Infection Control in The Intensive Care

Infection Control in the Intensive
Care Unit
H. K. F. van Saene L. Silvestri

M. A. de la Cal A. Gullo

Editors
Infection Control
in the Intensive
Care Unit
Third Edition
Foreword by Julian Bion
123
H. K. F. van Saene M. A. de la Cal
Institute of Aging and Chronic Diseases Department of Intensive Care Medicine
University of Liverpool Hospital Universitario de Getafe
Liverpool Getafe, Madrid
UK Spain
L. Silvestri A. Gullo
Department of Emergency and Unit of Department of Anesthesia
Anesthesia and Intensive Care and Intensive Care
Presidio Ospedaliero di Gorizia School of Medicine
Gorizia University Hospital Catania
Italy Catania
Italy
ISBN 978-88-470-1600-2 e-ISBN 978-88-470-1601-9

DOI 10.1007/978-88-470-1601-9
Springer Milan Heidelberg Dordrecht London New York
Library of Congress Control Number: 2011929635
Springer-Verlag Italia 1998, 2005, 2012

This work is subject to copyright. All rights are reserved, whether the whole or part of the material is
concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcast-
ing, reproduction on microfilm or in any other way, and storage in data banks. Duplication of this
publication or parts thereof is permitted only under the provisions of the Italian Copyright Law in its
current version, and permission for use must always be obtained from Springer. Violations are liable to
prosecution under the Italian Copyright Law.
The use of general descriptive names, registered names, trademarks, etc. in this publication does not
imply, even in the absence of a specific statement, that such names are exempt from the relevant
protective laws and regulations and therefore free for general use.
Product liability: The publishers cannot guarantee the accuracy of any information about dosage and
application contained in this book. In every individual case the user must check such information by
consulting the relevant literature.
Printed on acid-free paper
Springer is part of Springer Science+Business Media (www.springer.com)

The essential of intensive care is the
prevention of complications
C. P. Stoutenbeek 19471998
Foreword
In 1847, Ignatius Semmelweiss friend and colleague Jakob Kolletschka died of

sepsis after his finger had been cut during a post-mortem examination at the
Allgemeine Krankenhaus in Vienna. Semmelweis made the connection between
the process which caused the death of his friend, and that which caused the post-
partum deaths of so many of the mothers in his obstetric clinic at the hospital. His
study of the prevention of puerperal sepsis through effective hand hygiene, and his
subsequent career, are classical examples of how inspired insight may fail to be
translated into effective action because of defective communication, professional
resistance to change, cultural incomprehension that beneficent individuals could
also be agents of harm, and lack of an underpinning scientific mechanism.
No such criticisms can be made of the editors and contributors for this valuable
and successful book, now in its third edition, which brings together international
experts in infection and infection control to review the most recent scientific
evidence in preventing critically ill patients from suffering additional harm through
the acquisition of autogenous and exogenous infections during their hospital stay.
Wider attitudes to one of the components discussed, selective digestive decon-
tamination, do bear some comparison with the Semmelweis story in terms of the
gap between the scientific evidence and implementation in practice. Future edi-
tions of this book will no doubt contain additional reflections from the behavioural
sciences. In the meantime, intensive care and infection control practitioners will
find both fact and wisdom in this compendium to guide their practice and improve
patient care.
November 2011 Julian Bion

Professor of Intensive Care Medicine
University Department of Anaesthesia and ICM
Queen Elizabeth Hospital
Edgbaston, Birmingham, UK
vii
Preface
A week-long postgraduate course was organised in Trieste, Italy, in 1994. This

course was extremely popular Europe wide. Participants were so impressed that
they asked for copies of the lectures, and as a result of the many requests, lecturers
were asked to provide a manuscript of their lecture(s). These manuscripts resulted
in the first edition of this book, published in 1998.
This first edition contained five sections, each based on a day of the course,
which comprised six lectures. The five sections Essentials in Clinical Microbiology,
Antimicrobials, Infection Control, Infections on ICU, and Special Topics. The
format remains the same today.
There are two previous editions to this 2011 edition: 1998 and 2005. The
differences between the first edition and this latest one are in the first and last
sections. Two chapters from the first edition are merged in the first section:
Carriage, and Colonisation and Infection. This occurred because 85% of all
infections are endogenous and characterised by these three stages. The other
difference is a chapter on microcirculation and infection in Section 5. Perhaps
the most important difference between the previous editions and this most recent
edition is pictured on the front cover: 15% of all infections are exogenous, and
research over the 6 years since the last edition has shown that topically applied
antimicrobials are able to control exogenous infections. However, topically
applied antimicrobials should only be part of the prophylactic protocol when
exogenous infections are endemic.
This third edition is current, with references to publications from 2011. We
regard it as important that all statements are justified by the best available evi-
dence. All authors have made efforts to avoid unsubstantiated expert opinion.
Although prevention is not entirely separate from therapy, prevention rather than
cure is pivotal in this publication.
ix
x Preface
We are grateful to Donatella Rizza, Catherine Mazars and Hilde Haala for the
their superb assistance. We hope that this third edition is instructive, and helpful in
your daily practice and that you enjoy it.
November 2011 H. K. F. van Saene

L. Silvestri
M. A. de la Cal
A. Gullo
Contents
Part I Essentials in Clinical Microbiology
1 Glossary of Terms and Definitions . . . . . . . . . . . . . . . . . . . . . . . 3

R. E. Sarginson, N. Taylor, M. A. de la Cal
and H. K. F. van Saene
2 Carriage, Colonization and Infection. . . . . . . . . . . . . . . . . . . . . . 17

L. Silvestri, H. K. F. van Saene and J. J. M. van Saene
3 Classification of Microorganisms According

to Their Pathogenicity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29
M. A. de la Cal, E. Cerd, A. Abella and P. Garcia-Hierro
4 Classification of ICU Infections . . . . . . . . . . . . . . . . . . . . . . . . . . 41

L. Silvestri, H. K. F. van Saene and A. J. Petros
5 Gut Microbiology: Surveillance Samples for Detecting

the Abnormal Carrier State in Overgrowth. . . . . . . . . . . . . . . . . 53
H. K. F. van Saene, G. Riepi, P. Garcia-Hierro,
B. Ramos and A. Budimir
Part II Antimicrobials
6 Systemic Antibiotics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 67
A. R. De Gaudio, S. Rinaldi and C. Adembri
7 Systemic Antifungals . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99
C. J. Collins and Th. R. Rogers
xi
xii Contents
8 Enteral Antimicrobials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

M. Snchez Garca, M. Nieto Cabrera, M. A. Gonzlez Gallego
and F. Martnez Sagasti
Part III Infection Control
9 Evidence-Based Infection Control in the Intensive Care Unit . . . . 145

J. Hughes and R. P. Cooke
10 Device Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

A. R. De Gaudio, A. Casini and A. Di Filippo
11 Antibiotic Policies in the Intensive Care Unit . . . . . . . . . . . . . . . 173

