concerns diagnostic criteria for Dysautonomia. As noted earlier,
sympathetic arousal is an extremely common feature during the early post-acute TBI recovery
period and it is not clear where such hyperactivity should become classified as Dysautonomia.
There are currently 5 sets of diagnostic criteria3,5,6,10,71 that appear to be minor variations
based around similar constructs. More recently, the concept of ‘triggering’ of paroxysms
following minor stimuli has been proposed as a clinical sign that may help to differentiate
Dysautonomic from non-dysautonomic individuals.55 While moves to critically appraise these
criteria have begun,57,91 there is insufficient research available to be able to establish consensus
criteria at this time.
Linked in with the issue of the lack of diagnostic criteria, there has also been a lack of
quantitative mechanisms to investigate the pharmacological management of dysautonomia.
Dysautonomic subjects tend to show marked day to day variability in their presentation (as
illustrated in Figure 15.1) and an appropriate research tool would need to accurately measure
the frequency, severity and duration of paroxysms over time. Recently, a number of tools
have been proposed to overcome this limitation including event related heart rate changes
and heart rate variability paradigms over short periods55 or over a 24 hour time scale.31
Research techniques such as these are required to raise awareness of the syndrome and to
properly evaluate treatment efficacy.
CONCLUSION
Research shows that dysautonomia is a distinct but under-recognised clinical syndrome that
can result from numerous forms of acquired brain injury. While short duration sympathetic
hyperactivity does not appear to be associated with worse outcome, clinical research shows
that prolonged dysautonomia places a considerable burden on both patient and health care
services. There is potential for reducing this burden through timely recognition and intervention,
however, the field is hampered by a lack of clinical awareness, a poor understanding of
pathophysiology and anecdotal management protocols. Investigatory techniques are becoming
available that will allow the development of evidence based treatment paradigms for the first
time. However, due to the relatively low incidence of the condition, multicentre research using
standardised nomenclature and diagnostic criteria will be required to achieve this aim. It is
hoped that more effective treatment protocols will, in turn, result in improved outcomes for
individuals affected by this condition.
REFERENCES
1. Jennett B, Teasdale G, Galbraith S, Pickard J, Grant H, Braakman R, et al. Severe head injuries in three
countries. Journal of Neurology, Neurosurgery and Psychiatry 1977;40;291-8.
2. Baguley IJ, Slewa-Younan S, Heriseanu RE, Nott MT, Mudaliar Y, Nayyar V. The incidence of Dysautonomia
and its relationship with autonomic arousal following traumatic brain injury. Brain Inj. 2007;21;1175-82.
3. Baguley IJ, Nicholls JL, Felmingham KL, Crooks J, Gurka, JA, et al. Dysautonomia after traumatic brain
injury: a forgotten syndrome? Journal of Neurology, Neurosurgery and Psychiatry 1999;67;39-43.
4. Baguley IJ. Autonomic complications following central nervous system injury. Seminars in Neurology 2008.
5. Fernandez-Ortega JF, Prieto-Palomino MA, Munoz-Lopez A, Lebron-Gallardo M, Cabrera-Ortiz H, Quesada-
Garcia G. Prognostic influence and computed tomography findings in dysautonomic crises after traumatic
brain injury. Journal of Trauma-Injury Infection and Critical Care 2006;61;1129-33.
6. Hendricks HT, Guerts ACH, van Ginnekin BC, Heeren AJ. Brain injury severity and autonomic dysregulation
accurately predict heterotopic ossification in patients with traumatic brain injury. Clinical Rehabilitation
2007;21;545-53.
7. Fearnside MR, Cook RJ, McDougall P, McNeil RJ. The Westmead Head Injury Project outcome in severe
head injury. A comparative analysis of pre-hospital, clinical and CT variables. Br.J.Neurosurg. 1993;7;267-
79.
8. Krach LE, Kriel RL, Morris WF, Warhol BL, Luxenberg MG. Central autonomic dysfunction following
acquired brain injury in children. J Neurol Rehabil 1997;11;41-5.
9. Dolce G, Quintieri M, Leto E, Milano M, Pileggi A, Lagani V, et al. Dysautonomia and Clinical Outcome
in Vegetative State. Journal of Neurotrauma 2008;25;1079-82.
10. Rabinstein AA. Paroxysmal sympathetic hyperactivity in the neurological intensive care unit. Neurological
Research 2008;29;680-2.
11. Strich SJ. Diffuse degeneration of the cerebral white matter in severe dementia following head injury. J
Neurology, Neurosurgery and Psychiatry 1956;19;163-85.
12. Bullard DE. Diencephalic Seizures: Responsiveness to Bromocriptine and Morphine. Ann.Neurol.
1987;21:609-11.
13. Sneed RC. Hyperpyrexia associated with sustained muscle contractions: an alternative diagnosis to central
fever. Arch.Phys.Med.Rehabil. 1995;76;101-3.
14. Pranzatelli MR, Pavlakis SG, Gould RG, De Vivo DC. Hypothalamic-Midbrain Dysregulation Syndrome:
Hypertension, Hyperthermia, Hyperventilation, and Decerebration. Journal of Child Neurology 1991;6:115-
22.
15. Baguley IJ. Nomenclature of “paroxysmal sympathetic storms”. Mayo Clinic Proceedings 1999;74;105.
16. Senno RG, Anderson V, Ahmed G, Duraski SA. Intrathecal baclofen administration in the management
of hyperadrenergic state in an adult with a severe anoxic brain injury: a case report. Archives of Physical
Medicine and Rehabilitation 2004;85(9);e15.
17. Diesing TS, Wijdicks EF. Arc de Cercle and dysautonomia from anoxic injury. Movement Disorders
2006;21;868-9.
18. Turner MS. Early use of Intrathecal bacolfen in brain injury in pediatric patients. Acta Neurochirurgica -
Supplement 2003;87;81-3.
19. Leow MKS, Loh KC, Kwek TK, Ng PY. Catecholamine and metanephrine excess in intracerebral
haemorrhage: revisiting an obscure yet common “pseudophaeochromocytoma”. Journal of Clinical
Pathology 2007;60;583-4.
20. Becker R, Benes L, Sure U, Hellwig D, Bertalanffy H. Intrathecal baclofen alleviates autonomic dysfunction
in severe brain injury. Journal of Clinical Neuroscience 2000;7;316-9.
21. Rossitch E, Bullard DE. The Autonomic Dysfunction Syndrome: aetiology and treatment. British Journal
of Neurosurgery 1988;2;471-8.
22. Tong C, Konig MW, Roberts PR, Tatter SB, Li X. Autonomic dysfunction secondary to intracerebral
haemorrhage. Anesthesia and Analgesia 2000;91;1450-1.
23. McLean AJ. Autonomic epilepsy: reports of a case with observtions at necropsy. Archives of Neurology
and Psychiatry 1934;189-97.
24. Evans CH, Westfall V, Atuk NO. Astrocytoma mimicking the features of pheochromocytoma. New England
Journal of Medicine 1972;286;1397-9.
25. Soloman GE. Diencephalic autonomic epilepsy caused by a neoplasm. J Pediatrics 1973;83;277-80.
26. Oh SJ, Hong YK, Song E. Paroxysmal autonomic dysregulation with fever that was controlled by propanolol
in a brain neoplasm patient. Korean J Int Med 2007;22;51-4.
27. Scott JS, Ockey RR, Holmes GE, Varghese G. Autonomic dysfunction associated with locked-in syndrome
in a child. Am.J.Phys.Med.Rehabil. 1997;76;200-3.
28. Talman WT, Florek G, Bullard DE. A hyperthermic syndrome in two subjects with acute hydrocephalus.
Archives of Neurology 1988;45;1037-40.
29. Bhigjee AI, Ames FR, Rutherford GS. Adult aqueduct stenosis and diencephalic epilepsy. A case report.
Journal of the Neurological Sciences 1985;71;77-89.
30. Rey M, Borrallo JM, Vogel CM, Pereira MA, Varela MA, Diz JC. Paroxysmal sympathetic storms after type
a dissection of the aorta. Journal of Cardiothoracic and Vascular Anesthesia. 1919;654-5.
31. Baguley IJ, Nott MT. Quantitating the Efficacy of Gabapentin in a Novel Case of Dysautonomia.
Neurorehabilitation and Neural Repair 2008;22;570-1.
32. Baguley IJ. The excitatory:inhibitory model (EIR model): An integrative explanation of acute autonomic
overactivity syndromes. Medical Hypotheses 2007;70;26-35.
33. Onofrj M, Thomas A. Acute akinesia in Parkinson disease. Neurology 2005;64;1162-9.
34. Kipps CM, Fung VS, Grattan-Smith P, de Moore GM, Morris JG. Movement disorder emergencies. Movement
Disorders 2005;20;322-34.
35. Boyer EW,.Shannon M. The serotonin syndrome. New England Journal of Medicine 2005;352;1112-20.
36. Coffey RJ, Edgar TS, Francisco GE, Graziani V, Meythaler JM, Ridgely PM, et al. Abrupt withdrawal from
intrathecal baclofen: recognition and management of a potentially life-threatening syndrome.
Arch.Phys.Med.Rehabil. 2002;83;735-41.
37. Mohammed I,.Hussain A. Intrathecal baclofen withdrawal syndrome- a life-threatening complication of
baclofen pump: a case report. BMC Clinical Pharmacology 2004;4;6.
38. Macrokanis CJ, Hall NL, Mein JK. Irukandji syndrome in northern Western Australia: an emerging health
problem. Medical Journal of Australia 2004;181;699-702.
39. Reed MJ, Clegg GR. Paroxysmal sympathetic surge associated with gamma hydroxybutyrate. European
Journal of Emergency Medicine. 13(1);41-2, 2006.
40. Penfield W. Diencephalic autonomic epilepsy. Archives of Neurology and Psychiatry 1929;22;358-74.
41. Thorley RR, Wertsch JJ, Klingbeil GE. Acute hypothalamic instability in traumatic brain injury: a case
report. Arch.Phys.Med.Rehabil. 2001;82;246-9.
42. Do D, Sheen VL, Broomfield E. Treatment of paroxysmal sympathetic storm with labetalol. Journal of
Neurology, Neurosurgery and Psychiatry 2000;69;832-3.
43. Baguley IJ, Heriseanu RE, Gurka JA, Slewa-Younan S. Gabapentin for the management of dysautonomia
following severe traumatic brain injury: a cautionary tale. Proceedings 4th ISPRM meeting, Korea 2007.
44. Baguley IJ, Heriseanu RE, Cameron ID, Nott MT, Slewa-Younan S. A critical review of the pathophysiology
of Dysautonomia following traumatic brain injury. Neurocritical Care 2008;8;293-300.
45. Penfield W, Jasper H. Somatic motor seizures. In Penfield W, ed. Epilepsy and the functional anatomy of
the human brain, pp 350-88. London: J and A Churchill Ltd, 1954.
46. Sandel ME, Abrams PL, Horn LJ. Hypertension after Brain Injury: Case Report. Arch.Phys.Med.Rehabil.
1986;67:469-72.
47. Boeve BF, Wijdicks EF, Benarroch EE, Schmidt KD. Paroxysmal sympathetic storms (“diencephalic seizures”)
after severe diffuse axonal head injury. Mayo Clinic Proceedings 1998;73;148-52.
48. Russo RN, O’Flaherty S. Bromocriptine for the management of autonomic dysfunction after severe traumatic
brain injury. Journal of Paediatrics and Child Health 2000;36 :283-5.
49. Cuny E, Richer E, Castel JP. Dysautonomia syndrome in the acute recovery phase after traumatic brain
injury: relief with intrathecal Baclofen therapy. Brain Inj. 2001;15;917-25.
50. Baguley IJ, Cameron ID, Green AM, Slewa-Younan S, Marosszeky JE, Gurka JA. Pharmacological
management of Dysautonomia following traumatic brain injury. Brain Inj. 2004;18;409-17.
51. Lemke DM. Riding out the storm: sympathetic storming after traumatic brain injury. Journal of Neuroscience
Nursing 2004;36;4-9.
52. Baguley IJ, Heriseanu RE, Gurka JA, Nordenbo A, Cameron ID. Gabapentin in the management of
dysautonomia following severe traumatic brain injury: a case series. Journal of Neurology, Neurosurgery
and Psychiatry 2007;78;539-41.
53. Srinivasan S, Lim CCT, Thirugnanam U. Paroxysmal autonomic instability with dystonia. Clin Auton Res
2007;17;378-81.
54. Lemke DM. Sympathetic storming after severe traumatic brain injury. Critical Care Nurse 2007;27;30-7
55. Baguley IJ, Nott MT, Slewa-Younan S, Heriseanu RE, Perkes IE. Diagnosing Dysautonomia following
acute traumatic brain injury: evidence for over-responsiveness to afferent stimuli. Arch.Phys.Med.Rehabil.
2008;(in press).
56. Cervero F,.Laird JM. Mechanisms of touch-evoked pain (allodynia): a new model. Pain 1996;68;13-23.
57. Rabinstein AA. Paroxysmal Autonomic Instability after Brain Injury. Arch Neurol 2004;61;1625.
58. Baguley IJ, Heriseanu RE, Felmingham KL, Cameron ID. Dysautonomia and Heart Rate Variability
Following Severe Traumatic Brain Injury. Brain Inj. 2006;20;437-44.
59. Baena RC, Busto R, Dietrich WD, Globus MY, Ginsberg MD. Hyperthermia delayed by 24 hours aggravates
neuronal damage in rat hippocampus following global ischemia. Neurology 1997;48;768-73.
60. Dietrich WD, Alonso O, Halley M, Busto R. Delayed posttraumatic brain hyperthermia worsens outcome
after fluid percussion brain injury: a light and electron microscopic study in rats. Neurosurgery 1996;38;533-
41.
61. Stocchetti N, Rossi S, Zanier E, Colombo A, Beretta L, Citerio G. Pyrexia in head-injured patients admitted
to intensive care. Intensive Care Medicine 2002;28;1555-62.
62. Diringer M, Reaven N, Funk S, Uman G. Elevated body temperature independently contributes to increased
length of stay in neurologic intensive care unit patients. Critical Care Medicine 2004;32;1489-95.
63. Ryan JB, Hicks M, Cropper JR, Garlick SR, Kesteven KH, Wilson MK, et al. Functional evidence of reversible
ischaemic injury immediately after the sympathetic storm associated with experimental brain death. J
Heart Lung Transplant 2003;22;922-8.
64. McLeod AA, Neil-Dwyer G, Meyer CH, Richardson PL, Cruickshank J, Bartlett J. Cardiac sequelae of
acute head injury. British Heart Journal 1982;47;221-6.
65. Hamill RW, Woolf PD, McDonald JV, Lee LA, Kelly M. Catecholamines predict outcome in traumatic
brain injury. Annals of Neurology 1987;21;438-43.
66. Clifton GL, Robertson CS, Choi SC. Assessments of nutritional requirements of head-injured patients.
Journal of Neurosurgery 1986;64;895-901.
67. Moore R, Najarian MP, Konvolinka CW. Measured Energy Expenditure in Severe Head Trauma. J.Trauma.
1989;29;1633-6.
68. Ott L, Young B, Phillips R, McClain C, Adams L, Dempsey, et al. Altered gastric emptying in the head-
injured patient: relationship to feeding intolerance. Journal of Neurosurgery 1991;74;738-42.
69. Mehta NM, Bechard LJ, Leavitt K, Duggan C. Severe weight loss and hypermetabolic paroxysmal
Dysautonomia following hypoxic ischemic brain injury: the role of indirect calorimetry in the Intensive
Care Unit. Journal of Parenteral and Enteral Nutrition 2008;32:281-4.
70. Kishner S, Augustin J. Post Head Injury Autonomic Complications. emedicine.com . 4-10-2006.
71. Blackman JA, Patrick PD, Buck ML, Rust RS Jr. Paroxysmal autonomic instability with dystonia after
brain injury. Archives of Neurology 2004;61;321-8.
72. Bricolo A, Turazzi S, Alexandre A, Rizzuto A. Decerebrate rigidity in acute head injury. Journal of
Neurosurgery 1977;47;680-98.
73. Rabinstein AA, Benarroch EE. Treatment of paroxysmal sympathetic hyperactivity. Current Treatment
Options Neurology 2008;10:151-7.
74. Francois B, Vacher P, Roustan J, Salle J-Y, Vidal J, Moreau JJ, et al. Intrathecal baclofen after traumatic
brain injury: early treatment using a new technique to prevent spasticity. Journal of Trauma-Injury Infection
and Critical Care 2001;50;158-61.
75. Anderson VL, Ahmed G, Duraski SA, Senno RG. Alternative treatment in the management of combined
hyperadrenergia and spasticity in the adult with a severe traumatic brain injury: A case report. Archives
of Physical Medicine and Rehabilitation 85(9), e15. 2004.
76. Follett KA, Burchiel K, Deer T, DuPen S, Prager J, Turner MS, et al. Prevention of Intrathecal Drug Delivery
Catheter-Related Complications. Neuromodulation 2003;6;32-41.
77. Ivanhoe CB, Tilton AH, Francisco GE. Intrathecal baclofen therapy for spastic hypertonia. Physical Medicine
and Rehabilitation Clinics of North America 2001;12;923-38.
78. Guillaume D, Van Havenbergh A, Vloeberghs M, Vidal J, Roeste G. A clinical study of intrathecal baclofen
using a programmable pump for intractable spasticity. Arch.Phys.Med.Rehabil. 2005;86;2165-71.
79. Meythaler JM, Stinson AM. Fever of central origin in traumatic brain injury controlled with propranolol.
Arch.Phys.Med.Rehabil. 1994;75;816-8.
80. Chiolero RL, Breitenstein E, Thorin D, Christin L, de Tribolet N, Freeman J, et al. Effects of propranolol
on resting metabolic rate after severe head injury. Critical Care Medicine 1989;17;328-34.
81. Silver JK, Lux WE. Early onset dystonia following traumatic brain injury. Arch.Phys.Med.Rehabil.
1994;75;885-8.
82. Wortsman J, Burns G, Van Beek AL, Couch J. Hyperadrenergic state after trauma to the neuroaxis.
Journal of the American Medical Association 1980;243;1459-60.
83. Riordan WP Jr., Cotton BA, Norris PR, Waitman LR, Jenkins JM, Morris JA Jr. Beta-blocker exposure in
patients with severe traumatic brain injury (TBI) and cardiac uncoupling. J Trauma 2007;63;503-10.
84. Salim A, Hadjizacharia P, Brown C, Inaba K, Teixeira PG, Chan L, et al. Significance of troponin elevation
after severe traumatic brain injury. Journal of Trauma-Injury Infection and Critical Care 2008;64;46-52.
85. Robertson CS, Clifton GL, Taylor AA, Grossman RG. Treatment of hypertension associated with head
injury. Journal of Neurosurgery 1983;59;455-60.
86. Feibel JH, Baldwin CA, Joynt RJ. Catecholamine-associated refractory hypertension following acute
intracranial haemorrhage: control with propranolol. Annals of Neurology 1981;9;340-3.
87. Payen D, Quintin L, Plaisance P, Chiron B, Lhoste F. Head injury: clonidine decreases plasma
catecholamines. Critical Care Medicine 1990;18;392-5.
88. Metz SA, Halter JB, Porte D Jr., Robertson RP. Autonomic epilepsy: clonidine blockade of paroxysmal
catecholamine release and flushing. Anns Int .Med 1978;88;189-93.
89. Dunne, S. L. Clonidine for the treatment of paroxysmal autonomic instability with dystonia following
traumatic brain injury. Neonatal and Paediatric Pharmacists Group (NPPG) Conference . 2007.
90. Mysiw WJ, Bogner JA, Corrigan JD, Fugate LP, Clinchot DM, Kadyan V. The impact of acute care
medications on rehabilitation outcome after traumatic brain injury. Brain Injury 2006;20:905-11.
91. Baguley IJ, Nott MT. Investigating the Impact of Differing Diagnostic Criteria for Dysautonomia and ‘PAID’:
Determining Caseness. Neurorehabilitation and Neural Repair 2008;22:570.
Section E: Tropical Diseases in ICU
Sixteen
Artesunate as First Line Therapy for
Severe Falciparum Malaria
Ravikiran Sonawane, Dilip R Karnad
INTRODUCTION
Falciparum malaria is one of the most menacing public health problems faced worldwide
especially in tropical countries. The number of cases of malaria worldwide appears to be
growing because of an increased risk of transmission in areas where malaria control has
declined, increasing prevalence of drug-resistant strains of parasites (e.g., chloroquine resistance)
and increasing international travel. The annual clinical caseload Is estimated to be well over
500 million, that, in turn leads to between 1 and 3 million deaths, mainly among young
children.1 Of all patients with falciparum infection, 90% of cases occur in Africa, followed by
Asia. Nearly 80% of cases in Asia occur in India. In India, the mortality rate has steadily
increased from 0.09/100,000 population in 2000 to 0.15/ 100,000 in 2006.2 The case fatality
rate among critically ill patients (admitted to ICU) with severe malaria is estimated to range
between 10 to 50%.
Severe falciparum malaria, as defined by the WHO criteria (Table 16.1),3 is treated by
parenteral antimalarials like quinine or artusunate. Quinine has traditionally been the drug of
first choice for severe malaria. However, with the advent of the artemesinin derivatives, there
has been a reassessment of priorities centred on the hope that more rapidly parasiticidal
activity could potentially reduce mortality in severe malaria. Unfortunately, initial randomized
controlled trials in Asia and Africa have failed to show any benefit of artemisin derivatives
over quinine. Recent studies have suggested that artesunate may be better suited for treatment
of severe malaria than the other artemesinin derivatives because of its superior pharmaco-
dynamics. This review focuses on the evidence for and against the viewpoint that supports
the use of artusunate as first line therapy of complicated falciparum malaria in select patient
populations.
ANTIMALARIAL THERAPY4
The principal effect of anti-malarials is to inhibit parasite maturation and multiplication. Left
untreated, the parasite can multiply at a rate limited only by the average number of viable
Artesunate as First Line Therapy for Severe Falciparum Malaria 167
Table 16.1: Case definition of severe falciparum malaria3
Cerebral malaria Unrousable coma (GCS < 11/15), with peripheral P.falciparum parasitaemia
after exclusion of other causesof encephalopathy
Severe anaemia Normocytic anaemia with haemoglobin <5 g /dl(haematocrit<15%) in