H. K. F. van Saene, N. J. Reilly, A. de Silvestre and F. Rios
12 Outbreaks of Infection in the ICU: Whats up at the Beginning

of the Twenty-First Century? . . . . . . . . . . . . . . . . . . . . . . . . . . . 189
V. Damjanovic, N. Taylor, T. Williets and H. K. F. van Saene
13 Preventing Infection Using Selective Decontamination

of the Digestive Tract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203
L. Silvestri, H. K. F. van Saene and D. F. Zandstra
Part IV Infections on ICU
14 Lower Airway Infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219

J. Almirall, A. Liapikou, M. Ferrer and A. Torres
15 Bloodstream Infection in the ICU Patient . . . . . . . . . . . . . . . . . . 233

J. Valls and R. Ferrer
16 Infections of Peritoneum, Mediastinum, Pleura, Wounds,

and Urinary Tract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251
G. Sganga, G. Brisinda, V. Cozza and M. Castagneto
17 Infection in the NICU and PICU. . . . . . . . . . . . . . . . . . . . . . . . . 289

A. J. Petros, V. Damjanovic, A. Pigna and J. Farias
18 Early Adequate Antibiotic Therapy. . . . . . . . . . . . . . . . . . . . . . . 305

R. Reina and M. A. de la Cal
Contents xiii
19 ICU Patients Following Transplantation . . . . . . . . . . . . . . . . . . . 315

A. Martinez-Pellus and I. Corts Puch
20 Clinical Virology in NICU, PICU and AICU . . . . . . . . . . . . . . . . 333

C. Y. W. Tong and S. Schelenz
21 AIDS Patients in the ICU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353

F. E. Arancibia and M. A. Aguayo
22 Therapy of Infection in the ICU . . . . . . . . . . . . . . . . . . . . . . . . . 373

J. H. Rommes, N. Taylor and L. Silvestri
Part V Special Topics
23 The Gut in the Critically Ill: Central Organ in Abnormal

Microbiological Carriage, Infections, Systemic Inflammation,
Microcirculatory Failure, and MODS . . . . . . . . . . . . . . . . . . . . . 391
D. F. Zandstra, H. K. F. van Saene and R. E. Sarginson
24 Nonantibiotic Measures to Control

Ventilator-Associated Pneumonia . . . . . . . . . . . . . . . . . . . . . . . . 401
A. Gullo, A. Paratore and C. M. Celestre
25 Impact of Nutritional Route on Infections: Parenteral

Versus Enteral . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 411
A. Gullo, C. M. Celestre and A. Paratore
26 Gut Mucosal Protection in the Critically Ill Patient:

Toward an Integrated Clinical Strategy. . . . . . . . . . . . . . . . . . . . 423
D. F. Zandstra, P. H. J. van der Voort, K. Thorburn
27 Selective Decontamination of the Digestive Tract:

Role of the Pharmacist . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 433
N. J. Reilly, A. J. Nunn and K. Pollock
28 Antimicrobial Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451

N. Taylor, I. Corts Puch, L. Silvestri, D. F. Zandstra
xiv Contents
29 ICU-Acquired Infection: Mortality, Morbidity, and Costs . . . . . . 469

J. C. Marshall and K. A. M. Marshall
30 Evidence-Based Medicine in ICU . . . . . . . . . . . . . . . . . . . . . . . . 485

A. J. Petros, K. G. Lowry, H. K. F. van Saene and J. C. Marshall
Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 507
Contributors
A. Abella Department of Intensive Care Medicine, Hospital Universitario de

Getafe, Madrid, Spain
C. Adembri Section of Anesthesiology and Intensive Care, Department of Med-
ical and Surgical Critical Care, University of Florence, Azienda Ospedaliero-
Universitaria Careggi, Florence, Italy
M. A. Aguayo Unidad de Cuidados Intensivos, Instituto Nacional Trax, Santiago,
Chile
J. Almirall PhD, MD Pneumology, Consorci Sanitari del Maresme, Barcelona,
Spain
F. E. Arancibia Unidad de Cuidados Intensivos, Instituto Nacional Trax,
Santiago, Chile
G. Brisinda Istituto di Clinica Chirurgica, Universit Cattolica del Sacro Cuore,
Rome, Italy
A. Budimir Department for Clinical and Molecular Microbiology, University of
Zagreb School of Medicine, University Hospital Centre Zagreb, Zagreb, Croatia
A. Casini MD Careggi Teaching Hospital, Section of Anaesthesia, Department of
Critical Care, University of Florence, Florence, Italy
M. Castagneto Istituto di Clinica Chirurgica, Universit Cattolica del Sacro
Cuore, Rome, Italy
C. M. Celestre MD Department of Anesthesia and Intensive Care, School of
Medicine, University Hospital Catania, Catania, Italy
E. Cerd Department of Intensive Care Medicine, Hospital Universitario de Parla,
Madrid, Spain
xv
xvi Contributors
C. J. Collins Clinical Microbiology, Trinity College Dublin, Dublin, Leinster,

Ireland
R. P. Cooke Department of Microbiology, University Hospital Aintree NHS
Foundation Trust, Liverpool, Merseyside, UK
I. Corts Puch Intensive Care Unit, Hospital Universitario de Getafe, Getafe,
Madrid, Spain
V. Cozza Dipartimento di Chirurgia F. Durante, Sapienza Universit di
Roma, Rome, Italy
V. Damjanovic School of Clinical Sciences, University of Liverpool, Liverpool,
UK
A. R. De Gaudio MD Careggi Teaching Hospital, Section of Anaesthesia,
Department of Critical Care, University of Florence, Florence, Italy
M. A. de la Cal Department of Intensive Care Medicine, Hospital Universitario de
Getafe, Getafe, Spain
A. de Silvestre Department of Anesthesiology and Intensive care, University
Hospital of S. Maria della Misericordia, Udine, Italy
A. Di Filippo MD Careggi Teaching Hospital, Section of Anaesthesia, Depart-
ment of Critical Care, University of Florence, Florence, Italy
J. Farias Paediatric Intensive Care Unit, Childrens Hospital Ricardo Gutierrez,
Buenos Aires, Argentina
M. Ferrer PhD, MD Servei de Pneumologia i Allrgia Respiratria, Hospital
Clnic, IDIBAPS, CibeRes (CB06/06/0028), Barcelona, Spain
R. Ferrer Critical Care Center, Hospital Sabadell, Sabadell, Barcelona, Spain
P. Garcia-Hierro Department of Medical Microbiology, Hospital Universitario de
Getafe, Madrid, Spain
M. A. Gonzlez Gallego Servicio de Medicina Intensiva, Hospital Clnico San
Carlos Universidad Complutense, Madrid, Spain
A. Gullo Department of Anesthesia and Intensive Care, School of Medicine,
University Hospital Catania, Catania, Italy
J. Hughes Infection Prevention and Control, 5 Boroughs Partnership NHS
Foundation Trust/University of Chester Warrington, Winwick, Warrington,
Cheshire, UK
A. Liapikou MD 1st Department of Respiratory Medicine, SOTIRIA Regional
Chest Disease Hospital of Athens, Athens, Greece
K. Lowry Intensive Care Unit, Royal Victoria Hospital, Belfast, Northern Ireland,
UK
Contributors xvii
J. C. Marshall Department of Surgery and Intensive Care, St Michaels Hospital,

Ontario, Canada
Department of Surgery and Interdepartmental Division of Critical Care, General
Hospital and University of Toronto, Toronto, Canada
K. A. M. Marshall Department of Surgery and Interdepartmental Division of
Critical Care, General Hospital and University of Toronto, Toronto, Canada
A. Martinez-Pellus Intensive Care Unit, University Hospital Virgen de la Ar-
rixaca, Murcia, Spain
F. Martnez Sagasti Servicio de Medicina Intensiva, Hospital Clnico San Carlos
Universidad Complutense, Madrid, Spain
M. Nieto Cabrera Servicio de Medicina Intensiva, Hospital Clnico San Carlos
Universidad Complutense, Madrid, Spain
A. J. Nunn Pharmacy Department, Alder Hey Childrens NHS Foundation Trust,
Liverpool, Merseyside, UK
A. Paratore MD Department of Anesthesia and Intensive Care, School of Med-
icine, University Hospital Catania, Catania, Italy
A. J. Petros Pediatric Intensive Care Unit, Great Ormond Street Childrens
Hospital, London, UK
A. Pigna Neonatal Intensive Care Unit, San Orsola Ospedale, Bologna, Italy
K. Pollock Department of Pharmacy, Western Infirmary, K Pollock, Senior
Pharmacist, Glasgow, UK
B. Ramos Department of Medical Microbiology, University Hospital Getafe,
Madrid, Spain
N. J. Reilly Pharmacy Department, Royal Liverpool Childrens NHS Trust, Alder
Hey, Liverpool, UK
R. Reina Intensive Care Unit, Hospital Interzonal de Agudos General San
Martn, La Plata, Buenos Aires, Argentina
G. Riepi Faculty of Medicine, University of Montevideo, Montevideo, Uruguay
S. Rinaldi Section of Anesthesiology, Villa Fiorita Hospital, Prato, Italy
F. Rios Pharmacy, Hospital Nacional Alejandro Posadas, Buenos Aires, Argentina
Th. R. Rogers Clinical Microbiology, St. Jamess Hospital and Trinity College
Dublin, Dublin, Leinster, Ireland
J. H. Rommes PhD, MD Gelre Ziekenhuizen Apeldoorn, Intensive Care,
Apeldoorn, The Netherlands
xviii Contributors
M. Snchez Garcia PhD, MD Servicio de Medicina Intensiva, Hospital Clnico

San Carlos Universidad Complutense, Madrid, Spain
R. E. Sarginson Paediatric Anaesthesia, Royal Liverpool Childrens NHS Trust,
S. Schelenz Institute of Biomedical and Clinical Sciences, School of Medicine,
Health Policy and Practice Faculty of Health University of East Anglia, Norwich,
UK
G. Sganga Istituto di Clinica Chirurgica, Universit Cattolica del Sacro Cuore,
Rome, Italy
L. Silvestri Department of Emergency and Unit of Anesthesia and Intensive Care,
Presidio Ospedaliero di Gorizia, Gorizia, Italy
N. Taylor School of Clinical Sciences, University of Liverpool, Liverpool,
Merseyside, UK
K. Thorburn Paediatric Intensive Care Unit, Royal Liverpool Childrens NHS
Trust Alder Hey, Liverpool, UK
C. Y. William Tong Infection, Guys and St Thomas NHS Foundation Trust and
Kings College London School of Medicine, London, UK
A. Torres MD Servei de Pneumologia i Allrgia Respiratria, Hospital Clnic,
Barcelona, Spain
J. Valls Critical Care Center, Hospital Sabadell, Sabadell, Barcelona, Spain
H. K. F. van Saene Institute of Ageing and Chronic Diseases, University of
Liverpool, Duncan Building, Liverpool, UK
School of Clinical Sciences, University of Liverpool, Liverpool, UK
J. J. M. van Saene School of Clinical Sciences, University of Liverpool,
P. H. J. van der Voort Department of Intensive Care Medicine, Onze Lieve
Vrouwe Gasthuis, Amsterdam, The Netherlands
T. Williets School of Clinical Sciences, University of Liverpool, Liverpool, UK
D. F. Zandstra Department of Intensive Care Unit, Onze Lieve Vrouwe Gasthuis,
Amsterdam, The Netherlands
Part I
Essentials in Clinical Microbiology
Glossary of Terms and Definitions
1
R. E. Sarginson, N. Taylor, M. A. de la Cal
1.1 Introduction
Defining terms is important to avoid ambiguity, particularly in the era of global

communication. Words, such as sepsis, nosocomial, colonization, and infection, are
often used in an imprecise fashion. Although standardization in terminology is
useful, revisions will be needed in the light of progress in biomedical knowledge.
Definitions can be based on a variety of concepts, varying from abnormalities in
patients physiology and clinical features to sophisticated laboratory methods.
A thoughtful introduction to clinical terminology can be found in the extensive
writings of Feinstein [1, 2], who made use of set theory and Venn diagrams to
categorize clinical conditions. The choice of boundaries between sets or values on
measurement scales can be difficult. In practice, such boundaries are often somewhat
fuzzy, for example in the diagnosis of ventilator-associated pneumonia [3].
The situation is further complicated by considering problems in measurement.
An apparently simple temperature measurement is subject to variation in time,
site, and technique, as well as to errors from device malfunction, displacement,
or misuse. Most definitions of infection at various sites include fever as a
necessary criterion, typically a temperature of C38.3C. Do we have good
evidence that this measurement is a reliable discriminator, in conjunction with
other necessary criteria, in distinguishing the presence or absence of a par-
ticular type of infection [4]?
R. E. Sarginson (&)
Paediatric Anaesthesia, Royal Liverpool Childrens NHS Trust,
e-mail: [email protected]
H. K. F. van Saene et al. (eds.), Infection Control in the Intensive Care Unit, 3
DOI: 10.1007/978-88-470-1601-9_1, Springer-Verlag Italia 2012
4 R. E. Sarginson et al.
Bone raised some important issues [58] for the terms sepsis and inflammation,
a debate that continues. Other interesting approaches in the fields of sepsis, sys-
temic inflammatory response, and multiple organ dysfunction are the use of
physiological state space concepts by Rixen et al. [9] and ideas from complex
adaptive system and network theory [1013]. A number of consensus confer-
ences have been held in recent years to seek agreement on definitions of infections
as they apply to patients in the intensive care unit (ICU) [14].
The glossary outlined here forms a basis for our clinical practice in various
aspects of intensive care infection and microbiology. We advocate definitions that
are usable in routine clinical practice and that emphasize the role of surveillance
samples in classifying the origins of infection.
1.2 Terms and Definitions
1.2.1 Acquisition
A patient is considered to have acquired a microorganism when a single surveil-

lance sample is positive for a strain that differs from previous and subsequent
isolates. This is a transient phenomenon, in contrast to the more persistent state of
carriage.
1.2.2 Bloodstream Infection
Bloodstream infections (BSI) were classified into primary, secondary, and catheter
related by the International Consensus Forum on ICU infections in 2005 [14].
Debate continues over the number and type of cultures required to detect patho-
gens in the blood [15]. The clinical impact of BSI depends on the pathogenicity of
the invading microorganism, together with the nature and severity of the host
response (see Microorganisms, and Systemic inflammatory response syndrome
(SIRS), sepsis, and septic shock definitions).
1.2.2.1 Primary Bloodstream Infection