presence of parasitaemia >10 000/ ml
Respiratory distress Pulmonary oedema or acute respiratory distress syndrome (ARDS)Would
now also include rapid laboured ‘acidotic’ breathingsometimes abnormal
in rhythm
Renal failure Urine output of less than 400 ml in 24 h (or <12 ml/kg in children) and a
serum creatinine >265 mmol/l (>3.0 mg/dl)
Hypoglycaemia Whole blood glucose <2.2 mmol l”1 (40 mg/dl)
Circulatory collapse Systolic blood pressure <70 mmHg or core-skin temperaturedifference
>10°C
Coagulation failure Spontaneous bleeding and/or laboratory evidence of disseminated
intravascular coagulation
Complicated malaria Such patients with complicated malaria should be managed assevere
Impaired consciousness of any malaria, i.e. with parenteral antimalarials even thoughthey do not
degree, prostration, jaundice, necessarily meet the criteria of severe disease.
intractable vomiting, parasitaemia
>2% in non-immune individuals
(no previous malaria)
merozoites per mature schizont. Anti-malarial drugs exert their maximum effects (Emax) by
reducing the parasite load by a factor of between 10 and 10,000 per cycle—a process that
appears to follow first order kinetics throughout. In simple terms, this means that in the presence
of minimum parasiticidal concentration (MPC) of an antimalarial, a fixed fraction of the parasite
population is removed at each successive asexual cycle. Patients with acute malaria may
have an estimated 1012 parasites in the circulation. Even with kill rate efficiency as high as
99.99%, it takes no less than three life cycles (6 days) to completely eradicate all parasites.
Thus, anti-malarial treatment must provide therapeutic drug concentrations for a minimum of
seven days (covering four cycles) to affect a cure. For drugs rapidly eliminated from the body,
this means the course of treatment must also be of seven days duration.
In the treatment of malaria, the stage of anti malarial activity is also important, as the
object of treatment is to stop parasite maturation, particularly from the less pathogenic ring
forms in circulation to the more virulent cyto-adherent stages. Drugs used for treatment of
severe malaria act predominantly in the middle third of the life cycle, during which there is
the greatest increase in parasite synthetic and metabolic activity. None of the drugs prevent
rupture of RBC and reinvasion of cells, once the schizont has formed. Young rings are also
relatively resistant to therapy, particularly to quinine and pyrimethamine. Artemisinin
compounds have the broadest time window of antimalarial action and the most rapid in vivo
activity. These compounds prevent maturation of ring stages, thereby inducing accelerated
clearance and reducing subsequent cyto-adherence; an action quite unlike that of quinine.
QUININE
Quinine was the primary treatment of malaria all over the world till the end of World War II
when it was replaced by the less toxic chloroquine. By 1960, plasmodium falciparum resistant
to chloroquine was identified in Thailand and Colombia.5 Fifty years later, there are very few
areas in the world, which are without chloroquine resistant plasmodium falciparum. Hence,
quinine has re-emerged as the principal anti-malarial agent for severe malaria (except in the
United States, where quinidine is used). Quinine is a blood schizonticidal agent and is active
against the asexual erythrocytic forms but is not active against sporozoites, pre-erythrocytic,
exoerythrocytic forms or gametocytes of P falciparum. It is believed to act by inhibiting the
enzyme heme polymerase, that allows an accumulation of its cytotoxic substrate, heme.6
Pharmacokinetics and Dose

Quinine is readily absorbed orally as well as parenterally reaching peak concentrations in
4 hours. Bioavailability is approximately 80% in healthy subjects. Quinine is 70 to 85% protein
bound. In malaria, the volume of distribution and clearance time is reduced in direct proportion
to the severity of illness. As a result, blood concentrations are highest in patients with severe
malaria. Quinine is extensively metabolized, mainly in the liver (>80%) and the metabolites
are excreted in urine, mainly in the form of hydroxy derivatives; small amounts also appear
in the feces, gastric juice, bile and saliva.7
It can be given orally or via the intravenous (IV) or intramuscular (IM) route. For intravenous
use, 20 mg/kg of quinine in 100 ml of 25% of dextrose is infused as a loading dose over
3-4 hours followed by 10 mg/kg in 100 ml 25% of dextrose (over 3-4 hrs), every 8 hours.
Intramuscular dose is similar to the IV dose. Oral dose is 10 mg/kg 8 hourly for 7 days.8
Problems with Quinine

Quinine is not considered as an ideal anti-malarial drug mainly because of its narrow
therapeutic window and adverse effects like cardiotoxicity, cinchonism, hyperinsulinemic,
hypoglycaemia and painful local reactions after intramuscular administration. Moreover, it
acts mainly on the middle third of the lifecycle of the parasite and does not prevent parasite
maturation. It also does not act on the young ring forms which go on to mature to cause cyto-
adherance.The Emax is less compared to artemisinin derivatives; this accounting for its longer
parasite clearance time.9 Finally, recent studies, especially from South-East Asia, have
demonstrated decreased efficacy as well as resistance to quinine.10
ARTESUNATE
In 1971, a new class of antimalarial drug was discovered from the Chinese medicinal herb
qinghao (Artemisia annua).11 The herb (sweet wormwood) was recommended for fevers in
the Chinese materia medica texts since 341 AD. The highly active chemical from qinghao is
now called artemisinin.4 Artemesinin derivatives combine potent, rapid antimalarial activity
with a wide therapeutic index along with an absence of clinically important resistance.
Artemisinin containing regimens met the urgent need to find effective treatments for multidrug
resistant malaria and have recently been advocated for more widespread use.
Pharmacokinetics
Artemisinin is a sesquiterpene lactone structure which is highly crystalline compound that
does not dissolve in oil or water. It has poor bioavailibilty; hence, semisynthetic derivatives
have been developed. They have been chemically modified to produce artesunate, artemether,
arteether, dihydroartemisinin, and artelinic acid. These compounds have variously been
formulated for oral, rectal, and parenteral use.
In the body, most artemisinin derivatives are converted to the active form, dihydroarte-
misinin (DHA). Artesunate is rapidly hydrolysed to DHA within minutes and its antimalarial
activity is largely mediated by DHA. DHA is mostly (90%) bound to plasma proteins and is
inactived by metabolism by hepatic cytochrome P-450 and other enzyme systems. DHA is
itself a potent antimalarial with an elimination half life of about 45 minutes.4 Despite this,
once daily dosing of oral artesunate results in parasite clearance kinetics comparable to twice
daily dosing, suggesting that the biological effects of artemisinins extend beyond the duration
of their therapeutic concentrations in plasma, similar to a post-antibiotic effect.
The absolute bioavailability of antimalarial activity after a single dose of oral artesunate
in uncomplicated adult malaria is about 60%.12 Time to maximum DHA concentration is
typically one to two hours. Studies suggest that clearance time of artesunate is reduced during
acute infection compared to the recovery period, either through disease effects on
pharmacokinetics or via enzyme autoinduction. Parenteral artesunate is pharmacokinetically
superior to artemether for the treatment of severe malaria, whether given intravenously or by
the intramuscular route (to children).13 Absorption from the intramuscular site in both adults
with uncomplicated malaria and children with severe malaria is rapid with peak DHA
concentrations achieved within an hour and with a bioavailability of over 80%. Severity of
malaria infection seems to have no significant influence on artesunate pharmacokinetics but
age may have.
Rectal artesunate in African children with moderate malaria (defined as being unable to
take oral medications or prostration/obtundation) shows rapid but variable absorption with
peak plasma DHA concentrations appearing in about two hours and bioavailability of between
20% and 60%.14
Antimalarial Activity
The antimalarial action of artemisinins has been attributed to their chemical capability of
generating free radicals. An alternative mechanism of action is inhibition of the malarial
parasite’s sarcoplasmic endoplasmic reticulum calcium ATPase, SERCA.15
Artemisinins kill all species of plasmodium that infect humans. In P falciparum infection,
they produce a 10, 000-fold reduction in parasite load per asexual cycle.9 Besides the large
ring stage of infection, artemisinins also target early tiny ring stages of plasmodia (present
only a few hours after red cells are invaded by merozoite stages). Artemisinins do not interfere
with hepatic stages of parasite development but they are active against early gametocyte
stages of development and have the potential to interfere with mosquito transmission.4
Problems with Artemisinins

Though they have a faster parasite clearance rate, the cure rate is poor with monotherapy.
High recrudescence rates are common and are attributed to the short half life of artemisinins,
rather than drug resistance. Hence combination therapy is advocated. Animal studies have
demonstrated neurotoxicity in the form of brainstem toxicity; a single case of cerebellar ataxia
has been reported after treatment with artesunate and lumefantrine. Gastrointestinal
disturbance, allergic reactions and hemolysis are rare. Fetotoxicity is a concern based on
animal studies; hence these drugs are not advised during first trimester of pregnancy.
DETERMINANTS OF AN IDEAL ANTIMALARIAL DRUG

The important features required in an ideal anyimalarial drug are:
1. Stage specificity—Plasmodia pass through different stages of form, location and clinical
consequence. Susceptibility to antimalarial drugs varies with each stage. Whereas quinine
has a narrow stage spectrum, atremisinin derivatives have a much wider spectrum and act
on many stages of the plasmodial lifecycle.
2. Parasite clearance—The ability of the drug to cause fractional decline in parasites with
each asexual cycle (Emax). Artemisinins have better Emax than quinine. The larger burden
of parasites in severe falciparum malaria also require the drug to be present for longer in
the blood, hence pharmacokinetics of the drug also contibute to its efficiency.
3. Safety and tolerability—A drug with better safety and tolerability profile would definitely
be preferred.
4. Resistance—Parasites causing malaria exhibit a range of intrinsic susceptibility to antimalarial
agents which is species and stage specific. Moreover, acquired resistance to antimalarials
has been a major problem in Africa, South-East Asia.
If we look at all these parameters, Artesunate scores over quinine in all aspects, except for
cost, availability and probably the cumulative clinical experience built up with use of quinine
over the decades.
However, in the treatment of severe malaria, these abovementioned features would only
qualify as secondary or less important end points. The main benefit that needs to be looked
for is a meaningful measure of outcome. Parameters such as fever clearance time, parasite
clearance time, and coma resolution time have limited value. The most important primary
outcome measure is mortality.
ARTEMESININ DERIVATIVES VS. QUININE

Initial randomized trials were done comparing artemether or arte-ether with quinine. A
randomized controlled trial comparing intramuscular artemether to intramuscular quinine in
560 Vietnamese adults with severe malaria reported that parasite clearance was faster with
artemether (72h vs 90h with quinine), but fever clearance time, coma resolution time and
hospital stay was longer than with quinine. Difference in mortality (13% with artemether and
17% with quinine) was not statistically significant.16 Another randomized trial in 576 Gambian
children with severe malaria showed that incidence of residual neurological deficits and
mortality in the two treatment groups (20.5% with artemether vs 21.5% with quinine) were
not statistically significant.17 However, both studies showed that adverse events were
significantly and expectedly higher with quinine.
Apart from these two large studies, a number of small studies from all over the world have
compared artemether with quinine. In 1999, a meta-analysis done by Pittler and Ernst, on
the results of 9 studies comparing artemether with quinine,18 revealed that there was no
significant difference in mortality with artemether and quinine in studies from the Subsaharan
region. However, studies from Southeast Asia showed mortality rate to be 40% lower among
patients treated with artemether when this was compared with quinine. This result is probably
due to an increase in resistance of P. falciparum to quinine in Southeast Asian countries.
Overall, both the agents were equal in terms of efficacy, with a trend toward greater effectiveness
of artemether in regions where there is recognized quinine resistance.
A Cochrane review comparing another artemesinin derivative, arte-ether with quinine in
the treatment of severe malaria identified only 2 randomized controlled studies. Four Indian
studies on alpha, beta arte-ether, a drug that was developed in India, were excluded as these
studies did not have a control group. In 194 patients from the 2 randomized studies that were
included in the meta analysis, there was no statistically significant difference in mortality.19
Both arteether and artemether are oil based formulations that release the drug slowly and
erratically from the intramuscular injection site, whereas artesunate can be given intravenously
and is also absorbed rapidly and reliably after intramuscular injection with peak concentrations
arising within 1 hour.20 Therefore, subsequent investigators have focussed their attention on
water soluble and pharmacokinetically superior artesunate in severe malaria.
The largest multicentre trial comparing intravenous artesunate with intravenous quinine is
called the SEQUAMAT (South East Asian Quinine Artesunate Malaria Treatment Group)
study.21 It included adults and children from four South-East Asian countries, namely
Bangladesh, India, Indonesia and Myanmar. This study enrolled 730 patients to randomly
receive artesunate (2.4 mg/kg bolus at 0,12, 24 h and then daily for 7 days) and compared
outcomes with 731 patients who received quinine (20 mg/kg loading dose infused over
4 hours followed by 10 mg/kg infused over 2-8 hours three times a day for 7 days). Both
medications were started as intravenous therapy but switched to the oral route as soon as
feasible. This study concluded that artesunate reduced mortality in patients with severe malaria
by over a third as compared to quinine (15% in artesunate group vs 22% in the quinine
group; ARR = 34.7%, 95% CI 18.5–47.6; p = 0.0002).
A recent meta-analysis combined data from 6 trials comparing artesunate with quinine
in severe malaria; all trials were conducted in Asia. This included 1664 adults and 274
children. Treatment with artesunate significantly reduced the risk of death (RR 0.62, 95%
CI 0.51-0.75). Parasite clearance time and incidence of hypoglycaemia too was lower with
artesunate. However, neurological sequelae, coma recovery time, fever clearance time,
hospital stay and other adverse effects were comparable with the 2 drugs.22
RECOMMENDATIONS
Of the 1461 patients enrolled in the SEQUAMAT study, 141 were Indians. Subgroup analysis
revealed that there was no significant difference in mortality in the Indian subgroup. While
this may be due to the lower power of the subgroup analysis to detect any mortality benefit,
this difference could also be due to absence of resistance to quinine in India. A decreased
efficacy of quinine has been reported in South-East Asia, including Myanmar and Indonesia.
In this situation, the SEAQUAMAT study could be interpreted to suggest that the benefit of
artesunate over quinine may be due to partial quinine resistance rather than due to the efficacy
of artesunate. In other words, whether artesunate is superior to quinine in areas where there
is no quinine resistance (like India) is not clear. A small study done from India in 2005 also
showed no mortality benefit of artesunate over quinine.24
National Drug Policy on Malaria25