A recognized pathogen, which is not regarded as a common skin contaminant, is
cultured from one or more blood cultures and the cultured organism is not related
to an infection at another site, including intravenous-access devices. A primary
BSI may also be present when a common skin organism, such as coagulase-
negative staphylococci, is cultured repeatedly from peripheral cultures.
1.2.2.2 Secondary Bloodstream Infection

A recognized pathogen is cultured from one or more blood cultures and is identical
to an organism responsible for an infection at another site.
1 Glossary of Terms and Definitions 5
1.2.2.3 Catheter-Related Bloodstream Infection

A pathogen is isolated from one or more blood cultures and is shown to be
simultaneously present in an intravascular device, together with clinical signs of
infection. No other source of the pathogen is identified in the patient. In practice, it
may be difficult to distinguish between an endogenous and exogenous source
unless surveillance cultures are available. If the patient has overgrowth of the
relevant pathogen in the gastrointestinal tract, translocation is another possible
mechanism for bacteremia.
1.2.3 Carriage/Carrier State
The same strain of microorganism is isolated from two or more surveillance

samples in a particular patient. In practice, consecutive throat and/or rectal sur-
veillance samples, taken twice a week (Monday and Thursday), yield identical
strains.
1.2.3.1 Normal Carrier State

Surveillance samples yield only the indigenous aerobic and anaerobic flora,
including Escherichia coli in the rectum. Varying percentages of people carry
normal potential pathogens in the throat and/or gut. Streptococcus pneumoniae
and Haemophilus influenzae are carried in the oropharynx by more than half of the
healthy population. Staphylococcus aureus and yeasts are carried in the throat and
gut by up to a third of healthy people.
1.2.3.2 Abnormal Carrier State

Opportunistic abnormal aerobic Gram-negative bacilli (AGNB) or methicillin-
resistant S. aureus (MRSA) are persistently present in the oropharynx and/or
rectum. MRSA and AGNB are listed under abnormal microorganisms. E. coli,
isolated from the oropharynx in overgrowth concentrations [[2+ or [105 colony-
forming units (CFU)/ml], also represents an abnormal carrier state.
1.2.3.3 Primary Carriage

Primary carriage is the persistent presence of both normal and abnormal potential
pathogens in the admission flora surveillance (throat and rectum) samples.
1.2.3.4 Secondary Carriage

Secondary carriage is the persistent presence of abnormal bacteria in throat and/or
rectum acquired during treatment in the ICU and which were not present in the
admission flora. Commonly used antibiotics eliminate normal bacteria, such as
S. pneumoniae or H. influenzae, but promote the acquisition and subsequent carriage
of abnormal AGNB and MRSA. This phenomenon is sometimes referred to as
super or secondary carriage. Overgrowth with microorganisms of low patho-
genicity, such as coagulase-negative staphylococci and enterococci, can also occur
during selective decontamination of the digestive tract (SDD).
1.2.4 Central Nervous System Infections
This important group of infections includes meningitis, meningoencephalitis,

encephalitis, ventriculitis, and shunt infection. These conditions have some overlap
and may also coexist with sinus or mastoid infections and septicemia. Microbio-
logical diagnosis usually rests on culture of cerebrospinal fluid (CSF). Frequently,
lumbar puncture is contraindicated in suspected meningitis [16]. For example, in
meningococcal infection, contraindications include coagulopathy or when com-
puted tomography (CT) scan features suggest a risk of tentorial pressure coning if
lumbar puncture were to be done. Also, empirical antibiotics have frequently been
started prior to hospital admission. These issues are particularly important in
pediatric practice, where meningococcal DNA detection in blood and/or CSF by
polymerase chain reaction (PCR) assays, together with bacterial antigen tests,
improves diagnostic yield [17]. The use of molecular techniques, including PCR, in
detecting septicemia in critically ill patients is still in the developmental stage but
shows great promise [18]. In CNS infections, the usual nonspecific criteria of fever
or hypothermia, leukocytosis or leukopenia, and tachycardia are present, with
specific symptoms that may include headache, lethargy, neck stiffness, irritability,
fits, and coma. Cutoff values depend on age and should be defined at age-specific
percentile thresholds for physiological variables, e.g., [90th percentile for heart
rate. Detailed definitions are not given here, as they would require a separate chapter.
1.2.5 Colonization
Microorganisms are present in body sites that are normally sterile, such as the
lower airways or bladder. Clinical features of infection are absent. Diagnostic
samples yield B1+ leukocytes per high power field (HPF) [19], and microbial
growth is \2+ or \105 CFU/ml.
1.2.6 Defense
1.2.6.1 Against Carriage

The defense mechanisms of the oropharynx and gastrointestinal tract, e.g., fibro-
nectin, saliva, and gastric secretions, help prevent abnormal carrier states.
1.2.6.2 Against Colonization

Defense mechanisms of internal organs against microbial invasion, e.g., the
mucociliary elevator in the airways and secreted immunoglobulins.
1.2.6.3 Against Infection

Defense mechanisms of the internal organs, beyond skin and mucosa, which
include antibodies, lymphocytes, and neutrophils.
1.2.7 Endemicity
Endemicity is defined as at least one new case per month having a diagnostic
sample positive for the outbreak strain. Endemicity can be interpreted as an
uncontrolled, ongoing outbreak.
1.2.8 Infection
Infection can be remarkably difficult to define in clinical circumstances.

Patients have often received empirical antibiotics. In principle, infection is a
microbiologically proven clinical diagnosis of local and/or generalized
inflammation. The microbiological criteria conventionally include C105 CFU/ml
of diagnostic sample from the infected organ and C2+ leukocytes present per
HPF in the sample. The thresholds chosen for clinical features and laboratory
measurements depend on patient age and assessment timing. Assessment may
include temperature changes, heart rate, changes in heart rate variability [10],
white blood cell (WBC) counts, C-reactive protein [20], and procalcitonin [21,
22]. Infections can be classified according to the concept of the carrier
state [23]:
primary endogenous infection is caused by microorganisms carried by the
patient at the time of admission to the ICU and include both normal and
abnormal microorganisms;
secondary endogenous infection is caused by microorganisms acquired on
the ICU and not present in the admission flora. These microorganisms
usually belong to the abnormal group. Potentially pathogenic microorgan-
isms are acquired in the oropharynx and followed by carriage and over-
growth in the digestive tract. Subsequently, colonization and then infection
of internal organs may occur following migration from the oropharynx into
the lower airways or translocation across the gut mucosa into the lymphatics
or bloodstream;
exogenous infection is caused by microorganisms introduced into the
patient from the ICU environment. Organisms are transferred directly,
omitting the carriage stage, to a site where colonization and then infection
occur.
1.2.9 Inflammatory Markers
Inflammatory markers are cells and proteins associated with the proinflammatory
process. These include C-reactive protein [20], procalcitonin [21, 22], tumor
necrosis factor alpha (TNF)-a, interleukin (IL)-1 and IL-6 [24], lymphocytes, and
neutrophils. The onset, magnitude, and duration of changes in these factors vary
with infection site and severity.
1.2.10 ICU infection
ICU infection refers to secondary endogenous and exogenous infections, which are
infections due to organisms not carried by the patient at the time of ICU admission
and transmitted via hands of carers [25]. The term nosocomial (literally, related to
the hospital) is widely used but lacks a precise definition.
1.2.11 Intra-Abdominal Infection
Intra-abdominal infection occurs in an abdominal organ and the peritoneal cavity

(peritonitis). Peritonitis can be a local or general inflammation of the peritoneal
cavity. Local signs such as tenderness and guarding may be difficult to elicit in
sedated ICU patients. Generalized, nonspecific features are fever (tempera-
ture C 38.3C), leukocytosis (WBC [ 12,000/mm3), or leukopenia (WBC \
4,000/mm3). Ultrasonography and/or CT evaluation may contribute to the diag-
nosis. Isolation of microorganisms from diagnostic samples at a concentration of
C2+ or C105 CFU/ml, with C2+ leukocytes, confirms the diagnosis. Specific
examples include fecal peritonitis due to colon perforation and peritonitis asso-
ciated with peritoneal dialysis.
1.2.12 Isolation
Patients are nursed in separate cubicles or rooms, with strict hygiene measures,
including protective clothing and hand washing by the staff, to control transmis-
sion of microorganisms. These measures particularly apply to patients infected
with high-level pathogens or resistant microorganisms and those with impaired
immunity.
1.2.13 Microorganisms
1.2.13.1 Normal Microorganisms

Normal microorganisms are carried by varying percentages of healthy people and
include S. aureus, S. pneumoniae, H. influenzae, Moraxella catarrhalis, E. coli,
and Candida albicans [26].
1.2.13.2 Abnormal Microorganisms

Abnormal microorganisms are carried by people with chronic disease or those
admitted to the ICU from inpatient wards or other hospitals. These are typically
AGNB or MRSA. AGNB include Pseudomonas, Acinetobacter, Klebsiella,
Citrobacter, Enterobacter, Serratia, Proteus, and Morganella spp. These organ-
isms are rarely carried by healthy people [27, 28]. Microorganisms can be ranked
by pathogenicity into three types:
highly pathogenic microorganisms, e.g., Salmonella spp, may cause infection in

an individual with a normal defense capacity;
potentially pathogenic microorganisms, e.g., S. pneumoniae in community
practice and P. aeruginosa in hospital practice, can cause infection in a patient
with impaired defense mechanisms. These two types of microbes cause both
morbidity and mortality;
microbes of low pathogenicity cause infection under special circumstances
only, e.g., anaerobes can cause abscesses when tissue necrosis is present. Low-
level pathogens in general cause morbidity and little mortality.
Intrinsic pathogenicity refers to the capacity to cause infection. The intrinsic
pathogenicity index (IPI) is defined as the ratio of the number of patients who
develop an infection due to a particular microorganism and the number of
patients who carry the organism in the throat and/or rectum. Indigenous flora,
including anaerobes and S. viridans, rarely cause infections despite being carried
in high concentrations. The IPI is typically in the range of 0.010.03. Coagulase-
negative staphylococci and enterococci are also carried in the oropharynx in high
concentrations but are unable to cause lower airway infections. High-level
pathogens, such as Salmonella spp, have an IPI approaching 1 in the gut.
Potentially pathogenic microorganisms have an IPI in the range of 0.10.3, and
include the normal and abnormal potential pathogens, which are the targets
of SDD.
1.2.14 Migration
Migration is the process whereby microorganisms carried in the throat and gut
move to colonize and possibly infect internal organs. Migration is promoted by
underlying chronic disease, some drugs, and invasive devices.
1.2.15 Outbreak
An outbreak is defined as an event in which two or more patients in a defined

location are infected by identical, often multidrug-resistant, microorganisms
transmitted via the hands of health care workers, usually within an arbitrary time
period of 2 weeks. There are two different types of infection involved in out-
breaks: secondary endogenous and exogenous. Outbreaks of secondary endoge-
nous infection are preceded by outbreaks of carriage of abnormal flora, whereas
outbreaks of exogenous infection are not. Outbreaks of carriage of microbes may
therefore have considerable significance for infection control. These two types of
outbreaks require different management approaches: SDD is designed to prevent
secondary endogenous types of outbreaks, whereas emphasis on hygiene pro-
cedures, such as handwashing and cohort nursing, is needed to prevent exoge-
nous outbreaks. SDD paste, applied to tracheostomy wounds, can reduce the risk
of exogenous transmission during an outbreak. Such outbreak episodes often

occur with multidrug-resistant microorganisms, such as Pseudomonas, MRSA, or
vancomycin-resistant enterococci (VRE). In the pediatric ICU, viruses such as
respiratory syncytial virus or rotavirus can also be a major problem.
1.2.16 Overgrowth
Overgrowth is defined as the presence of a high concentration of potentially

pathogenic microorganisms, C2+ or C105 CFU/ml, in surveillance samples from
the digestive tract [29]. Gut overgrowth can harm the critically ill patient, as it can
cause immunosuppression [30], inflammation [31], infection [32], and antimi-
crobial resistance [33]. Overgrowth control is the main mechanism of action of
SDD. SDD restores immune function [34] and reduces inflammation [35], infec-
tion rates [36], and antimicrobial resistance [37].
1.2.17 Pneumonia
1.2.17.1 Microbiologically Confirmed Pneumonia

presence of new or progressive infiltrates on a chest X-ray for C48 h, and
fever C38.3C, and
leukocytosis (WBC [ 12,000/ml) or leukopenia (WBC \ 4,000/ml), and
purulent tracheal aspirate containing C2+ WBC/HPF, and
tracheal aspirate specimen yielding C105 CFU/ml, or
protected brush specimen (PBS) yielding [103 CFU/ml, or
bronchoalveolar lavage (BAL) specimen yielding [104 CFU/ml.
1.2.17.2 Clinical Diagnosis Only