The National Drug Policy on Malaria (2007) recommends that in severe and complicated
P. falciparum malaria cases, intravenous quinine is still the drug of choice. Parenteral artesunate
is recommended as second line therapy.
WHO Treatment Guidelines8

The World Health Organization released revised treatment guidelines for the treatment of
malaria in 2006. These guidelines, based on the SEQUAMAT study, recommended use of
artemisin compounds for the treatment of severe falciparum malaria as summarised in Table
16.2. Notwithstanding these recommendations, Artesunate is not yet licenced for use in the
United States, but can only be obtained from the Centres for Disease Control (CDC, Atlanta)
on a case-by-case basis.23
CONCLUSION
Artesunate is not only an effective alternative to quinine in treatment of severe malaria but
an attractive first line option. It has faster parasite clearance time, lesser incidence of post-
treatment hypoglcemia, and a potential mortality benefit especially in regions with resistance
to quinine (SE Asia). Other advantages include ease of administration (bolus instead of
Table 16.2: WHO treatment guidelines8

1. Artesunate 2.4 mg/kg body weight IV or IM given at admission (Time 0), then at 12 h and 24 h. Thereafter,
once a day is recommended for treatment of severe P falciparum malaria in low transmission areas or
outside malaria endemic areas.
2. For children with severe malaria in high transmission areas, both artemisinin and quinine are recommended
as there is insufficient evidence to recommend any of these antimalarial medicines over another.
3. For severe malaria during pregnancy, use the parenteral antimalarial treatment available locally should be
used in full doses. Where available, Artesunate should be the first line, and artemether should be the second
line option during the second and third trimesters. During the first trimester, until more evidence becomes
available, both artesunate and quinine may be considered as options.
4. If used, a loading dose of 20 mg /kg body weight of quinine is recommended. Rate controlled IV infusion
is preferred, but if this cannot be given safely, then IM injection is a satisfactory alternative.
5. Artemether IM is an acceptable alternative to IV quinine. Use of artemotil is NOT recommended unless
alternatives are not available. Use of artemisinins rectally for treatment is suggested ONLY when parenteral
antimalarial treatment is not possible.
6. Quinidine is considered more toxic than quinine and should only be used if none of the other effective
parenteral drugs are available.
infusion), less frequent dosing, better tolerability and lesser adverse effects. Artesunate is equally
effective as quinine in children with regards to mortality. As in adults, it has better safety
profile, ease of administration and better tolerability.
The outcome benefits of artesunate are probably due to pharmacodynamic differences
between artesunate and quinine. Artesunate kills circulating ring-stage parasites, which can
then be removed by the spleen, whereas quinine does not. Thus, artesunate prevents
maturation of younger parasites into mature forms and this prevents sequestration of RBC16,
and consequent microcirculatory disturbances. Artesunate has an advantage in patients with
high parasitaemia load.4
Though artesunate can be considered as first line drug, it may not be useful as monotherapy
in severe malaria due to its pharmacokinetics and may lead to recrudescence and poor cure
rates; hence, it should always be used in combination with another slower acting antimalarial.
The recommended regimen is seven days of oral doxycycline whenever patient can take oral
drugs. Clindamycin can be used instead of doxycycline in children and pregnant women in
whom doxycycline is contraindicated.4,8
REFERENCES
1. Snow RW, Guerra CA, Noor AM, et al. The global distribution of clinical episodes of Plasmodium falciparum
malaria. Nature 2005;434:214-7.
2. http://www.searo.who.int/LinkFiles/Malaria_in_the_SEAR_SEAmal_mob_mort_India06.pdf, accessed on Jan
3, 2009.
3. WHO. Severe falciparum malaria. World Health Organization, Communicable Diseases Cluster. Trans R
Soc Trop Med Hyg 2000;94(Suppl 1):S1-90.
4. Woodrow CJ, Haynes RK, Krishna S. Artemisinins. Postgrad Med J 2005;81:71-8.
5. Stephen L Hoffmann. Artemether in Severe Malaria- Still too many Deaths. N Engl J Med 1996;335:124-
6.
6. http://www.malariasite.com/malaria/quinine.htm, accessed on Jan 3, 2009.

7. Quinine Pharmacokinetics in Chronic Hemodialysis Patients. Br J Clin Pharmacol 2002;54:604-9.
8. World Health Organization. Roll Back Malaria Dept. Guidelines for the treatment of malaria [WHO/HTM/
MAL/2006.1108]. Geneva, World Health Organization, 2006.
9. White NJ. Assessment of the Pharmacodynamic Properties of Antimalarial Drugs in vivo. Antimicrob Agents
Chemother 1997;41:1413-22.
10. Pukrittayakamee S, Supanaranond W, Looaresuwan S, et al. Quinine in Severe Falciparum Malaria:
Evidence of Declining Efficacy in Thailand. Trans R Soc Trop Med Hyg 1994;88:324-7.
11. HeinTT, White NJ. Quinghaosu. Lancet 1993;341:603-8.
12. Batty KT, Thu LT, Davis TM et al. A Pharmacokinetic and Pharmacodynamic Study of Intravenous versus
Oral Artesunate in Uncomplicated Falciparum Malaria. Br J Clin Pharmacol 1998;45:123-9.
13. Nealon C, Dzeing A, Muller Romer U et al. Intramuscular Bioavailability and Clinical Efficacy of Artesunate
in Gabonese Children with Severe Malaria. Antimicrob Agents Chemother 2002;46:3933-9.
14. Barnes KI, Mwenechanya J, Tembo M, et al. Efficacy of rectal artesunate compared with parenteral
quinine in initial treatment of moderately severe malaria in African children and adults: a randomised
study. Lancet 2004;363:1598-605.
15. Eckstein-Ludwig U, Webb R, van Goethem ID, et al. Artemisinins target the SERCA of Plasmodium
falciparum. Nature 2003;424:957-61.
16. Hein TT, Day NPJ, Phu NH, et al. A controlled trial of artemether or quinine in Vietnamese adults with
severe falciparum malaria. N Engl J Med 1996;335:76-83.
17. van Hensbroek MB, Onyiorah E, Jaffar S, Schneider G, Palmer A, Frenkel J, et al. A trial of artemether
or quinine in children with cerebral malaria. N Engl J Med 1996;335:69-75.
18. Pittler MH, Ernst E. Artemether for severe malaria: a meta-analysis of randomized clinical trials. Clin
Infect Dis 1999;28:597-601.
19. Afolabi BB, Okoromah CN. Intramuscular arteether for treating severe malaria. Cochrane Database of
Systematic Reviews 2004, Issue 4. Art. No.: CD004391. DOI: 10.1002/14651858.CD004391.pub2.
20. Hein TT, Davis TM, Chuang LV, et al. Comparative Pharmacokinetics of Intramuscular Artesuante and
Artemether in Patients with Severe Falciparum Malaria. Antimicrob Agents Chemother 2004;48:4234-9.
21. South East Asian Quinine Artesunate Malaria Trial (SEAAQUAMAT) Group: Artesunate versus quinine for
treatment of severe falciparum malaria: a randomized trial. Lancet 2005;136:717-7.
22. Jones KL, Donegan S, Lalloo DG. Artesunate versus quinine for treating severe malaria. Cochrane Database
of Systematic Reviews 2007, Issue 4. Art. No.: CD005967. DOI: 10.1002/14651858.CD005967.pub2.
23. Rosenthal PJ. Artesunate for the treatment of severe falciparum malaria. N Engl J Med 2008;358:1829-
36.
24. Haroon N, Amichandwala K, Solu M. Comparative efficacy of quinine and artesunate in the treatment of
severe malaria: A randomized controlled trial. JK Science 2005;7(1):32-5
25. http://www.nvbdcp.gov.in/Doc/Revised%20drug%20policy.pdf , accessed on Jan 3, 2009.
Management of Airway and Autonomic Dysfunction in Tetanus 175
Seventeen
Management of Airway and
Autonomic Dysfunction in Tetanus
FE Udwadia
INTRODUCTION
Tetanus is a major health problem in poor developing countries of the world and is still
occasionally seen in the developed world. The WHO estimated that there were approximately
1,000,000 deaths from tetanus worldwide in 1992. This included 582,000 deaths from neonatal
tetanus, of which 210,000 were in South-East Asia and 152,000 in Africa.1 In 2001, the
WHO estimated that worldwide around 281,000 deaths occurred annually, of which 210,000
were among children under the age of five (including neonatal tetanus).2 These numbers
should be interpreted with caution, as there is a great variability in reporting. Even so, it is
safe to conclude that the incidence and mortality from tetanus have declined due to more
widespread use of preventive measures.
Modern intensive care involving efficient ventilatory support can prevent death from acute
respiratory failure even in severe cases of tetanus. However, death in these patients can still
result from cardiovascular complications due to severe autonomic dysfunction as also from
other complications related to a prolonged critical illness.
SEVERITY GRADING
Based on our experience on 270 patients of tetanus over many years, we formed detailed
clinical criteria for grading tetanus. Our criteria are a modification of Albetts original
classification.3
Grade I (Mild): Mild to moderate trismus, general spasticity, no respiratory embarrassment,
no spasms, little or no dysphagia.
Grade II (Moderate): Moderate trismus, well marked rigidity, mild to moderate, but short
lasting spasms, moderate respiratory embarrassment with tachypnoea over 30-35/minute,
mild dysphagia.
Grade III (Severe): Severe trismus, generalized spasticity, reflex and often spontaneous
prolonged spasms, respiratory embarrassment with tachypnoea over 40/minute, apnoeic spells,
severe dysphagia, tachycardia usually over 120/minute; a steady moderate increase in

autonomic nervous system activity.
Grade IV (Very Severe): Features of Grade Ill plus violent autonomic disturbances often resulting
in what may be aptly termed ‘autonomic storms’ involving the cardiovascular system.
The above classification is not sacrosanct. Both, the severity of seizures and “autonomic
storms” are independent indices of severity. Though severe autonomic dysfunction generally
occurs in patients with grade III (severe) tetanus, it may occasionally occur in Grade II or
moderately severe disease. It is also important to be aware that patients presenting with Grade
I tetanus can graduate within 3 to 7 days to Grade III (severe) or even Grade IV (fulminant)
tetanus. Close observation and re-grading of severity is therefore absolutely necessary.
The grading of the severity of tetanus is not just of academic interest. It is of help in
making important management decisions. Grade I tetanus should be treated conservatively
with the use of long acting sedatives alone. In grade II tetanus, in addition to the use of
sedatives, the airway should be secured and muscle relaxants should be used. Grade III and
IV tetanus require sedation, tracheostomy and continuous ventilatory support following induced
paralysis with neuroparalytic drugs till such time as spasms relent and recovery occurs. Overall
quick diagnosis and prompt treatment of complications that arise in moderate or severe forms
of the disease is vital if the management strategies outlined above are to succeed.
PRINCIPLES OF AIRWAY MANAGEMENT IN TETANUS

Securing and maintaining an artificial airway is of utmost importance in the management of
severe tetanus. If an artificial airway is not promptly secured and effectively managed, death
often results from acute respiratory failure due to airway obstruction, hypoxia, aspiration or
pneumonia. The artificial airway that is preferred is a tracheostomy. Endotracheal intubation
is not preferred for three reasons. Firstly, most patients with tetanus who require an airway
need it for well over a week; patients with severe tetanus may require an airway for as long
as six to eight weeks. Secondly, it is extremely difficult and often impossible to intubate a
patient without triggering off spasms. Finally, clearance of tracheobronchial secretions in severe
tetanus through an endotracheal tube may not be effective.
Endotracheal intubation is, however, often done to temporarily secure the airway. It is
then a prelude to be followed promptly by an electively performed tracheostomy. Endotracheal
intubation in severe tetanus can be notoriously difficult or even impossible. It should be
attempted only by an expert. Intravenous diazepam or midazolam in combination with propofol
and neuroparalytic agents should be used rpior to and during the procedure.
Tracheostomy is mandatory in Grade III and Grade IV tetanus, not only to ensure control
and patency of the airways but also for ventilatory support. It should also be considered in
grade II tetanus because even in an intensive care environment, an important cause of death
in these patients is sudden prolonged unrelieved laryngeal spasm leading to fatal asphyxia.
The inability of patients with grade II tetanus to handle upper respiratory secretions, the
presence of dysphagia with the associated danger of aspiration pneumonia and the presence
of secretions in the upper airways which need to be aspirated to ensure their patency are all
indications for an elective tracheostomy. The use of heavy sedation in many cases of Grade
II tetanus can depress respiration and airway reflexes. An elective tracheostomy is thus
necessary, as it allows for immediate ventilator support under these circumstances.
Our experience4 has been with open surgical tracheostomy performed under local
anaesthesia together with the use of sedatives given intravenously just before the procedure.
All tracheostomies were successfully performed, while observing every possible aseptic
precaution within the ICU itself. Patients therefore, need not be moved to the operation theatre,
as there are inherent risks in moving patients with severe tetanus out of the ICU. In the eighties
and nineties the technique of percutaneous tracheostomy had not been devised. Today it is
a safe, established technique used in many intensive care units around the world. It is
particularly suitable for patients with tetanus. Potential reported advantages over traditional
open surgical techniques include reduced blood loss, lower morbidity and few long-term
sequelae.5 The single step forceps dilatational method is rapid and can be used for emergency
airway management while the serial dilatational technique is useful for patients who have
already been intubated.6,7
Percutaneous tracheostomy should be performed only by those familiar with its technique.
There are not many intensive care units in our country that have the necessary experience
with this procedure. Under these circumstances, an open tracheostomy is to be preferred.
The principles in management of tracheostomy are the same as for any other critical illness.
Only those aspects of tracheostomy care, which are of special relevance to tetanus are briefly
discussed below.
1. A blocked tracheostomy tube or disconnection from the ventilator are disasters to be avoided
and if they occur, promptly recognized and treated. They are more frequent with severe
tetanus because of the prolonged nature of the illness, the copious secretions in the
respiratory tract and the use of neuroparalytic drugs. A sudden rise in peak and pause
pressures on the ventilator, a fall in oxygen saturation, a worsening in the circulatory
hemodynamics or a difficulty in negotiating a suction catheter through the tracheostomy
tube should alert the physician to the possibility of a blocked tube. When in doubt one
must always change the tracheostomy tube. Disconnection of the tube from the ventilator
in a paralysed patient if not promptly corrected can lead to hypoxic cardiac arrest and also
cause surgical emphysema of the neck, chest and at times, of the mediastinum.
2. Suction should be done as often as is necessary. Tetanus is generally associated with copious
respiratory secretions, requiring suction through the tracheostomy every 15 to 30 minutes.
In patients who are hypoxic because of respiratory complications, the FiO 2 should be
sharply stepped up for 5 minutes before suctioning. Suction should not last for more than
10 seconds at a time. Sudden bradycardia with hypotension or even a cardiac arrest may
occur in severe tetanus during suction. Hence, the doctor or nurse should keep a careful
watch on the monitor and stop suction if the heart rate starts to fall. Atropine to counter
a bradyrhythm should be at hand.
3. Good physiotherapy is vital for successful outcome and may well make the difference
between life and death. It is given four times during the day, twice at night. Frequent
change in posture from side to side is necessary. Physiotherapy should be timed when the
patient is under the maximum influence of a neuroparalytic drug. Patients not on ventilatory
support, an IV dose of 10 to 20 mg of diazepam may be necessary before physiotherapy.
Even so, segmental atelectasis frequently and invariably occurs; the atelectatic segments
should be opened up promptly. Delay in doing so compounds the problem. The use of
a fiberoptic bronchoscope to open an atelectatic lobe not responding to physiotherapy is
at times necessary. However the frequent use of the bronchoscope to aspirate secretions
is inadvisable and invariably promotes infection. A daily or even more frequent X-ray of
the chest during the critical phase of the disease is mandatory to help diagnose early signs
of pneumonia or the onset of the acute respiratory distress syndrome.
4. Every attempt should be made to prevent iatrogenic pulmonary infection, more so because
of the severity of the illness and its prolonged duration. Every single suction should be like
a surgical procedure, using the no-touch technique; the tracheostomy wound should receive
meticulous attention. A yellowish or greenish tracheal aspirate points to tracheobronchial
or pulmonary infection. A culture of the tracheal aspirate plus antibiotic sensitivity tests
help in the choice of an appropriate antibiotic. Nosocomial pneumonia is invariably due
to gram negative infection, rarely due to staphylococcal infection.
5. Other complications incidental to a tracheostomy are the same as those encountered in
any other critical illnesses. Tracheal stenosis is a feared long-term sequalae, which can be
avoided by meticulous care of the tracheostomy tube and by periodic deflation of the low-
pressure high volume cuff. Since these patients also have a nasogastric tube in place for
many weeks, a tracheo-oesophageal fistula is a possible complication that needs to be
borne in mind.
6. Extubation should be done only when seizures cease, rigidity particularly of the intercostals
is markedly reduced, tidal volume and vital capacity are adequate and blood gases are
satisfactory. There should be haemodynamic stability and no major pulmonary
complications. The patient should be able to cough well, swallow satisfactorily and thereby
protect his airway. The tracheostomy tube may be necessary for quite a few days after
weaning from ventilator support.
AUTONOMIC DYSFUNCTION
Severe autonomic dysfunction and in particular ‘autonomic storms’ usually occur in patients
with severe or grade III tetanus thereby further intensifying the course and duration of the
disease. Occasionally, particularly in elderly patients, severe autonomic dysfunction may
complicate moderately severe (Grade II) tetanus and may prove fatal. These disturbances
always herald danger and worsen prognosis. Both sympathetic and parasympathetic
dysfunction may be present in varying degrees.
Cardiovascular complications in severe tetanus are chiefly mediated through the autonomic
nervous system.8-10 Sustained tachycardia, often greater than 170-180/min, can persist for
hours to days. Labile hypertension, moderately severe persistent systolic and diastolic
hypertension, and severe peripheral vasoconstriction with a shock-like state are all observed.
A dreaded complication is the occurrence of “autonomic storms”, characterized by episodes
of marked sinus tachycardia (HR over 150/min) with severe hypertension (blood pressure in
the range of 180-220/110-120 mm Hg), followed within minutes or hours by a sudden slowing
of the heart rate (50-90/min) with a fall in blood pressure to less than 100 mm Hg systolic,
sometimes as low as 60-70 mm Hg. Such cardiovascular instability may be a forewarning of
cardiac arrest and death. Sudden severe reflexly induced bradyrhythms may also occur
independently of autonomic storms. Severe sinus bradycardia or a slow junctional rhythm
leading at times to a cardiac arrest can be occasionally induced after tracheal suction or
during a painful venous or arterial puncture due to increased vagal activity.
Arrhythmias
Tachyarrhythmias11,12 may well represent excessive sympathetic activity. Supraventricular and
ventricular extrasystoles are frequent. Short bursts of ventricular tachycardia that revert
spontaneously or with a bolus of intravenous lignocaine are also observed. Short runs of
paroxysmal atrial tachycardia and junctional tachycardia are also seen; these almost always
revert spontaneously and rarely require specific treatment.
Disturbances in Conduction13
A prolonged PR interval is observed fairly frequently. Slow nodal or junctional rhythms are
probably related to suppression of SA node activity, secondary to increased vagal tone.
Sudden Death
Sudden cardiac arrest causing death,4,14,15 remains even today, the single most dreaded
complication of moderate and severe tetanus. Severe autonomic dysfunction contributes to
this complication. Sudden death can occur under the following circumstances:
a. Cardiac arrest can occur during periods of marked cardiovascular instability due to
fluctuating sympathetic tone.
b. Excessive vagal tone can reflexly induce bradycardia and cardiac arrest. Impaired
conduction of the cardiac impulse can do likewise.
c. An abrupt rise in core or rectal temperature (over 41°C) can cause sudden cardiovascular
collapse. This can be missed, if the rectal temperature is not monitored, as these patients
do not necessarily register a corresponding increase in the axillary temperature.
d. Sudden death can occur unrelated to severe autonomic dysfunction, as after pulmonary
embolism, myocardial infarction; unrecognized hypokalaemia or perhaps due to toxic
myocarditis.
e. Severe hypoxia due to unrelenting seizures, laryngeal spasms or prolonged periods of
apnoea, can singly or in combination lead to sudden death in patients who do not have
a secure airway and who are not on ventilator support.
f. Finally cardiac arrest may well occur for no ascertainable reason.
Other complications due to disturbances in the Autonomic Nervous System