The first four microbiological criteria are fulfilled, but tracheal aspirates, PBS, or
BAL are sterile. Criteria for the diagnosis of pneumonias remain controversial [3].
The situation is sometimes complicated by viral etiologies and/or prior antibiotic
treatment, particularly in infants and children. There is also overlap with other
pathophysiological terms, such as pneumonitis and bronchiolitis.
1.2.18 Resistance
A microorganism is considered to be resistant to a particular antimicrobial agent if:

the minimal inhibitory concentration of the antimicrobial agent against a col-
onizing or infecting microbial species is higher than the nontoxic blood con-
centration after systemic administration;
the minimum bactericidal concentration of the antimicrobial agent against
microbes carried in throat and gut is higher than the nontoxic concentration
achieved by enteral administration.
1.2.19 Samples
1.2.19.1 Diagnostic
Diagnostic or clinical samples are taken from sites that are normally sterile in
order to diagnose infection or evaluate response to therapy. Samples are taken on
clinical indication only from blood, lower airways, CSF, urinary tract, wounds,
peritoneum, joints, sinuses, or conjunctiva.
1.2.19.2 Surveillance
Surveillance samples are taken from the oropharynx and rectum on admission and
subsequently at regular intervals (usually twice weekly). These specimens are
needed to:
evaluate the abnormal carriage level of potentially pathogenic microorganisms,
in particular, overgrowth;
assess the eradication of potential pathogens by enteral nonabsorbable anti-
microbial regimens used in SDD protocols;
detect the carriage of resistant strains.
1.2.20 Selective Decontamination of the Digestive Tract
SDD is an antimicrobial prophylaxis method consisting of parenteral cefotaxime

and enteral and topical polymyxin E/tobramycin/amphotericin B (PTA) to prevent
severe endogenous and exogenous infections of lower airways and blood in the
critically ill patient requiring treatment in the ICU. The full SDD protocol has four
components [25, 38, 39]:
parenteral antibiotic (e.g., cefotaxime), is administered for the first few days to
prevent or control primary endogenous infection;
nonabsorbable antimicrobials are administered into the oropharynx and gas-
trointestinal tract when surveillance cultures show abnormal carriage; the usual
combination is PTA;
a high standard of hygiene is required to prevent exogenous infection episodes;
regular surveillance samples of throat and rectum are obtained to diagnose
carrier states and monitor SDD efficacy.
The policy at Alder Hey, Liverpool, UK is to use SDD a la carte, guided by the
abnormal carrier state detected by surveillance samples. However, most ICUs that
use SDD start the regimen on admission, irrespective of surveillance swab results.
1.2.21 Systemic Inflammatory Response Syndrome,

Sepsis, and Septic Shock
Definitions for SIRS, sepsis, severe sepsis, and septic shock have been extensively
reviewed in recent years, particularly in relation to the inclusion criteria for
clinical trials [8, 40, 41]. Consensus definitions form categories based on cutoff
points in the value distributions of a number of variables. Cutoff points based

perfusion indices can be difficult to evaluate in practice. Furthermore, a patients
clinical state can change rapidly [42]. Microbiological confirmation of infection
may occur a considerable time after the clinical diagnosis of septic states. Cutoffs
and thresholds must be adjusted in the pediatric population [43].
1.2.21.1 Systemic Inflammatory Response Syndrome

SIRS can be caused by a wide variety of clinical insults [8, 44, 45] and is man-
ifested by two or more of the following:
temperature [38 or \36C;
heart rate [90 bpm;
respiratory rate [20 breaths/min;
WBC count [12,000/mm3 or \4,000/mm3, or [10% immature forms.
These variables must be adjusted in infants and children [46].
1.2.21.2 Sepsis
Sepsis is defined as SIRS with a clear infectious etiology.
1.2.21.3 Septicemia
Septicemia is sepsis with a positive blood culture. In contrast, bacteremia
is defined as a positive blood culture in a patient exhibiting no clinical
symptoms.
1.2.21.4 Severe Sepsis

Severe sepsis is defined as sepsis with organ dysfunction, hypoperfusion, or
hypotension. Manifestations of hypoperfusion may include, but are not limited to,
lactic acidosis, oliguria, and acute alterations in mental state.
1.2.21.5 Septic Shock

Septic shock is sepsis-induced hypotension, persisting despite adequate fluid
resuscitation, together with manifestations of hypoperfusion. Hypotension is
defined as a systolic blood pressure\90 mmHg or a reduction of[40 mmHg from
baseline in the absence of other causes of hypotension.
1.2.22 Sinusitis
Sinusitis is infection of the paranasal sinusesmaxillary, ethmoidal, frontal, or

sphenoidal. Symptoms and signs such as localized tenderness and purulent dis-
charge may be absent in the sedated ICU patient. Fever (temperature C 38.3C)
and leukocytosis (WBC [ 12,000/mm3) or leukopenia (WBC \ 4,000/mm3) are

the main clinical features. Plain radiographs or CT imaging may show fluid levels
of obliteration in the sinus air spaces. Surgical drainage is performed to obtain
microbiological confirmation (C2+ or C105 CFU/ml of pus, together with C2+
leukocytes).
1.2.23 Tracheitis/Bronchitis
In the absence of pulmonary infiltrates on chest X-ray, tracheitis/bronchitis is

defined as:
purulent tracheal aspirate, and
fever [38.3C, and
leukocytosis (WBC [ 12,000/mm3) or leukopenia (WBC \ 4,000/mm3);
C2+ or C105 CFU/ml of tracheal aspirate.
1.2.24 Translocation (Transmural Migration)
Translocation is defined as the passage of viable microorganisms from the throat

and gut through mucosal barriers to regional lymph nodes and internal organs,
including the blood.
1.2.25 Transmission
Transmission is defined as the spread of microorganisms between patients by

means of vectors such as a carers hands. Transmission of potential pathogens is
the crucial stage in the pathogenesis of secondary endogenous and exogenous
infections. Measures to control transmission include isolation, hand washing,
protective clothing, and care of equipment.
1.2.26 Urinary Tract Infection
Urinary tract infection is defined as infection of the urinary tract, most fre-
quently the bladder. The common clinical features of dysuria, suprapubic pain,
frequency, and urgency are often absent in the sedated ICU patient. The
diagnosis rests on a freshly voided catheter urine specimen or suprapubic
sample containing C105 bacteria or yeasts per milliliter of urine and C5
WBC/HPF.
1.2.27 Wound Infection
Wound infection is defined as purulent discharge from wounds, signs of local