These include drenching sweats, excessive salivation and excessive tracheobronchial secretions
producing difficult problems in respiratory care. Hyperthermia (over 41º C rectal) as also
hypothermia can occur suggesting hypothalamic involvement. We have also occasionally
encountered the syndrome of inappropriate anti-diuretic hormone secretion (SIADH), again
pointing to hypothalamic involvement.4
Haemodynamics in Tetanus with Autonomic Dysfunction

Our study in patients with severe (grade III) tetanus8 and in patients with fulminant tetanus
(grade IV) during autonomic storms revealed interesting results.
1. Severe uncomplicated tetanus is characterized by a high cardiac output, hyperdynamic
circulatory state, increased basal sympathetic tone and a slight increase in the core
temperature.
2. Autonomic dysfunction and in particular “autonomic storms” in patients with grade IV
tetanus involved both the sympathetic and parasympathetic systems.
Sympathetic disturbances could take the form of either a sustained increase in sympathetic
tone or a seizure like discharge from the sympathetic nervous system. The former leads to
tachycardia, and sustained hypertension with well-marked increase in the systemic vascular
resistance. A seizure-like discharge could lead to sudden tachycardia, sudden spikes in systemic
blood pressure, associated with a rise in central venous pressure (CVP) and a sharp rise in
the systemic vascular resistance.8 Noradrenaline concentrations are known to rise 10 fold
during these sympathetic storms.16 The rise in catecholamines is similar to that observed in
patients with pheochromocytoma.17 Cessation of seizure-like discharge leads to bradycardia,
hypotension and is associated with a fall in the CVP and also a sharp fall in systemic vascular
resistance.8 Most importantly, the cardiac index and the left heart filling pressure were unaltered
during these periods of fluctuating sympathetic activity.8 The failure of left ventricular stroke
work index to rise significantly during hypertensive spells suggested an impairment of
myocardial function.8 Whether this could be due to tetanus toxin itself or to excessive
catecholamine levels in blood is difficult to ascertain. Histological changes in the hearts of
patients dying from severe tetanus bear a strong resemblance to the changes observed in
patients with pheochromocytoma, the likely cause being persistently raised catecholamine
levels.17
Parasympathetic disturbances are evident in the marked bradycardia following the cessation
of the seizure-like sympathetic discharge, the presence of excessive salivation, excessive
tracheobronchial secretions, sudden bradyrhythms at times observed after tracheal suction
and the spontaneous occurrence of transient junctional bradyrhythms.
Management of Autonomic Dysfunction in Tetanus

A number of drugs and methods have been tried for the management of autonomic cardio-
vascular disturbances in severe tetanus. It is the general consensus that no drug singly or in
combination, has proved to be consistently effective in the control of autonomic disturbances.
Drugs that have been tried include the use of heavy sedation, intravenous propanolol or
labetalol.18 Esmolol, a short acting beta-blocker is perhaps safer than intravenous propanolol
which carries the risk of sudden death. Other reported treatments include the use of clonidine19
and intravenous atropine.20 Epidural or spinal bupavacaine has been tried,21,22 as also
intrathecal baclofen (a GABA antagonist).23 Perhaps the only treatment that holds promise is
the use of intravenous magnesium sulphate as reported by James and Manson in 1985.24
They reported two deaths in 10 patients treated with intravenous infusion of magnesium
sulphate. Each patient was given a loading dose of 70 mg/kg of magnesium sulphate as an
infusion over 5 minutes; this was followed by an infusion at a rate required to maintain a
steady state magnesium level of between 2.5 and 4 mmol/l. The magnesium infusion was
gradually reduced after two weeks or earlier if the patient’s condition so permitted.
In our haemodynamics study,8 we reported just 2 deaths among 32 patients treated in the
ICU. Fourteen of these 32 patients had Grade IV tetanus with autonomic disturbances and
there was just one death in this group. We attribute our low mortality to good overall critical
care, efficient cardiorespiratory support and, above all, to the avoidance of drugs that strongly
depress either the central or autonomic nervous system. The following two principles (based
on our study) are recommended in the treatment of autonomic cardiovascular disturbances.
Drugs acting on the cardiovascular system should be used, under constant monitoring,
and only when deemed absolutely necessary to treat grossly deranged haemodynamic functions
or parameters. Hypotensive episodes should be treated with a volume load. If this is
contraindicated (as in patients with pulmonary congestion or with ARDS) or ineffective, we
recommend catecholamine infusion titrated to maintain a systolic blood pressure just above
100-120 mm Hg. The infusion rate is reduced and if possible stopped when the systolic
pressure steadies between 110 and 120 mm Hg. Hypertensive episodes with systolic blood
pressure over 200 mm Hg or a diastolic blood pressure over 110 mm Hg are best treated with
a small oral dose of propranolol (5-10 mg) or sublingual nifedipine (2.5 mg) to which these
patients are exquisitely sensitive. A titrated dose of Esmolol, a short acting beta-blocker has
been effectively used to control a hypertensive crisis when small doses of oral propranolol
have failed. We have avoided both intravenous propranolol, or labetolol in these patients for
fear of severe hypotension, bradyrhythms, and cardiac arrest. In some patients with persistent
hypertension, the oral dose of propranolol was carefully increased but never exceeded 40 mg
thrice daily.
Bradyrhythms are treated with titrated boluses of intravenous atropine rather than by a
continuous infusion of the drug. Persistent sinus tachyrhythms over 180/minute in young
adults and over I50/minute in older patients are best controlled with a titrated dose of oral
verapamil 40 mg twice or thrice daily.
When autonomic disturbances take the form of severe persistent hypertension with cold
peripheries, a low cardiac index and a marked increase in the systemic vascular resistance,
intravenous morphine sulphate in small bolus doses of 2-6 mg titrated to produce a fall both
in the systolic blood pressure, and in the systemic vascular resistance, is of benefit. The use
of a calcium-channel blocker like nifedipine is an alternative modality of treatment in these
patients.
If the above regime fails to control autonomic cardiovascular ‘storms’, it would perhaps be
wiser to try an intravenous infusion of magnesium sulphate as advocated by James and
Manson.24 In our experience, autonomic ‘storms’ at least in patients under 50 years of age
have invariably been weathered by the simple regime outlined above. Close haemodynamic
monitoring is however necessary, if this therapeutic regime is to succeed.
Large doses of intravenous sedatives or anticonvulsants should not be given to these
patients. We generally do not use more than 60 mg/day of diazepam intravenously. This is
in striking contrast to the high doses of diazepam (200-400 mg/day intravenously), that form
the cornerstone of treatment in almost all tetanus units in this country and in the West. Spasms
are controlled effectively with titrated doses of neuroparalytic agents; diazepam being used
chiefly for sedation and relief of anxiety, and not primarily with the objective of controlling
muscular activity.
Large doses of diazepam and other drugs that strongly depress the central and autonomic
nervous systems in patients with severe autonomic dysfunction have inherent dangers; They
may either predispose to cardiac arrest (particularly after a period of hypotension or severe
bradycardia), or may render resuscitation from cardiac arrest more difficult or even impossible,
because of their strong depressant effect on medullary centers (including the vasomotor center),
particularly when used in high doses over several days or weeks.
CONCLUSION
In conclusion though the reported mortality of severe tetanus with well marked autonomic
dysfunction has been reported to be around 50%.1 close haemodynamic monitoring with
management strategies outlined above have, in our experience, led to significantly improved
results. Overall critical care with meticulous attention to detail is vital for patient survival. Not-
withstanding recent advances in intensive care, severe tetanus remains a formidable challenge
due to the intense nature of the illness, its prolonged course, and the occurrence of complications
that can involve every organ system of the body.
REFERENCES
1. Towey R. Tetanus: A review. Update in Anaesthesia, Issue 19, 2005.
2. Data from the www.nathnac.org. Travel Health Surveillance section of the Health Protection Agency,
Communicable Disease Surveillance center.
3. Albett JJ. In: Ellis M (Ed). Symposium on Tetanus in Great Britain, Leeds United Hospitals 1967.
4. Udwadia FE. Tetanus. Oxford University Press, New Delhi 1994.
5. Griggs WM, Myburg JA, Worthley LI. A prospective comparison of a percutaneous tracheostomy technique
with a standard surgical tracheostomy. Intensive Care Med 1991;17:261-3.
6. Dob DP, McLure HA, Soni N. Failed intubation and emergency percutaneous tracheostomy. Anaesthesia
1998;53:72-4
7. Griggs WM, Worthley LI, Gilligan JE, et al. A simple percutaneous tracheostomy technique. Surg Gynaecol
Obstet 1990;170:543-5.
8. Udwadia FE, et al. Hemodynamic studies during management of severe tetanus. Q J Med 1992;83:449-
60.
9. De Michele DJ, Taviera da Silva AM. Cardiovascular findings in a patient with severe tetanus. Crit Care
Med 1983;11:828-9.
10. Sutton DN, Tremlett MB, Woodcock TE, et al. Management of autonomic dysfunction in severe tetanus:
the use of magnesium sulphate and clonidine. Intensive Care Med 1990;16:75-80.
11. Trujillo MH, Castillo A, Espana J, et al. Impact of intensive care management on the prognosis of tetanus.
Analysis of 641 cases. Chest 1987;92(1):63-5.
12. Luisto M. Tetanus in Finland: diagnostic problems and complications. Ann Med 1990;22(1):15-9.
13. Drost R, Manz R, Finsterer U, et al. On cardiac involvement in tetanus. Anaesthetist 1970;19:109-12.
14. Udwadia FE, Lall A, et al. Tetanus and its complications: intensive care and management experience in
150 Indian patients. Epidemiol Infect 1987;99:675-84.
15. Ramirez MR, Sicilia LT, Aguilera JL, et al. Tetanus in adults: Study of 130 cases. Enferm Infecc Microbiol
Clin 1990;8(6):338-43.
16. Bleck TP. Tetanus: dealing with the continuing clinical challenge. J Crit Illn1987;2:41-52.
17. Rose AG. Catecholamine induced myocardial damage associated with pheochromocytoma and tetanus.
South Afr Med J 1974;48:1285-9.
18. Wesley AG, Hariparsad D, Pather M, et al. Labetolol in tetanus. The treatment of sympathetic nervous
system overactivity. Anaesthesia 1983;38:243-9.
19. Gregorakos L, Kerezoudi E, Dimopoulos G, et al. Management of blood pressure instability in severe
tetanus: the use of clonidine. Intensive Care Med 1997;23:893-5.
20. Dolar D. The use of continuous atropine infusion in the management of tetanus. Intensive Care Med
1992;18:26-31.
21. Southorn PA, Blaise GA. Treatment of tetanus-induced autonomic nervous system dysfunction with
continuous epidural blockade. Crit Care Med 1986;14:251-2.
22. Shibuya M, Sugimoto H, Sugimoto T, et al. The use of continuous spinal anaesthesia in severe tetanus
with autonomic disturbance. J Trauma 1989;29:1423-9.
23. Dressnandt J, Konstanzer A, Weinzierl FX, et al. Intrathecal baclofen in tetanus: four cases and a review
of reported cases. Intensive Care Med 1997;23:896-902.
24. James MFM, Manson EDM. The use of magnesium infusions in the management of very severe tetanus.
Intensive Care Med 1985;11:5-12.
Eighteen
Management of Haemodynamic
Instability after Scorpion Sting
HS Bawaskar, PH Bawaskar
INTRODUCTION
Scorpion sting is a public health problem. Its occurrence is routinely reported from parts of
rural India, Middle East, Latin America, Africa and Saudi Arabia.1 Envenomation following
a scorpion sting is potentially a life threatening medical emergency, which, if not treated properly
in a time may result in death of the victim. Treating doctors in developing nations have the
impression that scorpion sting means excruciating pain but most are not aware of fatality
associated with the condition. Standard medical textbooks devote only a few lines on the
subject while emphasising the role of antivenin therapy, which by itself has a limited role
once signs and symptoms of “autonomic storm” set in.
The WHO reported that the true incidence of scorpion sting is unknown because of a lack
of accurate statistical data. Many cases do not seek medical attention and those who do, are
in rural areas in developing countries where medical facilities are rudimentary. Till recently,
the WHO was not even aware of case fatality rate due to scorpion sting.2 Scorpion envenoma-
tion occurs largely in the villages in India and therefore, remains under reported and unnoticed.
SCORPION STING
In India, there are two types scorpion seen commonly—a big black colour scorpion called
Palmanmeus Gravimanus3 and small sized red coloured scorpion called Mesobuthus Tamulus
(Fig. 18.1). Mesobuthus Tamulus, the Indian red scorpion is the most lethal among 46 species
of poisonous scorpions reported from India. Poisonous scorpion species are seen in Thane,
Raigad, Ratanagiri, Sholapur, parts of Aurangabad and Nanded districts of Maharashtra;
Kurnool, Anantapur, Hyderabad and Tirupati from Andhra Pradesh; Bellary, Bagalkot, Bijapur
areas of Karnataka; Saurashtra and Kutch areas of Gujarat as well as Chennai and Pondicherry.
Fatality rates as high as 30% due to refractory pulmonary edema and multi-organ failure due
to scorpion sting have been reported from Maharashtra, Chennai, Andhra Pradesh and
Pondicherry.
Management of Haemodynamic Instability after Scorpion Sting 185
Fig. 18.1: Indian red scorpion Mesobuthus Tamulus
Scorpions
Scorpions are venomous arthropods in the class Archnida. Worldwide, there are around 1400
species of scorpions but only 45-50 of these are potentially lethal to humans. Scorpions are
well-adapted arthropods that are able to survive heat, drought and desert conditions including
freezing temperatures for weeks. Even more remarkable is the fact that scorpions can withstand
starvation for months or total immersion in water for days. This allows scorpions to survive
at times of adverse environmental conditions and gives the species an unbroken continuity.
Scorpions are strictly carnivorous, feeding for the most part on other insects. They are viviparous
and give birth to young ones directly. Sometimes, the mother tries to eat the young, but more
often, the young ones nibble the mother to death. Scorpion seek shelter under the bark of
trees, dry firewood, cow dung, in piles of bricks, paddy husk, beddings, loose tiles of hut
dwellings and at times in the shoe left over night or even in pockets of trousers and shirts left
unattended for sometime. In a tropical country like India, birds occasionally bring small
scorpions along with dry grass for building a nest in a pucca house. Farmers and farm labourers
often get stung while handling paddy husk or when harvesting grass in the months of September
to November or while preparing the land for sowing in the month of May and June. Travelers
get stung while walking barefoot in the desert or in farmland. Children are often stung in their
sleep by a scorpion falling off from the loose tiles of a hut roof.
Venom
Tail end bulb of scorpion contains two telson glands, which actively secrete the venom at the
time of sting. The venom finds its way into the victim or prey through the sharp stinger in the
tail. Venom is not stored but secreted at the time of stinging. Scorpion venom is a cocktail of
various active substances. Of these, neurotoxins are the most important.3 Neurotoxins secreted
by the scorpion consist of different small size proteins with sodium and potassium cations,
which interfere with neurotransmission in the victim. Peptide neurotoxin is venom that opens
the Na+ channels (β toxin). Sodium is primarily an extracellular ion responsible for maintaining
the electric voltage difference across the cell membrane. Venom depolarizes the cell membrane
and, in addition, it also inhibits the deactivation of Na+ channels (α toxin). There is, as a
result, a massive release of endogenous catecholamine into the circulation due to delayed
inactivation of Na+ neuronal channels by venom.1,3 Thus, the scorpion venom causes
autonomic storm by stimulating sympathetic and parasympathetic nervous system. Mesobuthus
Tamulus venom is a potent neuronal Na+ channel activator. Iberiotoxin isolated from
Mesobuthus Tamulus venom inhibits the calcium dependent K+ channels.4
PATHOPHYSIOLOGY
Delayed inactivation of neuronal Na+ channels and inhibition of calcium dependent K+
channels result in a sudden outpouring of endogenous catecholamines into the circulation.
This forms the basis of the “autonomic storm” seen in patients with severe envenomation
following a scorpion sting. Stimulation of a adrenergic receptors plays an important role in
the pathogenesis of pulmonary oedema5. Accumulation of Ca++ ions in the myocardium
due to rise in circulating catecholamine, coronary spasm,6 accompanied with increased oxygen
requirement resulting in systolic and diastolic myocardial dysfunction. Rise in renin and
angiotensin II, which is injurious to myocardium. Angiotensin II activates the alpha-1 receptors
in the myocardium. Venom reaches the myocardium within 3-5 minute of sting. There is no
evidence of a direct effect of venom on the myocardium. Reversible scorpion cardiomyopathy
seen after a scorpion sting is usually attributed to the catecholamine excess in the acute phase.
In our series that included follow-up for 10 years, not a single case developed signs and
symptoms of long standing cardiomyopathy. Pulmonary oedema is often secondary to
myocardial dysfunction. However, acute lung injury pattern or adult respiratory distress
syndrome (ARDS) attributed to secretory pulmonary oedema has been reported from Brazil.7
Venom induced liberation of IL-6, tumour necrosis factor (TNF), kallikrenin and platelets
activating factor (PAF) are believed to contribute to the pathogenesis of pulmonary oedema.8
Autopsy of fatal cases following a scorpion sting shows fluid accumulation in alveoli with
leukocytes infiltration and hyaline membranes formation in the lung along with contraction
band necrosis in the myocardium. The histological pattern is similar to cases of pheochroma-
cytoma or acute stress induced apical ballooning syndrome.9,10
Hypertension and not the hypotension is a dose dependent phenomenon. Venom induces
transient parasympathetic and prolonged sympathetic stimulation. Sweating, priapism,
bradycardia, ventricular ectopy and salivation suggest ongoing action of circuiting venom.
CLINICAL MANIFESTATIONS
Clinical manifestations are due to stimulation of the autonomic nervous system (both
parasympathetic and sympathetic). Close to 80% of victims complain of severe and excruciating
pain at the site of sting. This radiates along the corresponding dermatomes and is accompanied
by local oedema, sweating, and muscular fasciculations at the site. Sometimes, pain is so
severe; it results in local muscle spasm. Envenomation by an Indian red scorpion evokes an
“autonomic storm” due to excessive stimulation or delayed closure of neuronal Na+ channels.
This is characterized by transient parasympathetic over activity manifested as vomiting, profuse
sweating, salivation, bradycardia and priapism in males. A phase of sustained sympathetic
stimulation manifest as hypertension, tachycardia, non-sustained runs of ventricular tachy-
cardia, cold extremities, pulmonary edema and shock.10 Vomiting, sweating, hypotension
and cool extremities are premonitory signs of systemic envenomation.11
Systemic involvement has been reported in up to 15% of a case series of 3546 scorpion
sting cases from three districts in the Konkan region of peninsular India.12 Cardiovascular
involvement has been described in reports of envenomation by Mesobuthus Tamulus, Leiurus
Quinquestriatus, Androctonus Mauretanicus , Buthgus Occipitus , Androctonus Crassicauda,
Tityus zulianus, Titus Serrulatus from India, Israel, Saudi Arabia, Brazil, South Africa and
Mexico. It is important to note that irrespective of the species of poisonous scorpion involved,
similar cardiovascular manifestations have been reported.1
Grading of Systemic Features