inflammation, and a culture yielding C2+ or C105 CFU/ml. The isolation of skin
flora in the absence of these features is considered contamination.
References
1. Feinstein AR (1967) Clinical judgment. Williams and Wilkins, Baltimore
2. Feinstein AR (1994) Clinical judgment revisited: the distraction of quantitative models. Ann
Intern Med 120:799805
3. Bonten M (1999) Controversies on diagnosis and prevention of ventilator-associated
pneumonia. Diagn Microbiol Infect Dis 34:199204
4. Toltzis P, Rosolowski B, Salvator A (2001) Etiology of fever and opportunities for reduction
of antibiotic use in a pediatric intensive care unit. Infect Control Hosp Epidemiol 22:499504
5. Bone RC (1991) Lets agree on terminology: definitions of sepsis. Crit Care Med 19:973976
6. Canadian Multiple Organ Failure Study Group (1991) Sepsisclarity of existing
terminologyor more confusion? Crit Care Med 19:996998
7. Bone RC, Grodzin CJ, Balk RA (1997) Sepsis: a new hypothesis for pathogenesis of the
disease process. Chest 112:235243
8. Levy MM, Fink MP, Marshall JC et al (2003) 2001 SCCM/ESICM/ACCP/ATS/SIS
international sepsis definitions conference. Crit Care Med 31:12501256
9. Rixen D, Siegel JH, Friedman HP (1996) Sepsis/SIRS, physiologic classification, severity
stratification, relation to cytokine elaboration and outcome prediction in post trauma critical
illness. J Trauma 41:581598
10. Seeley AJE, Christou NV (2000) Multiple organ dysfunction syndrome: exploring the
paradigm of complex non-linear systems. Crit Care Med 28:21932200
11. Toweill DL, Goldstein B (1998) Linear and nonlinear dynamics and the pathophysiology of
shock. New Horiz 6:155168
12. Aird WC (2002) Endothelial cell dynamics and complexity theory. Crit Care Med
30(suppl):S180S185
13. Strogatz SH (2001) Exploring complex networks. Nature 410:268276
14. Calandra T, Cohen J (2005) The international sepsis forum consensus conference on
definitions of infection in the intensive care unit. Crit Care Med 33(7):15381548
15. Lee A, Mirrett S, Reller LB et al (2007) Detection of bloodstream infections in adults: how
many blood cultures are needed? J Clin Microbiol 46(3):35463548
16. Smith TL, Nathan BR (2002) Central nervous system infections in the immune-competent
adult. Curr Treat Options Neurol 4:323332
17. Carrol ED, Thomson AP, Shears P et al (2000) Performance characteristics of the polymerase
chain reaction assay to confirm clinical meningococcal disease. Arch Dis Child 83:271273
18. Cursons RTM, Jeyerajah E, Sleigh JW (1999) The use of the polymerase chain reaction to
detect septicemia in critically ill patients. Crit Care Med 27:937940
19. ACourt CHD, Garrard CS, Crook D et al (1993) Microbiological lung surveillance in
mechanically ventilated patients, using non-directed bronchial lavage and quantitative
culture. Q J Med 86:635648
20. Reny JL, Vuagnat A, Ract C et al (2002) Diagnosis and follow up of infections in intensive
care patients: value of C-reactive protein compared with other clinical and biological
variables. Crit Care Med 30:529535
21. Claeys R, Vinken S, Spapen H et al (2002) Plasma procalcitonin and C-reactive protein in
acute septic shock: clinical and biological correlates. Crit Care Med 30:757762
22. Christ-Crain M, Jaccard-Stolz D, Bingisser R et al (2004) Effect of procalcitonin-guided