Clinical manifestations depend upon the amount of time elapsed from sting to presentation
at a healthcare facility.13 According to clinical manifestations, victims are graded by increasing
severity of signs and symptoms from grade 1 to 5.
Grade 1: Severe excruciating local pain at the site of sting, which radiates along with
corresponding dermatomes without any signs and symptoms of systemic involvement.
Grade 2: Local signs associated with profuse sweating all over body, salivation, hyper or
hypotension, cold extremities, ventricular premature contractions and /or bradycardia.
Grade 3: Pulmonary oedema with hypertension or hypotension and cold extremities.
Grade 4: Warm extremities, tachycardia, hypotension, pulmonary oedema, air hunger and
shock.
Grade 5: Death.
Hypertension
45% victims with systemic involvement have a raised blood pressure 140/90–230/160 mmHg
associated with bradycardia (42-60 /min) soon after the sting. Children look anxious, agitated,
confused with propped up eyes and face. Adults complain of headache, paraesthesia around
face and limbs and report a chest discomfort as well. Raised blood pressure is seen within 15
minutes and up to 11 hours of a sting. There is parasternal systolic lift due to vigorous
myocardial contraction due to raised catecholamines. A loud summation gallop or transient
systolic murmur due to mitral regurgitation as a result of papillary muscle dysfunction may
also be seen.
Hypotension
Hypotension soon after the sting is due to loss of fluid due to vomiting, salivation and sweating.
This is further aggravated by restriction of fluid intake advocated by relatives or local healers.
Bradycardia, premature atrial and ventricular ectopic beats and runs of non-sustained
ventricular tachycardia contribute to hypotension. Subsequent hypotension is due to cardiac
failure with low output. Asymptomatic hypotension with bradycardia observed after 24-36
hours of hospitalisation is due to depletion of catecholamine stores and is a self-limiting
phenomenon. This does not occur, if victim reports earlier and is treated with scorpion
antivenin.
Warm shock is a complex condition seen in pediatric cases presenting late after irrational
management at the periphery by excessive fluid, steroids, anti-histamines, digoxin, or atropine.
Characteristic features are hypotension, warm extremities, tachypnoea, air hunger and a rapid
heart rate (up to 200 beats per min). There is cadaver pallor due to irreversible severe persistent
tissue anoxia.14
Cardiac Arrhythmias
Sinus bradycardia, ventricular premature contractions, runs of ventricular tachycardia,
ventricular fibrialltion, supraventricular tachycardia at 110-240 beats per minute, delay in
atrioventricular conduction, low voltage pattern ECG, junctional rhythm are abnormalities
that have been described with scorpion sting. If not managed properly, prolonged QTc with
torsades in a recovering hospitalised victim can result in sudden death.15
Pulmonary Oedema
Pulmonary oedema occurs in 27-30% of severe scorpion sting cases and may or may not be
associated with respiratory failure. Pulmonary oedema develops within 30 minute to up to
10 hours after the sting. Pulmonary oedema is characterized by dyspnoea, othopnoea,
intractable cough, reddish expectoration, bilateral moist rales, loud summation gallop with
transient systolic murmur, cold extremities, low volume pulse with hypotension. The early
manifestations are profuse sweating associated with raised blood pressure 160/120 to 240/
160 mmHg.
8% cases develop severe pulmonary edema, requiring acute medical intervention. A pattern
suggestive of acute myocardial infarction can be seen on ECG.
Haemodynamic Study
It is difficult to perform invasive haemodynamics in severely ill, scorpion sting patients. Karnad16
have studied haemodynamic pattern in victim with history of a Mesobuthus Tamulus
envenoming from Western Maharashtra. Mild envenomation causes severe vasoconstriction
and hypertension with pulmonary edema. In severe scorpion envenomation, predominant
left ventricular dysfunction with normal systemic vascular resistance and pulmonary oedema
is often seen. Similar pattern of haemodynamic changes has been reported from Tunisia,
Brazil and Israel.17,18
Echocardiography
Poor global myocardial contractility, poor ejection fraction, decreased systolic left ventricular
performance; mitral incompetence and abnormal diastolic filling can be demonstrated to persist
for upto 4 weeks.19 At times, apical ballooning of the heart similar to that seen in stress-
induced cardiomypathy can be observed.20 Echocardiographically, there is good correlation
between clinical improvement and return of left ventricular contractility towards normal.
MANAGEMENT
No scorpion sting should be taken lightly. On the basis of known pathophysiology, therapeutic
effort should be directed against the venom, and the resulting clinical manifestations. Venom
action on neurotransmitters in the body is rapid and potentially lethal. Although majority of
victims are active and healthy before the sting, they can succumb to the venom suddenly;
hence, an observation period of 24 hours is always recommended.
General Measures
Scorpion sting affects the extremities - fingers, toes, web spaces, sole and palm. Because of
pain, victim is more anxious, restless and uncomfortable. The stung part is more sensitive to
touch. Reassurance to the victim and relative reduces anxiety, which is also contributing factor.
Profound loss of fluids due to sweating, salivation, vomiting can occur and this should be
replaced with crystalloids.
Regular monitoring of temperature, pulse, respiration, blood pressure, auscultation of heart
and lungs and perfusion of extremities is recommended. Certain therapeutic agents should
be avoided, if possible, in scorpion sting. These include digoxin, diuretics,21 atropine, beta-
blockers, nifedipine, adrenaline and anti- histamines.
Local Pain
Pain at the site of the sting should be managed with locally active agents. Table 18.1 summarises
various regimens that can be tried. A simple tap at the site of the sting normally evokes severe
pain but this tap sign is not elicited in patients with severe vasoconstriction.
Table 18.1: Result of various regimen used for local pain
Regimen N Recovery in hrs (mean)
Local xylocaine (multiple doses) 30 12-72 (21)

Xylocaine + NSAID + diazepam 27 6-15 (9)
NSAID + Diazepam + cold therapy* 145 2-11 (5.2)
Dehydro emetine 19 2-8 (4)
*Cold therapy- cloth soaked in cool water or water from mud pot.
Scorpion Antivenin (SAV)

This is a specific antidote to circulating venom. SAV is species specific; thus antivenin prepared
from the Mesobuthus Tamulus venom cannot be used for envenomation by other species.
There are 46 poisonous species seen in India. Hence, vigilance and careful monitoring should
be continued even after giving antivenin the victim. Scorpion venom causes transient para-
sympathetic and prolonged sympathetic stimulation. The presence of parasympathetic features
indicates early activity of the venom and its presence in circulation. At this stage, antivenin
is effective and should be given along with prazosin. If the victim attends the hospital within
an hour of the sting, recovery time is faster with scorpion antivenin given early. Antivenin is
given diluted in normal saline over 30 minutes. If time permits, a test dose should be given
to rule out anaphylaxis. Turbid antivenin should not be used, as it is indicates the precipitation
of a protein, which may cause a severe systemic reaction. Scorpion antivenin is available
from Bharat Serum Laboratories and Haffkine Institute, Mumbai.22
Prazosin
Prazosin is a post-synaptic a1 blocker and is ideally suited to counteract the sympathetic over
activity seen with scorpion sting. It reduces preload, left ventricular impedance without changing
the heart rate; it is thus a balanced vasodilator. Prazosin inhibits the phospodiesterase enzyme
system resulting in an accumulation of cGMP, which enhances insulin secretion. In victims
there is already a high circulating level of potassium following an efflux from the liver cells
into the circulation due to catecholamine stimulation and an associated high blood sugar
level as well. Prazosin, by raising insulin levels, functions as a GIK (glucose + Potassium+
insulin) drip to control metabolic problems in circulation. Hence prazosin is not only a
physiological antidote but a poor man’s antivenin. Even after giving full dose of SAV, many
victims develop severe cardiovascular manifestations, which improve only after oral prazosin
is given. In agitated children prazosin can be given by a feeding tube or given by the mother
to the child with the warning how the vital and life saving the dose is.23-27
The recommended dose of prazosin is 500 micrograms in children and 1000 micrograms
in adults, to be repeated three hourly till extremities are cold then 6 hourly. One should be
careful that prazosin has reached the circulation and it is acting by noting the clinical
improvement. Since the advent of prazosin, fatality due to scorpion sting is dropped to <1%
from 29% (during pre prazosin era) in endemic regions of scorpion sting in India - Pondicherry,
Bellary, Karnataka, Andhra Pradesh and Saurashtra.
Special Situations
Mangement of massive pulmonary oedema requires intravenous Sodium nitroprusside (SNP)
3-5 microg/kg/min for up to 48 hours. Simultaneous use of dobutamine may improve prognosis
but this has not been studied. Glyceryl Nitrate (GTN) can be used, if SNP is not available.
While using SNP, one should be aware of cyanide toxicity, which can be avoided by giving
injectable vitamins (B-12). Once a treating doctor thinks of SNP he should start drip rather
than wait for advice from seniors, who might have not treated a case before. Non-invasive
ventilation and close monitoring are also required for management of pulmonary oedema.
CONCLUSION
Scorpion envenoming is an acute life threatening time limiting medical emergency, if not
treated properly in a time may result in death of a victim. Majority of treating doctor and even
a layperson has a idea that scorpion sting means severe excruciating pain only and not aware
of fatality, hence many time in early stage victim signs and symptoms are neglected. Even the
international medicine text book included few lines on this subjects and emphasis is given to
antivenin therapy which is no more benefits once the victim having signs and symptoms of
autonomic storm. Antivenin is an adjuvant with prazosin within 1-2 hours of sting is much
helpful for recovery. Being rural problem remained neglected for many years. Due to scarcity
and poverty many victim go uninvestigated. Even at tertiary care institute majority of treating
expert are not aware of severity of scorpion sting. Recently details investigation and its rational
approach is studied and reported from India and abroad.
REFERENCES
1. Ismail M. The scorpion envenoming syndrome. Toxicon 1995;33:825-58.
2. WHO. Rabies and envenoming: a neglected public health issue. Report of a consultative meeting WHO
Geneva 10th January 2007 page 1-32.
3. Bawaskar HS. Scorpion sting. In Shah SN (Ed). API Text Book of Medicine, 8 thedition, Assoc Physicians
India, Mumbai, 2008;1520-23.
4. Gwee MCE, Nirthanan S, Khoo H, Gopalakrishnakone P, Kini MR, Cheath IS. Autonomic effects of some
scorpion venoms and toxins. Clin Exp Pharmacol Physiol 2002;29:795-801.
5. Freire-Maia L, Pinto GI and Franco I. Mechanism of the cardiovascular effects produced by purified scorpion
toxin in the rat. J Pharmacol Exp Ther 1974;188:207-13.
6. Margulis, Sofer S, Zalstein E, Zunker, et al. Abnormal coronary perfusion in experimental scorpion
envenomation. Toxicon 1994;32;1675-8.
7. Amaral CFS, de-Rezende NA, Freire-Maia L. Acute pulmonary edema after Tityus Serralatus scorpion
sting in children. Am J Cardiol 1993;71:242-5.
8. Sofer S, Gueron M, White RM, Lifshitz M, Apte RN. Interleukin -6 release following scorpion sting in
children. Toxicol 1996;34:489-92.
9. Wittstein IS. Apical–Ballooning syndrome. Lancet 2007;370:545-7.

10. Benvenuti LA, Dpouetts KV, Cardoso JLC. Myocardial necrosis after envenomation by the scorpion Tityus
Serrulatus. Trans Roy Soc Trop Med Hyg 2002;96:275-6.
11. Bawaskar HS. Diagnostic cardiac premonitory signs and symptoms of red scorpion sting. Lancet 1982;
ii:552-4.
12. Bawaskar HS, Bawaskar PH. Peripheral doctors form backbone for management of acute life threatening
medical emergency evoked due to envenoming by Indian red scorpion: Mesobuthus Tamulus. Bombay
Hospital J 1997;39:710-4.
13. Bawaskar HS, Bawaskar PH. Sting by red scorpions (Mesobuthus Tamulus) in Maharashtra state: a clinical
study. Trans Roy Soc Trop Med Hyg 1989;83:858-60.
14. Bawaskar HS, Bawaskar PH. Cardiovascular manifestations of severe scorpion sting in India (review of
34 children). Ann Trop Pediatr 1991;11:381-7.
15. Bawaskar HS, Bawaskar PH. Management of the cardiovascular manifestations of poisoning by the Indian
red scorpion (Mesobuthus Tamulus). British Heart J 1992;68:478-80.
16. Karnad DR. Haemodynamic pattern in patients with scorpion envenomation. Heart 1998;79:485-9.
17. Abroug F, Boujdaria R, Belghith MB, Nouira S, Bouchoucha S. Cardiac dysfunction and pulmonary edema
following scorpion envenoming. Chest 1991;100:1057-9.
18. Abroug F, Ayari M, Nouira S, Gamra H, Boujdaria R, Elatrous S, Farhat MB, Bouchoucha S. Assessment
of left ventricular function in severe scorpion envenomation: combined hemodynamic and Echo-doppler
study. Intensive Care Med. 1995;21:629-35.
19. Gueron M, Margulis G, Sofer S. Echocardiographic and radionuclide angiographic observations following
scorpion envenomation by Leiurus quinquestriatus. Toxicon 1990;28:1005-9.
20. Rajasekhar D, Mohan A. Clinical and echocardiographic finding in patients with myocardial toxicity due
to scorpion sting. National Medical J India 2004;17:307-9.
21. Karnad DR, Deo AM, Apte N, Lohe AS, Thatte S, Tilve GH. Captopril for correcting diuretic induced
hypotension in pulmonary oedema after scorpion sting. BMJ 1989;298:1430-1.
22. Bawaskar HS, Bawaskar PH. Utility of scorpion antivenin Vs Prazosin in the management of severe
Mesobuthus Tamulus (Indian red scorpion) envenoming at rural setting. J Assoc Physicians India 2007;
55:14-21.
23. Bawaskar HS, Bawaskar PH. Severe envenoming by the Indian red scorpion Mesobuthus Tamulus : Use
of prazosin therapy. Q J Med 1996;89:701-4.
24. Bawaskar HS, Bawaskar PH. Prazosin in management of cardiovascular manifestations of scorpion sting.
Lancet 1986;ii:510-11.
25. Mahadevan S, Choudhury P, Puri RK, Srinivasan S. Scorpion envenomation and the role of lytic cocktail
in its management. Indian J Pediatr 1981;48:757-61.
26. Gupta V. Prazosin: a pharmacological antidote for scorpion envenomation. J Trop Pediatr 2006;52:150-
1.
27. Al-Asmari AK, Al-Seif AA, Hassen MA, Abdul-Makssod NA. Role of prazosin on cardiovascular manifes-
tations and pulmonary oedema following severe scorpion sting in Saudi Arabia. Saudi Med J 2008;29:
1296-9.
Heat Stroke 193
Nineteen
Heat Stroke
Kishore Pichamuthu, JV Peter
“It does not take long either to boil an egg or to cook neurons”1
INTRODUCTION
Heat stroke is part of the spectrum of heat related illness. Heat illnesses or thermic stress
syndromes as they are also known, result from an uncontrolled increase in the body’s core
temperature that occurs when exogenous heat or endogenous heat production overwhelms
thermoregulatory mechanisms. Heat stress, which is a mild form of heat related illness, is a
perceived discomfort or physiological strain, especially during physical work in a hot
environment. Heat oedema, heat cramps and tetany are part of heat stress illnesses. Heat
exhaustion is a moderately severe form of thermic stress due to salt or water depletion that
results from exposure to high environmental temperatures or strenuous exercise. The signs
and symptoms include intense thirst, weakness, discomfort, anxiety, dizziness, fainting and
headache. The core temperature is between 37°C and 40°C.2
Heat stroke is the most severe form of thermic stress and is defined as elevation of the core
body temperature above 40°C associated with central nervous system (CNS) dysfunction
such as delirium, convulsions and coma. Heat stroke can either result from exposure to
environmental heat (Classic Heat Stroke) or strenuous physical exercise (Exertional Heat
Stroke). From a pathophysiological perspective, it can be defined as a form of hyperthermia
associated with a systemic inflammatory response leading to a syndrome of multi-organ
dysfunction in which encephalopathy predominates.
Classical heat stroke or non-exertional heat stroke (NEHS) results from a failure of the
body’s heat dissipating mechanisms and is usually seen in the very young, the elderly,
chronically ill persons, those with reduced cardiovascular reserve and those with impaired
sweating. Exertional heat stroke however, results from increased heat production, which
overwhelms the body’s ability to dissipate heat. Exertional heat stroke affects young, healthy
individuals who engage in strenuous physical activity, particularly manual labourers, sports,
military and firefighting personnel. The features of the two forms of heat stroke are compared
in Table 19.13 whilst the risk factors for heat stroke are listed in Table 19.2.4
Table 19.1: Common and distinctive features of classic and exertional heat stroke3
Features Classic Exertional
Common
H y p e rthermia > 40°C > 40°C
C e n t r a l n e rvous system alteration Delirium, convulsion Delirium, convulsion
Hypotension 20%–30% Unknown
Distinctive
Age Elderly Young
Skin Hot, dry Hot, profuse sweating
Rhabdomyolysis Mild/moderate Severe
Renal failure Uncommon Common
Lactic acidosis Mild/moderate Severe
Glycaemia Hyperglycaemia Hypoglycemia
Disseminated intravascular coagulation Mild/moderate Severe
INCIDENCE
The true incidence of this common disorder is not accurately known. This is because of under
diagnosis, under reporting and a wide variation in the diagnostic criteria for a heat related
death. The incidence varies seasonally – being common in summer months and particularly
during heat waves. The geographical location also influences the incidence of heat stroke.
Although heat stroke, particularly the exertional variety has been reported from all over the
globe, it is commoner in the tropics where summer daytime environmental temperatures of
> 40°C are not uncommon. One study reported an incidence of 17.6–26.5 per 100,000
Table 19.2: Risk factors for developing heat stroke4
Classical heat stroke Exertional heat stroke
• Inability to control environment and fluid intake • Preceding viral infection

— Infants • Dehydration
— Elderly • Fatigue
— Chronically ill, bed bound individuals • Obesity
• Reduced cardiovascular reserves • Lack of sleep
— Elderly • Poor physical fitness
— Heart failure • Lack of acclimatization
• Impaired sweating • Intense exercise
— Elderly • Drug use
— Skin disease — Cocaine
• Dehydration — Amphetamine
• Lack of acclimatization • Status epilepticus
• Anti-cholinergic drugs
— Tricyclic antidepressants
— Anti-parkinsonian drugs
— Anti-psychotics
Heat Stroke 195
population during a heat wave in the United States,5 while estimates from Saudi Arabia during
the Haj (annual pilgrimage to Mecca) ranged from 22 to 250 per 100,000 population with
an overall mortality of 50%.2
There is scant data on the incidence of this disorder in India. The earliest report of heat
stroke in India was in 1873, when 123 out of 186 prisoners under the British Raj in the
“Black Hole of Calcutta’ died in one night.6 Deaths due to heat related illness are grossly
under diagnosed and under reported in India. Although the press reported the loss of 1600
lives during a heat wave in 2003, official statistics recorded only 807 heat stroke deaths for
the entire year.7, 8 It is most likely that heat related deaths probably occur in many thousands,
particularly in the Gangetic plains and the Deccan peninsula where it assumes a major public
health importance every summer.
With trends towards higher environmental temperatures globally due to global warming
and a predicted increase in heat waves even in temperate climates, heat related fatalities are
likely to increase and contribute significantly to preventable mortality.
PATHOPHYSIOLOGY
In the last few years, laboratory studies have elucidated the mechanisms for the multi-organ
injury seen in heat stroke and enhanced our understanding of the disorder at the bedside.
Extreme heat stress triggers three pathophysiological processes, namely thermoregulatory
failure, exaggerated acute phase response and altered expressions of heat shock proteins.
The inflammatory and coagulation responses to these processes and the direct cytotoxic effects
of heat are responsible for the multi-organ failure that ensues.2
Normally, heat is gained by the body from the environment as well as endogenously from
metabolism. Heat loss occurs from the body in 4 ways. Radiation accounts for 60% of the
body’s heat loss followed by 30% due to evaporative losses. Convective and conductive losses
make up the rest. When the ambient temperature is higher than 37°C, heat cannot be lost by
radiation or conduction. At these temperatures, evaporative losses through sweating become
the only mechanism of heat loss. The evaporation of 1.7 ml of sweat dissipates 1 kcal of heat.
However, the efficacy of sweating markedly reduces as humidity increases. Increasing the
movement of air around the body overcomes this to some extent. Most parts of India are hot
and humid and many do not have access to fans or air conditioning, creating a high potential
for heat strokes. Further, in overcrowded environments, an inability to dissipate heat effectively
combined with absorption of heat emitted by others in the crowd can lead to a heat stroke
situation even when the ambient temperature is not very high. This phenomenon, called the
“penguin effect,” is derived from a heat conserving method used by Antarctic penguins termed
“huddling”.9
Although sweating is an efficient way to lose heat, it results in significant salt and water
losses, which can be as high as 2L per hour. Additionally, in an attempt to re-distribute heat
from the core to the periphery, massive cutaneous vasodilatation occurs, thereby shunting
blood away from the splanchnic circulation. This can lead to gut ischaemia. There is also a
secondary increase in cardiac output due to vasodilatation. Conditions or medications that
impair sweating or an inability to increase cardiac output due to dehydration, cardiovascular
disease as well as medications contribute to thermoregulatory failure.
Heat stress triggers an acute phase response, releasing cytokines, predominantly TNF-α,
Interleukins 1, 6 and 10 from muscle, monocytes and endothelium. This results in a systemic
inflammatory response syndrome. The gastrointestinal tract fuels this inflammatory response.
The increased metabolic demands coupled with splanchnic hypoperfusion secondary to
shunting of blood leads to ischemia and hypoxia of the gut mucosa resulting in increased
intestinal permeability with endotoxins from the gut entering the circulation. This enhances
release of inflammatory cytokines and vasodilators, aggravating organ dysfunction and
precipitating hypotension. The subsequent endothelial activation and inhibition of fibrinolysis
leads to disseminated intravascular coagulation (DIC) that contributes to organ dysfunction.
It is evident that the pathophysiology of organ dysfunction in heat stroke bears similarities
with that of sepsis.
Heat also has a direct cytotoxic effect. It causes denaturation of all cellular proteins leading
to apoptosis. In addition, exposure to heat triggers the production of heat shock proteins,
which protect cellular proteins from further heat damage, preventing cell death. High
concentrations of heat shock proteins attenuate cerebral and cardiovascular failure. Ageing,
lack of acclimatization and genetic predisposition can lead to reduced expression of heat
shock proteins favoring progression to organ failure in heat stroke.2 Organs differ in their
susceptibility to heat damage, with the cerebellum being the most sensitive and the pancreas
being the most resilient to thermic stress.
MANIFESTATIONS
The presence of a core temperature higher than 40°C and CNS dysfunction are essential
criteria to diagnose heat stroke. Typically patients with classical heat stroke have dry and
warm skin, whilst those with exertional heat stroke are sweaty. However, some patients with
classical heat stroke may continue to sweat at the time of presentation, as sweating is abolished
late in its course. The temperature at presentation may sometimes be less than 40°C, particularly
in the exertional variant as cooling may have occurred after physical activity ceased.
CNS dysfunction may range from irritability and delirium to confusion and coma. Seizures
may occur, but are more likely during cooling. Cerebellar signs and dystonia are not uncommon.
It is important to keep in mind that many of these CNS manifestations may also be a result
of dyselectrolytemia, dysglycaemia, uraemia, fulminant hepatic failure or intracranial haemor-
rhage secondary to the coagulopathy.
Tachycardia and volume depletion are almost universal. Although blood pressure is usually
normal at presentation, most patients have a hyperdynamic state, with a wide pulse pressure
and a high cardiac output. Hypotension is common and usually develops > 2 hours after the
Heat Stroke 197
onset of hyperpyrexia. Less commonly, in the presence of pre-existing cardiac disease or

direct heat induced myocardial damage, a hypodynamic state with clammy extremities and
a low cardiac output may occur. This can be associated with a significant rise in serum troponin.
ECG abnormalities such as supra ventricular and ventricular arrhythmias and S-T segment
changes resembling acute myocardial infarction can also be encountered.10
All patients with heat stroke have a centrally driven hyperventilation with a respiratory
alkalosis. Hypoxia may result from aspiration, pulmonary edema or acute respiratory distress
syndrome (ARDS).
Gastrointestinal haemorrhage is common because of the loss of mucosal integrity and
coagulopathy. Heat damage to the liver usually manifests as markedly elevated transaminases
(>1000 U/L) and hyperbilirubinemia. This can progress to a fulminant hepatic failure, with
the ensuing encephalopathy and coagulopathy. Hypophosphatemia at admission can predict
development of acute liver failure.11 Thrombocytopenia, prolonged coagulation tests and
defibrination as part of DIC is very common. Other laboratory abnormalities such as leuco-
cytosis, haemo-concentration, hypocalcaemia, hypophosphatemia and hypokalaemia may
be observed. Varying degrees of rhabdomyolysis with serum creatinine kinase levels >1000
U/L can occur. Acute renal failure due to rhabdomyolysis, dehydration, shock and DIC is
seen in 50% of patients and can also be a result of heat damage of tubular cells.
High anion gap acidosis is the commonest abnormality on the arterial blood gas followed
by respiratory alkalosis. An elevated lactate (>3 mmol/dl) is almost always present in exertional
heat stroke. While such an elevation is unusual in classical heat stroke, when present, portends
a poor prognosis.12
Much of the organ failure takes time to manifest and it is not uncommon to encounter
heat stroke victims who are normotensive, non-oliguric with just mild irritability on presentation,
who go on to develop shock, renal failure, ARDS and DIC over the next few hours despite
adequate cooling. It is important to keep even apparently well patients monitored for 48
hours.4
The overall mortality of patients with heatstroke is 50 to 70% with 85% of patients
manifesting failure of multiple organs.2,13 However, it is important to note that the mortality
is markedly reduced with early initiation of cooling measures. Mortality can be reduced to as
low as 10% if cooling is initiated in the first hour in contrast with a mortality of 80% when
cooling is delayed beyond 2 hours. Some investigators have therefore suggested that the
term “golden hour” that is often used in trauma to describe the period until the initiation of
treatment is also relevant for the cooling of heat stroke patients. If cooling of heat stroke
patients is initiated within this limited “window time period” the prognosis may be favorably
influenced.14
Prognostically, metabolic acidosis, elevated creatinine kinase (>1000 U/L) and
transaminases elevated to more than twice normal have been found to predict multi-organ
failure.13 In addition, being bed bound, dependent on others for self care and previous
psychiatric, cardiovascular and pulmonary illnesses significantly increase the risk of death.15
MANAGEMENT
Heat stroke is a medical emergency. Prompt recognition, rapid cooling and support of failing
organ systems are the pillars of heat stroke management.16 Prompt recognition of heat stroke
is crucial as early therapy, as mentioned above, favourably impacts outcome. The most
common reason why the diagnosis may not be made early is due to a low clinical index of
suspicion. When a patient presents to the emergency services with high temperature and an
altered conscious state, valuable time is lost in trying to look for CNS infection, sepsis, malaria
or neuroleptic malignant syndrome and waiting for CSF studies, brain imaging and blood
tests.
Another group of patients in whom the diagnosis of heat stroke may go unconsidered are
those already admitted to hospital for an unrelated problem. Ambient temperatures in non
air-conditioned hospital wards in tropical countries are frequently above 37°C in the summer
months. Patients admitted with a febrile illness, elderly patients, those with cardiac, neurological
or psychiatric disorders and patients with organophosphate poisoning on atropine are
particularly at risk of developing a “nosocomial” heat stroke. Thus, a high index of suspicion
for heatstroke in the summer months, measuring core body temperature for all patients
presenting with high fever and treating all patients with a temperature > 40°C as heatstroke,
by aggressively cooling them, unless proved otherwise, are ways to prevent/reduce this
misdiagnosis.
Rapid and effective cooling of the patient is the most important aspect of therapy. The
aim of cooling is to bring the core body temperature down to 38.9°C as rapidly as possible.
This is done after measures to stabilize the airway, breathing and circulation have been initiated.
A target temperature of 38.9°C and not 37°C has been chosen to avoid deleterious overshoot
brain hypothermia. The rapidity with which this target temperature is achieved also influences
outcomes. One study showed a reduction in mortality to 15% when the target temperature
was reached within 60 minutes whilst the mortality was 33% when it took longer.17 Ideal
cooling rates are between 0.1°C/min and 0.2°C/min.
A number of methods have been tried to achieve this objective and these are enumerated
in Table 19.3.
Conductive Cooling Methods

Iced water immersion: After stabilizing the airway, breathing and circulation, the patients clothes
are removed, water proof ECG electrodes and a core temperature probe (oesophageal, urinary
or rectal) are placed, and the patient is immersed in a shallow tub filled with water and enough
crushed ice to form a slush. The vitals signs and temperature are closely monitored. Placement
of a Foley’s catheter to monitor urine output is helpful. Cooling is continued till the core
temperature reaches 38.9°C. Excessive perceived discomfort or shivering mandates removal
of the patient from the iced water. Although chlorpromazine has been advocated to abolish
mild shivering, some advice against it as it may worsen hyperpyrexia, lower seizure threshold
and contribute to hepatotoxicity.18 Benzodiazepines may be used to ameliorate shivering.
Heat Stroke 199
Table 19.3: Classification of cooling methods
• Non-invasive
— Conductive
– Whole body immersion in iced water
– Limb immersion in iced water
– Ice packs in neck, axilla, groin
— Evaporative
– Wet sheets
– Water sprays and fans
– Body cooling unit
• Invasive
— Iced gastric, bladder and peritoneal lavage
— Endovascular cooling probes
— Haemodialysis
— Cardiopulmonary bypass
• Chemically assisted
— Dantrolene
In the only controlled clinical trial of iced water immersion, 21 exertional heatstroke patients
were treated with either iced water immersion or wet towels over the body. The rate of cooling
was 0.2°C/min in the iced water immersion arm vs. 0.11°C/min in the wet towel arm.19
However, this was a non-randomized comparative study and there could have been a significant
allocation bias. In a case series, 28 patients with classical heatstroke were cooled with immersion
or massaged with ice cubes if immersion was not tolerated. Cooling to 38.9°C was achieved
within 45 minutes in all patients.12 Two studies done in classical heatstroke are summarised
in Table 19.4.3
Theoretically, this technique is disadvantageous as it may induce peripheral vasoconstriction
and reduce the rate of heat transfer. Although not proven experimentally, iced water immersion
may also induce shivering and potentially increase the core temperature. Further, there are
several logistic drawbacks of this method. The immersion tub must be available and filling it
with iced water takes time to organize. Monitoring of vital signs, cardiopulmonary resuscitation
and defibrillation of the patient become cumbersome or even impossible after immersion.
Immersion causes a lot of discomfort to conscious patients and to the medical personnel
handling the patient in the iced water. Very unpleasant and unhygienic conditions may arise
if vomiting or diarrhoea occurs.20
Limb immersion: This limited version of whole body immersion consists of immersion of the
lower and upper limbs in iced water to the extent possible. This can be considered in field
situations, e.g. at the scene of an endurance event.21 This method has not been evaluated in
heatstroke patients.
Ice packs: Ice or cooling packs are placed over areas where large vessels are in close proximity
to the skin, viz. over the neck, axillae and groin. This is easy to do and has been recommended
200
Table 19.4: Summary of data on cooling methods based on conduction in the treatment of classic heatstroke3
Study Population Study N Intervention Outcomes Results Limitations
(ref., author) design measured
Hart et al12 Classic heats Case 28 Iced water Cooling time; Cooling time: < 30 Patients switched to brisk
troke series immersion; brisk mortality; minutes, 93%; massage were not
massage with ice; morbidity cooling time: 30 to 45 identified
target Trect : mins, 7%; mortality:
< 38.9°C 14.3%; neurologic
morbidity: 14.3%
Vicario et al17 Classic Case 39 Ice packs to axilla Cooling time; Cooling time: <60 Retrospective assignment
heatstroke series and groin; cold mortality mins, 69% mortality: of group; comparability of
wet sheets applied 15%; Cooling time: the groups at baseline
to torso; ice water >60 mins, 31% questionable
lavage; cooling mortality: 33%
blankets; target
trect: < 38.9°C
Trect : rectal temperature.

Heat Stroke 201
for use in the field. It has been compared to evaporative techniques in volunteers with exercise
induced hyperthermia. The time to reduce temperature by 2°C (to baseline) was 73.6 minutes
in the ice pack arm and 59.8 minutes in the evaporative arm. When both were combined,
a cooling time of 53.6 minutes was achieved.21,22
Evaporative Cooling Methods

Evaporative methods seek to induce heat losses through evaporation and convection. The
efficacy of evaporation is increased by increasing the water vapor pressure gradient by having
warm air blow over the surface of wetted skin.
After stabilizing the airway, breathing and circulation, all the patient’s clothing is removed
and ECG electrodes are placed on the back to prevent loss of adhesion due to wetting.
Conventionally, evaporative cooling consists of placing tepid wet gauze sheets all over the
body or spraying tepid water (20°C to 40°C) and using a large fan to create air-flows across
the body. This is simple, easy and quick to set up. In a case series of 14 patients with classical
heat stroke, a cooling time of 34 to 89 minutes was achieved using conventional evaporative
methods.23 However, this cooling method was combined with other conductive cooling
strategies in some patients. In a crossover trial of evaporative cooling versus immersion in
6 volunteers with exertional hyperthermia, evaporative cooling led to faster cooling with a
mean cooling time of 50 minutes (rate of 0.07°C/min).24
This basic evaporative cooling principle has been used to design a specialized body cooling
unit. In this unit, water at 15°C is sprayed all over the body and warm air at 45°C is blown
over the surface to keep the temperature of the wetted skin between 32°C and 33°C. This
unit is used mostly during the Haj in Mecca, Saudi Arabia when a large number of heat
stroke cases are encountered each year. In 16 classical heat stroke patients treated with either
the body cooling unit or conventional evaporation, no significant difference was found in
cooling times.3,25 In studies done on 6 hyperthermic volunteers, the body cooling unit was
superior to iced water immersion with cooling times of 6.5 minutes (rate 0.31°C/min) in the
body cooling unit and 18.4 mins (rate 0.11°C/min) with immersion.26 Table 19.5 summarizes
studies done on evaporative cooling in classical heat stroke.3
Invasive Cooling Methods

Various invasive cooling methods have been described, including iced intravenous (IV) fluids,
cold gastric, thoracic, peritoneal and bladder lavages, endovascular cooling probes, cold
hemodialysis and cardiopulmonary bypass. All these have been reported in animal models
and case reports. With exception of iced IV fluids, iced gastric and bladder lavages, these
techniques need specialized technical skill and sophisticated equipment. They cannot be
organized as quickly as evaporative methods and should be reserved for patients refractory
to standard cooling measures. Although iced IV fluids are commonly used for volume
resuscitating patients with heat stroke, their contribution to the cooling is small and there is
no evidence to support its use.
202
Table 19.5: Summary of data on cooling methods based on evaporation in the treatment of classic heatstroke 3
Study Population Study design Intervention Outcomes Results Limitations

(Author, ref) measured
Graham et al23 n = 14 Case series Ice to the trunk and spraying Cooling time; Median (range) Combination of
of tepid water (40°C); with mortality; cooling time: 60 several cooling
fan; massage to torso and morbidity (34 to 89) minutes; techniques; relative
neck; chilled intravenous mortality: 7.1%; contribution of
solution; target neurologic each difficult to
Trect: < 39.4°C morbidity: 0% ascertain
Al-Aska et al3,28 n = 25 Case series Wet gauze sheet with water at Cooling time; Mean (range) No follow-up
20°C; fan with speed airflow mortality; cooling time: 40.4
of 2.6 m/s; target Trect: < 39°C morbidity (20 to 145) minutes;
mortality: 0%;
morbidity: 24%
Khogali et al3,29 n = 18 Case series Body cooling unit; target Cooling time; Cooling time: 26 No follow-up
Trect: < 38°C mortality; to 300 minutes;
mortality: 11.1%
Khogali et al3,27 n = 174 Case series Body cooling unit; Cooling time; Mean (range) No follow-up
target Trect : < 38°C mortality; cooling time: 78
(20 to 180) minutes;
mortality: 14.9%
Al-Harthi et al3,25 n = 16 Randomized Body cooling unit (n = 8) Cooling time; No significant Small sample size;
controlled versus conventional method mortality; difference in randomization
trial (wet gauze sheet with water morbidity cooling time; method not
at 25°C and fanning air at Mortality: 0%; specified; no
20°C) (n = 8); target neurologic follow-up
Trect: < 38.5°C morbidity: 25% vs.
12.5%
Trect : rectal temperature.

Heat Stroke 203
Choice of Cooling Method

The available literature suggests that iced water immersion and evaporative techniques are
equally efficacious in bringing the core body temperature down quickly. Hence, the choice of
technique depends on the available resources. Iced water immersion is the preferred way of
cooling patients in some US military establishments, where it is readily available for use and
staff are familiar with dealing with patients during immersion.21 In places like Mecca, Saudi
Arabia where large volumes of heat strokes are encountered, specialized and sophisticated
equipment like the body cooling unit may be ideal. Strategic placement of ice packs, limb
immersion in iced water and covering with wet sheets are the only practical ways to cool a
patient in the field. In combat situations, very rapid cooling can be achieved by using
evaporative methods in the downwash of a helicopter.21
Conventional evaporation with water sprays and fans should probably be the default cooling
method as it is effective, needs no special equipment, causes no discomfort to the patient and
has no other downside to it other than being labor intensive.
Chemically Assisted Cooling

Dantrolene is a hydantoin derivative, which inhibits calcium release from muscle sarcoplastic
reticulum, reducing muscle tone and contractility. Dantrolene has been useful in the treatment
of malignant hyperthermia and neuroleptic malignant syndrome. Since these disorders are
also part of the thermic stress syndromes, it was suggested that dantrolene may be useful in
heat stroke. Some early case reports reported a benefit in exertional heat stroke.30 In one
randomized controlled trial of dantrolene 4 mg/kg in 20 classical heatstroke patients who also
received evaporative cooling, reductions in cooling times, but no difference in morbidity or
mortality was observed.31 This was a small study with questionable randomization techniques.
This was closely followed by a larger, better designed randomized controlled study in
52 classical heat stroke patients, of 2 mg/kg dantrolene in addition to evaporative cooling.
Dantrolene did not affect cooling time, organ dysfunction, length of stay or mortality.32 These
2 studies are summarized in Table 19.6.3
With the evidence currently available, dantrolene does not appear to offer an additional
benefit over standard cooling therapies in patients with heat stroke.
Thermometry
Core body temperature measurements are essential for the diagnosis and management of the
heat stroke patient. A rectal temperature is the quickest way of determining core temperature
to diagnose heat stroke. However, it may not be the best way to monitor cooling. Changes
in rectal temperatures measured by a thermometer or a thermistor placed 15 cm into the
204
Table 19.6: Summary of data on pharmacological cooling in the treatment of classical heatstroke 3
Study Population Study design Intervention Outcomes Results Limitations

(Author, ref) measured
Channa et al31 n = 20 Randomized Evaporative cooling + Cooling time; Cooling time in Small sample size;
controlled dantrolene 2-4 mg/kg IV mortality; the dantrolene group randomization
study (n = 8) vs. morbidity lower than control method not

Evaporative cooling alone (49.7± 4.4 vs 69.2 specified;
(n = 12); ± 4.8 minutes; comparability of
Target Trect : < 38.9°C p < 0.01); baseline charac-
no difference in teristics
morbidity and questionable
mortality
Bouchama et al32 n = 52 Randomized Evaporative cooling + Cooling time; No significant None

controlled dantrolene 2 mg/kg IV organ difference for any
study (n = 26) vs. dysfunction; of the endpoints
evaporative cooling + length of
placebo (n = 26); hospital stay;
target Trect : < 39.4°C mortality
IV, intravenous; Trect : rectal temperature.

Heat Stroke 205
rectum, lag behind actual changes in core temperature. This may result in overcooling and
overshoot hypothermia. One study showed a 33% incidence of overshooting cooling targets.33
Infrared tympanic thermometry has a significant variability and is affected by head skin
temperature, age, measuring technique and presence of cerumen.18 Measuring pulmonary
arterial temperature using a Swan Ganz catheter is the gold standard, but is no longer done
routinely. Esophageal thermistors placed in the mid esophagus are reliable and closely track
changes in core temperature. They are easy to insert in emergency situations. Urinary
temperature measured with a thermistor tipped Foley’s catheter has been shown to have a
good correlation with pulmonary artery temperatures and a high fidelity to rapidly track changes
in core temperature.34 Hence, either esophageal or urinary temperatures should be used to
monitor patients during cooling. Monitoring of temperature should continue even after the
target temperature has been achieved as a rebound increase in core temperature may occur.
Supportive Therapy
Organ system failure in heat stroke is very similar to that observed in sepsis. The principles
of management are therefore very similar. Specific measures are discussed below.
CNS: Seizures should be treated with benzodiazepines or barbiturates as phenytoin is not
effective. The theoretical suppression of sweating by barbiturates is not clinically significant
and should not deter its use.4 Some seizures may be refractory till the temperature is normalized.
Such patients should be paralyzed and ventilated until cooling is completed.
Haemodynamics: The pattern of circulatory disturbance in heat stroke is that of vasodilatation
with relative or mild absolute hypovolaemia. Central venous pressure at presentation is usually
low but total resuscitation fluid requirements are modest averaging 1-1.5 litres of crystalloids.3
Volume loading should be monitored closely as such patients are prone to developing
pulmonary oedema. Pulmonary oedema can occur during the treatment of heat stroke due
to three reasons. First, cooling reverses the massive cutaneous vasodilatation, shunting blood
back into the central blood pool, which in combination with aggressive fluid resuscitation
rapidly raises left ventricular filling pressures. Second, some patients have a reduced cardiac
contractility due to direct thermal injury to the myocardium leading to a poor tolerance of a
fluid load. Third, thermal injury to the lung leads to increased capillary permeability and
pulmonary edema manifests at lower capillary pressures.
Vasoactive medication therapy in hypotension due to heat stroke differs from that in sepsis.
Predominant alpha agonists such as noradrenaline are best avoided in the initial phase of
management. This is because the total peripheral resistance is not markedly low as the
cutaneous vasodilatation is balanced by splanchnic vasoconstriction. By pharmacologically
increasing vasoconstriction, renal and hepatic ischemia may be aggravated. Additionally, this
vasoconstriction may also interfere with cutaneous heat exchange and reduce the efficacy of
cooling.18 After the initial phase however, the endotoxinemia and systemic inflammatory
response syndrome produces the exact haemodynamic profile of sepsis for which alpha-
adrenergic agents such as noradrenaline may then be used.
Twenty per cent of patients have a hypodynamic haemodynamic profile with a low cardiac
output secondary to direct thermal damage to the myocardium.3 Such patients need more
invasive monitoring including cardiac output monitoring. Dobutamine can be used, subject
to blood pressure considerations, to increase cardiac output.4
Low dose steroids may have a role in shock due to heat stroke only if there is evidence
of hypoadrenalism due to direct thermal injury.
Pulmonary: Pulmonary oedema, aspiration pneumonia and ARDS are managed according
to current established guidelines.
Hepatic: The reversible but sometimes severe fulminant hepatic failure that occurs is treated
conservatively with particular attention to prevention of hypoglycaemia, replacement of clotting
factors and measures to reduce encephalopathy.4 Molecular Adsorbent Recirculating System
(MARS) and liver transplantation have been reported in this setting.35,36
Rhabdomyolysis and acute renal failure: When rhabdomyolysis is significant in the presence
of a normal urine output, large volume maintenance fluids, mannitol and alkanization can be
used to prevent renal injury. Once oliguric, dialysis may be indicated. Limbs should be
monitored for compartment syndrome in the presence of muscle swelling and rhabdomyolysis.
A fasciotomy should be considered when intramuscular pressure exceeds 50 mm Hg.4
DIC: Blood product support for disseminated intravascular coagulation is indicated only in
the presence of clinically evident or suspected bleeding.
Metabolic: While hypo and hyperkalaemia, hypo and hyperphosphataemia are treated in the
usual way, hypocalcaemia should be treated only when ventricular ectopics, ECG abnormalities
or seizures are present as calcium may deposit in and aggravate damage to muscles in
rhabdomyolysis.4
Experimental Therapies
Similar to studies in sepsis, IL-1 receptor antagonists, anti-endotoxin antibodies and cortico-
steroids have been shown to benefit animal models of heat stroke but have not been studied
in humans.2
Since the activation of coagulation and fibrinolysis leading to DIC is similar to what happens
in sepsis, it might be expected that Activated Protein C (APC) might have a beneficial effect
similar to its effect in select patients with sepsis. Administering APC at the time of the heat
insult resulted in a reduction in mortality and organ injury in a mouse model of heat stroke.37
It has not been evaluated in humans.
Other molecules studied in rats include ketanserin, ipsaspirone, platonin and L-arginine.38,39
A number of investigators have been looking at the role of various candidate molecules in
prophylaxis, but these are not relevant to clinical heat stroke management.
Heat Stroke 207
RECOVERY
Patients with heat stroke need a prolonged convalescence and should avoid exposure to heat
till their heat intolerance and thermoregulation return to normal. This is of particular importance
to sports and military personnel with exertional heat stroke. Some military health organizations
place those recovering from heat stroke in a series of work profiles that exposes them to
gradually increasing heat stress and then return them to their full duties if they are able to
tolerate it. Others expose them to a single heat tolerance test to return them to duty.40 Similar
principles are applied to sports persons wishing to return to training.41
Follow-up of heat stroke victims also requires attention to the sequelae, which are usually
neurological. The incidence of neurological sequelae depends on the severity and duration
of the initial heat stress and is around 20%. Cerebellar dysfunction, paraparesis, quadriparesis,
seizures, dystonia, myoclonus, dementia, personality disorders, Kluver-Bucy syndrome, central
pontine myelinolysis and peripheral neuropathy are some of the sequelae that have been
described. A number of Magnetic Resonance Imaging abnormalities have also been noted,
especially cerebral oedema and diffusion hyperintensities in the cerebellum, caudate nuclei
and hippocampus.42
PREVENTION
It is important to note that this potentially fatal and disabling disorder is largely preventable.
It is very important to educate the general public through mass media about heat stroke and
the simple steps they can take to prevent it. These include getting acclimatized to the heat,
drinking fluids at regular intervals (not only when thirsty), avoiding working at the hottest
time of the day and wearing loose clothing.
The other cohort who needs to be educated are primary care physicians. Patients with
heat stroke present to them in the “Golden Hour” when there is the best odds of treating
successfully. Awareness of the illness will raise the index of suspicion for heat stroke among
such physicians and will lead to fewer cases being diagnosed late.
CONCLUSION
With the juggernaut of global warming on the roll, heat stroke is set to become an increasingly
important cause of mortality and morbidity. Awareness campaigns and focused continuing
medical education for doctors and nurses are imperative. Research on newer modalities of
treatment needs encouragement and funding. Only such large scale multi-faceted approaches
will enable us to tackle this colossal public health problem.
Key Message
• Heat-stroke is defined as a core body temperature > 40°C with involvement of the CNS.
• It can be exertional or classical (non-exertional).
• Organ failure results from a combination of thermoregulatory failure with direct thermal
injury, exaggerated acute phase response and altered expression of heat shock proteins.
• Manifestations bear many similarities to sepsis, with involvement of all organs with the
exception of the pancreas.
• Treatment involves early recognition by having a high index of suspicion, reliable assessment
of core body temperature, rapid cooling and adequate supportive care of failing organ
systems.
• Cooling should be achieved as quickly as possible down to a target temperature of 38.9°C
with either conductive or evaporative measures, depending on resource availability and
familiarity with the technique.
• Dantrolene is not effective as an adjuvant to cooling.
• Creating awareness of this medical emergency among lay persons and primary care
physicians can reduce the incidence and improve outcomes by facilitating early recognition
and rapid cooling.
REFERENCES
1. Hamilton D. The immediate treatment of heatstoke. Anaesthesia. 1976;31:270.
2. Bouchama A, Knochel JP. Heat stroke. N Engl J Med. 2002;346:1978-88.
3. Bouchama A, Dehbi M, Chaves-Carballo E. Cooling and hemodynamic management in heatstroke: practical
recommendations. Crit Care 2007;11:R54.
4. Helman RS, Habal R. Heatstroke. http://emedicine.medscape.com/article/166320-overview, accessed
December 2008.
5. Jones TS, Liang AP, Kilbourne EM, et al. Morbidity and mortality associated with the July 1980 heat wave
in St. Louis and Kansas city, Mo. JAMA 1982;247:3327-31.
6. Sharma HS. Heat related deaths are largely due to brain damage. Indian J Med Res 2005;121:621-3.
7. Sriramachari S. Heat hyperpyrexia: time to act. Indian J Med Res. 2004;119:vii-x.
8. Central bureau of health intelligence. Incidence of accidental deaths in India during 2001, ‘02, ‘03.
www.cbhidghs.nic.in/writereaddata/linkimages/1107350275235.pdf, accessed December 2008
9. Blows WT. Crowd physiology: the ‘penguin effect’. Accid Emerg Nurs 1998;6:126-9.
10. Wakino S, Hori S. A case of severe heat stroke with abnormal cardiac findings. Int Heart J 2005;46:543-
50.
11. Garcin JM, Bronstein JA. Acute liver failure is frequent during heat stroke. World J Gastroenterol
2008;14:158-9
12. Hart GR, Anderson RJ, Crumpler CP et al. Epidemic classical heat stroke: Clinical characteristics and
course of 28 patients. Medicine (Baltimore) 1982;61:189-97
13. Varghese GM, John G. Predictors of multi-organ dysfunction in heatstroke. Emerg Med J 2005;22:185-
7.
14. Heled Y, Rav-Acha M. The “Golden Hour” for heatstroke treatment. Mil Med 2004;169(3):184.
15. Bouchama A, Dehbi M, Mohamed G. Prognostic factors in heat wave related deaths. Arch Intern Med
2007;167:2170-6.
16. Delaney KA. Heatstroke: Underlying processes and lifesaving management. Postgrad Med 1992;91:379-
88.
17. Vicario SJ, Okabajue R, Haltom T. Rapid cooling in classical heat stroke – effect on mortality rates. Am
J Emerg Med 1986;4:394-8.
18. Roth BA, Benowitz NL, Olson KR. Emergency management of drug abuse. In: Karch SB, editor. Addiction
and the medical complications of drug abuse. Boca Raton: CRC press; 2007.p.100-3
Heat Stroke 209
19. Armstrong LE, Crago AE, Adams R. Whole body cooling of hyperthermic runners: comparison of two
field therapies. Am J Emerg Med 1996;14:355-8.
20. Khogali M. Evaluation and treatment of heat related illnesses. Am Fam Physician 2003;67:1439.
21. Smith JE. Cooling methods used in the treatment of exertional heat illness. Br J Sports Med 2005;39:503-
7.
22. Kielblock AJ, Van Rensburg JP, Franz RM. Body cooling as a method for reducing hyperthermia. An
evaluation of techniques. S Afr Med J 1986;9:378-80.
23. Graham BS, Lichtenstein MJ, Hinson JM, Theil GB. Nonexertional heatstroke. Physiologic management
and cooling in 14 patients. Arch Intern Med 1986;146:87-90.
24. Wyndham CH, Strydom NB, Cook HM, et al. Methods of cooling subjects with hyperpyrexia. J Appl
Physiol 1959;14:771-6.
25. Al-Harthi SS, Yaqub BA, Al-Nozha MM. Management of heat stroke patients by rapid cooling at Mecca
pilgrimage. Saudi Med J 1986;7:369.
26. Weiner JS, Khogali M. A physiological body cooling unit for treatment of heat stroke. Lancet 1980;i:507-
9.
27. Khogali M, Al-Khawashki M. Heat stroke during the Makkah Pilgrimage. Saudi Med J 1981;2:85-93.
28. Al-Aska AK, Abu-Aisha H, Yaqub B, Al-Harthi SS, Sallam A. Simplified cooling bed for heatstroke. Lancet
1987;1:381.
29. Khogali M, Weiner JS. Heat stroke: report on 18 cases. Lancet 1980;2:276-8.
30. Lydiatt JS, Hill GE. Treatment of heat stroke with dantrolene. JAMA 1981;246:41-2.
31. Channa AB, Seraj MA, Saddique AA, et al. Is dantrolene effective in heat stroke patients? Crit Care Med
1990;18:290-2.
32. Bouchama A, Cafege A, Devol EB, et al. Ineffectiveness of dantrolene sodium in the treatment of heatstroke.
Crit Care Med 1991;19:176-80.
33. Ash CJ, Cook JR, McMurray TA, et al. The use of rectal temperatures to monitor heat stroke. Mo Med.
1992;89:283-8
34. Moran JL, Peter JV, Solomon PJ, et al. Tympanic temperature measurements: Are they reliable in the
critically ill? A clinical study of measures of agreement. Crit Care Med 2007;35:155-64.
35. Sein Anand J, Chodorowski Z, Korolkeiwicz RP. Heat stroke complicated by liver failure and
hyperbilirubinemia – a case report. Przegl Lek 2007;64:344-5.
36. Berger J, Hart J, Millis M et al. Fulminant hepatic failure from heat stroke requiring liver transplantation.
J Clin. Gastroenterol 2000;30:429-31.
37. Chen CM, Hou CC, Cheng KC, et al. Activated protein C in a rat heat stroke model. Crit Care Med.
2006;34:1960-6.
38. Chang CP, Chen SH, Lin MT. Ipsapirone and ketanserin protects against circulatory shock, intracranial
hypertension and cerebral ischemia during heatstroke. Shock 2005;24:336-40.
39. Chatterjee S, Premachandran S, Sharma D et al. Therapeutic treatment with L-arginine rescues mice from
heat stroke-induced death: physiological and molecular mechanisms. Shock 2005;24:341-7.
40. O’ Connor FG, Williams AD, Blivin S, et al. Guidelines for return to duty (play) after heat illness: a
military perspective. J Sport Rehabil 2007;16:227-37.
41. Mc Dermott BP, Casa DD, Yeargin SW, et al. Recovery and return to activity following exertional heat
stroke: considerations for the sports medicine staff. J Sport Rehabil 2007;16:163-81.
42. Mahajan S, Shucany W. Symmetric bilateral caudate, hippocampal, cerebellar, and subcortical white matter
MRI abnormalities in an adult patient with heat stroke. Proc (Bayl Univ Med Cent) 2008;21:433-6.
Index
A C Diagnostic algorithm for hyponatremia

127
ABCDE protocol 16 Central line placement 39 Difficult communication in ICU and ER
Acute kidney injury 74 Cerebral salt wasting syndrome 127 12
biomarkers 74 Cervical spine clearance 146 Dysautonomia 155
serum biomarkers 75 assessment 147 clinical relevance 157
cystatin C 77 clearing cervical spine 147 controversies 160
serum creatinine 75 epidemiology 146 epidemiology 155
urinary biomarkers 77 radiological imaging 148 management 158
interleukin-18 78 CT scanning 148 pathophysiology 156
kidney injury molecule-1 78 dynamic fluoroscopy 150
Na/H exchanger isoform-3 80 MRI scanning 149
neutrophil gelatinase-associated E
plain X-ray 148
lipocalin 77 Coagulation derangements 98 Efficacy of amiodarone compared to
Administration of amiodarone 50 Colloids and renal function 116 other drugs 48
dose 50 albumin 116 amiodarone vs. diltiazem 50
route 51 synthetic colloids 116 amiodarone vs. flecainide 48
toxicity 51 Communication research in acute care amiodarone vs. ibutilide 49
Airway management in tetanus 175 areas 21 amiodarone vs. magnesium 50
principles 176 Communication skills 5 amiodarone vs. procainamide 48
severity grading 175
content skills 5 amiodarone vs. propafenone 50
Anticoagulation strategies 99
perceptual skills 5 amiodarone vs. sotalol 49
no anticoagulation 99
process skills 5 DC cardioversion for AF 48
unfractionated heparin 102
CPR before defibrillation 56 Electromagnetic interference 40
Artemesinin derivatives vs. quinine 171
Crystalloids and renal function 113
Atrial fibrillation 42
effect of chemical composition 113
AF after cardiac surgery 44
effect of tonicity 115 F
AF after major non-cardiac thoracic
surgery 45 Functioning pacemaker 35
epidemiology 42 D device-related infection 37
treatment 45 pacing-induced atrial fibrillation 37
Defibrillation 39
Autonomic dysfunction in tetanus 178 pacing-induced atrioventricular
Delirium in ICU 133
arrhythmias 179 regurgitation 37
CAM-ICU 138
complications 180 pacing-induced heart failure 35
impact of delirium 140
disturbances in conduction 179
incidence 135
haemodynamics 180
management 181 intensive care delirium screening H
sudden death 179 checklist 138
management 141 Haemodynamic instability after scorpion
risk factors 134 sting 184
B scoring for sedation 136 clinical manifestations 187
Bad news 13 delirium scoring 137 cardiac arrhythmias 188
Barriers to good communication 3 sedation scoring 136 echocardiography 189
explaining probability 3 treatment 142 grading of systemic features 187
information flood 4 Determinants of ideal antimalarial drug haemodynamic study 189
kuhnian gap 3 170 hypertension 187
linguistic differences 4 parasite clearance 170 hypotension 188
social distance 4 resistance 170 pulmonary oedema 188
vested interests 4 safety and tolerability 170 management 189
Biphasic defibrillation 60 stage specificity 170 general measures 189
local pain 189 negotiating with colleagues 8 S
prazosin 190 overcoming unrealistic expectations
scorpion antivenin 190 6 Scope of problem 10
pathophysiology 186 working within team 7 difficulties 11
Heat stroke 193 Mannitol in neuro ICU 130 dilemmas 11
incidence 194 Microcirculation 63 doctors 11
management 198 pathophysiology 64 Septic acute kidney injury 83
chemically assisted cooling 203 physiology 63 epidemiology 85
choice of cooling method 203 signs 66 characteristics 87
conductive cooling methods 198 symptoms 66 clinical outcomes 87
evaporative cooling methods Microcirculatory disturbances 67 risk factors 87
201 Microvascular flow abnormalities 66 pathophysiology 89
experimental therapies 206 Monophasic defibrillation 60 bioenergetics 91
invasive cooling methods 201 global renal blood flow 90
supportive therapy 205 N non-hemodynamic kidney injury
thermometry 203 92
manifestations 196 National drug policy on malaria 172 regional renal blood flow 91
pathophysiology 195 Need for anticoagulation 97 urinary changes 92
prevention 207 Need for good communication 1 histopathology 93
recovery 207 better outcomes 1 Severe falciparum malaria 166
risk factors 194 better quality and safety 2 antimalarial therapy 166
Hypernatremia 129 malpractice 2 artesunate 168
classification 129 physician satisfaction 2 antimalarial activity 169
hypervolaemic hypernatremia pharmacokinetics 169
129 O problems with artemisinins 170
inadequate fluid intake 129 quinine 168
Optimising microcirculation 70
increased water loss 129 pharmacokinetics and dose 168
diabetes insipidus 129 problems with quinine 168
diagnosis 129 P Single shock vs. 3 stacked shocks 59
treatment 130 Sodium-water disturbances 122
Pacemakers 30 abnormal sodium levels and their
Hyponatremia 124
classification 124 reprogramme 32 correction 123
abnormalities of pacemaker normal water metabolism 122
hypertonic 124
sensing 34 Spikes method 14
hypotonic 125
isotonic 125 battery depletion 32 E-empathy 16
capture failure 32 I-invitation 14
inappropriate pacemaker rate K-imparting knowledge 15
I 35 P-perception 14
Ineffective information transfer 22 lead malfunctions 35 S-setting 14
cross coverage 23 Syndrome of inappropriate antidiuretic
handover 22 Q hormone secretion 125
interruptions 23 causes 126
Interaction between doctors and nurses Quantifying microcirculatory flow 68 diagnosis 125
26 Quantifying therapeutic effect 71 pathophysiology 125
Inter-professional communication 24 treatment 126
R
M Regional citrate anticoagulation 103 T
Managing difficult communication tasks advantages 105
5 citrate solutions 105 Teamwork 25
special issues in developing Temporary pacing 30
avoiding collusion 7
countries 107 Traumatic brain injury 155
breaking bad news 7
conveying prognosis 6 Richmond agitation sedation scale
dealing with difficult questions 6 (RASS) 137 U
handling denial 6 Role of amiodarone in cardioversion of
managing distress 7 recent onset AF 46 Urinary changes in septic AKI 92

Critical Care Update 2008

Uploaded by

Copyright:

Available Formats

Critical Care Update 2008

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

Critical Care Update 2008

Uploaded by

Copyright:

Available Formats

Critical Care Update

JV Peter MD DNB FRACP FJFICM

Roop Kishen DA MD FRCA

S Srinivas MD FNB MRCP EDIC

JAYPEE BROTHERS MEDICAL PUBLISHERS (P) LTD

Critical Care Update 2008

First Edition: 2009

Sean M Bagshaw MD, MSc, FRCPC Mathew Joseph MCh (Neurosurgery)

Shalini Nair Saxon Ridley MD, FRCA

JV Peter MD, DNB, FRACP, FJFICM Ravikiran Sonawane MD

Sydney (Australia) Vineet Nayyar

Section B: Cardiovascular Problems in ICU

Section C: Renal Problems in ICU

Section D: Trauma/Neuro-critical Care

13. Delirium in the ICU and its Consequences 133

Section E: Tropical Diseases in ICU

“The secret of the care of a patient is caring for the patient.”

THE NEED FOR GOOD COMMUNICATION

result of effective communication. Examples include improved postsurgical pain, severity of

Better Quality and Safety

BARRIERS TO GOOD COMMUNICATION

The Kuhnian Gap

Social Distance and Linguistic Differences

COMMUNICATION SKILLS CAN BE TAUGHT

MANAGING DIFFICULT COMMUNICATION TASKS

The world of medicine has become increasingly

Overcoming Unrealistic Expectations

Unrealistic expectations can be avoided by assuring consistency in communication, early

Breaking Bad News

Working within a Team

Negotiating with Colleagues

“Practice of medicine involves communicating lots of little bad news everyday”

SCOPE OF THE PROBLEM

Doctors’ Dilemmas and Difficulties

What kind of difficult news is communicated in ICU and ER

What should the patients and the relatives be told

HOW SHOULD BAD NEWS BE BROKEN

S–Summary and Forward Planning

DO’s AND DON’Ts: SOME PRACTICAL POINTS6,13,18,19

Special Considerations in Indian Context

2. Matthews H. Ombudsman calls for better communication. Br Med J 1998;317:1472.

COMMUNICATION RESEARCH IN ACUTE CARE AREAS

INEFFECTIVE INFORMATION TRANSFER

Table 3.2: Comparison of bedside and office handover

INTERACTION BETWEEN DOCTORS AND NURSES

Narasimhan et al29 evaluated the effect of a standardised worksheet on physicians and

CALLING FOR HELP

What should an Intensivist Know

DOES MY PATIENT NEED TEMPORARY PACING

Table 4.1: Indications for temporary pacing in the absence of primary

Side effect/overdose of cardiac drugs Beta blockers

Is Temporary Pacing Programmed Appropriately

IS PACEMAKER THE REASON FOR THE PATIENT’S ADMISSION TO ICU

Table 4.2: Causes of capture failure

Lead related problems Dislodgment

Abnormalities of Pacemaker Sensing

Inappropriate Pacemaker Rate

IS A NORMALLY FUNCTIONING PACEMAKER THE REASON FOR ADMISSION

Pacing-induced Heart Failure

Table 4.3: Summary of RCTs comparing onset of heart failure among