treatment on antibiotic use and outcome in lower respiratory tract infections: cluster-
randomised, single-blinded intervention trial. Lancet 363:600607
23. van Saene HKF, Damjanovic V, Murray AE, de la Cal MA (1996) How to classify infections
in intensive care unitsthe carrier state, a criterion whose time has come? J Hosp Infect
33:112
24. Dinarello CA (2000) Pro-inflammatory cytokines. Chest 118:503508
25. Sarginson RE, Taylor N, Reilly N et al (2004) Infection in prolonged pediatric critical illness:
a prospective four-year study based on knowledge of the carrier state. Crit Care Med
32:839847
26. van Saene HKF, Damjanovic V, Alcock SR (2001) Basics in microbiology for the patient
requiring intensive care. Curr Anaesth Crit Care 12:617
27. Johanson WG, Pierce AK, Sanford JP (1969) Changing pharyngeal bacterial flora of
hospitalized patients. Emergence of gram-negative bacilli. N Engl J Med 281:11371140
28. Chang FY, Singh N, Gayowski T et al (1998) Staphylococcus aureus nasal colonization in
patients with cirrhosis: prospective assessment of association with infection. Infect Control
Hosp Epidemiol 19:328332
29. Husebye E (1995) Gastrointestinal motility disorders and bacterial overgrowth. J Intern Med
237:419427
30. Deitch EA, Xu DZ, Qi L, Berg RD (1991) Bacterial translocation from the gut impairs
systemic immunity. Surgery 104:269276
31. Silvestri L, van Saene HKF, Zandstra DF et al (2010) Selective decontamination of the
digestive tract reduces multiple organ failure and mortality in critically ill patients:
systematic review of randomized controlled trials. Crit Care Med 38:13701376
32. van Uffelen R, van Saene HK, Fidler V, Lwenberg A (1984) Oropharyngeal flora as a source
of bacteria colonizing the lower airways in patients on artificial ventilation. Intensive Care
Med 10:233237
33. van Saene HKF, Taylor N, Damjanovic V et al (2008) Microbial gut overgrowth guarantees
increased spontaneous mutation leading to polyclonality and antibiotic resistance in the
critically ill. Curr Drug Targ 9:419421
34. Horton JW, Maass DL, White J, Minei JP (2007) Reducing susceptibility to bacteremia after
experimental burn injury: a role for selective decontamination of the digestive tract. J Appl
Physiol 102:22072216
35. Conraads VM, Jorens PG, De Clerck LS (2004) Selective intestinal decontamination in
advanced chronic heart failure: a pilot trial. Eur J Heart Fail 6:483491
36. de la Cal MA, Cerd E, van Saene HKF et al (2004) Effectiveness and safety of enteral
vancomycin to control endemicity of methicillin-resistant Staphylococcus aureus in a
medical/surgical intensive care unit. J Hosp Infect 56:175183
37. de Smet AM, Kluytmans JA, Cooper BS et al (2009) Decontamination of the digestive tract
and oropharynx in ICU patients. N Engl J Med 360:2031
38. Baxby D, van Saene HKF, Stoutenbeek CP, Zandstra DF (1996) Selective decontamination
of the digestive tract: 13 years on, what it is, what it is not. Intensive Care Med 22:699706
39. Silvestri L, de la Cal MA, van Saene HKF (2009) Selective decontamination of the digestive
tract (SDD)twenty-five years of European experience. In: Gullo A, Besso J, Lumb PD,
Williams GF (eds) Intensive and critical care medicine. Springer, Milano, pp 273284
40. Brun-Buisson C (2000) The epidemiology of the systemic inflammatory response. Intensive
Care Med 26(suppl):S64S74
41. Cohen J, Guyatt G, Bernard GR et al (2001) New strategies for clinical trials in patients with
sepsis and septic shock. Crit Care Med 29:880886
42. Rangel-Frausto MS, Pittet D, Costigan M et al (1995) The natural history of the systemic
inflammatory response syndrome (SIRS). A prospective study. JAMA 273:117123
43. Watson RS, Carcillo JA, Linde-Zwirble WT et al (2003) The epidemiology of severe sepsis
in children in the United States. Am J Respir Crit Care Med 167:695701
44. Marik PE (2002) Definition of sepsis: not quite time to dump SIRS. Crit Care Med
30:706708
45. Zahorec R (2000) Definitions for the septic syndrome should be re-evaluated. Intensive Care
Med 26:1870
46. Lucy Lum Chai S (2005) Bloodstream infection in children. Pediatr Crit Care Med
6(Suppl):S42S44
Carriage, Colonization and Infection
2
L. Silvestri, H. K. F. van Saene
and J. J. M. van Saene
2.1 Introduction
Physiologically, internal organs such as lower airways and bladder, are sterile.
However, colonization of lower airways and bladder by potentially pathogenic
microorganisms (PPMs) is common in critically ill patients [1]. Colonization of
the internal organs generally follows impaired carriage defense of the digestive
tract, which promotes PPM carriage and overgrowth, and impaired defenses of the
host against colonization due to illness severity. Failure to clear colonizing
microorganisms from the internal organs invariably leads to high concentrations of
PPMs, predisposing to infection. The host mobilizes both humoral and cellular
defense systems to hinder the invading microorganisms. However, infection requires
not only invasion but severity of the underlying disease, which jeopardizes immu-
nocompetence. This chapter defines the concepts of carriage, colonization, and
infection and describes the host defense mechanisms against carriage, colonization,
and infection.
2.2 Definitions
Carriage is defined as the patients state in which the same strain is isolated from
at least two surveillance samples (saliva, gastric fluid, feces, throat, rectum) in
any concentration over a period of at least 1 week (Fig. 2.1). Overgrowth is
defined as the presence of C2+ or C105 colony forming units (CFU) per
L. Silvestri (&)
Department of Emergency, Unit of Anesthesia and Intensive Care,
18 L. Silvestri et al.
DIGESTIVE TRACT INTERNAL ORGANS
Acquisition
Carriage
Overgrowth
Colonization
Infection
Surveillance samples Diagnostic samples
Fig. 2.1 The slippery slope of the pathogenesis of infection in critically ill patients. Acquisition
develops if only one surveillance sample is positive for a potentially pathogenic microorganism
(PPM) that differs from the previous and following isolates. Acquisition refers to the transient
presence of a microorganism (usually in the oropharynx and gut), whereas carriage is a persistent
phenomenon. Carriage or carrier state is the patients state in which the same bacterial strain is
isolated from at least two surveillance samples (saliva, gastric fluid, feces, throat, rectum) in any
concentration over a period of at least 1 week. Overgrowth is defined as C105 colony-forming units
(CFU)/ml of saliva or gastric fluid or gram of feces and is nearly always present in the critically ill
intensive care unit (ICU) patient with impaired gut motility. Colonization is the presence of a PPM
in an internal organ that is normally sterile (e.g., lower airways, bladder) without inflammatory host
response. The diagnostic sample yields \105 CFU/ml of diagnostic sample. Infection is a
microbiologically proven clinical diagnosis of inflammation. Apart from the clinical signs of
infection, the diagnostic sample obtained from the internal organ contains C105 CFU/ml or is
positive in blood, cerebrospinal fluid, and pleural fluid. Surveillance samples are samples from body
sites where potentially pathogenic microorganisms are carried, such as digestive tract and skin
lesions (tracheotomy, wounds, pressure sores). A surveillance set comprises throat and rectal swabs
taken on admission and twice weekly thereafter, e.g., on Monday and Thursday. The purpose of
surveillance samples is to determine the microbiological endpoint of the level of PPM carriage.
Diagnostic samples are samples from internal organs that are normally sterile, such as lower
airways, blood, and bladder. The aim of diagnostic samples is clinical, i.e., to microbiologically
prove a diagnosis of inflammation, either generalized or local
2 Carriage, Colonization and Infection 19
milliliter of saliva and/or per gram of feces. Colonization must be distinguished

from carriage and is defined as the presence of a microorganism in an internal
organ that is normally sterile (e.g., lower airways, bladder), without any host
inflammatory response (Fig. 2.1). Diagnostic samples such as lower airway
secretions, wound fluid, and urine generally yield \105 CFU of PPMs per mil-
liliter of diagnostic sample. In general, few leukocytes are present in colonized
internal organs on a semiquantitative scale of + = few, ++ = moderate, and
+++ = many [2]. Carriage and colonization are two different stages in the
pathogenesis of endogenous infection in intensive care unit (ICU) patients.
The first stage is practically always the oropharyngeal and gastrointestinal carrier
state, which is followed by overgrowth. Once the PPMs are present in high
concentrations, in general C105 of potential pathogens per milliliter of saliva
and/or gram of feces, they migrate into the sterile internal organs and colonize
the lower airways and bladder. Unfortunately, the term colonization is often used
to cover both stages of carriage and colonization [3].
Infection is a microbiologically proven clinical diagnosis of inflammation either
local and/or generalized. This includes not only clinical signs but also the presence
of a moderate (++) number of leukocytes and of C105 CFU/ml of diagnostic
samples obtained from an internal organ, or the isolation of a microorganism from
blood, cerebrospinal fluid, or pleural fluid [2]. Diagnostic samples are collected
from internal organs that are normally sterile, such as the lower airways, bladder,
and blood. They are obtained when clinically indicated and allow the diagnosis of
colonization and infection [2].
Surveillance samples are taken from body sites where the potential pathogens
are carried, that is, the digestive tract and skin lesions (e.g., tracheostomy, pressure
sores). Generally, a set of surveillance samples consists of throat and rectal swabs
taken on patient admission to the ICU and twice weekly thereafter. The purpose
of surveillance samples is to determine the microbiological endpoint of the level of
PPM carriage. They are not useful for diagnosing infection of internal organs, as
diagnostic samples are required for this purpose [2].
2.3 Throat/Gut Flora and Internal Organs

in Health and Disease
2.3.1 Carriage
Microorganisms are carried in the oropharynx, gut, and vagina. Microorganisms

present in healthy people are usually normal flora. They are mainly anaerobes and
aerobes of the indigenous flora, together with community microorganisms such
as Streptococcus pneumoniae, methicillin-sensitive Staphylococcus aureus,
Haemophilus influenzae, Moraxella catarrhalis, Escherichia coli, and Candida
albicans. Abnormal flora is uncommon in healthy people and may be only tran-
siently present [4], whereas disease promotes oropharyngeal and gastrointestinal
carriage of these abnormal microorganisms. Abnormal microorganisms include the
eight aerobic Gram-negative bacilli (AGNB), Klebsiella, Proteus, Morganella,

Enterobacter, Citrobacter, Serratia, Acinetobacter, and Pseudomonas spp., as
well as methicillin-resistant S. aureus (MRSA). Approximately one-third of
patients with an underlying chronic conditionsuch as diabetes, alcoholism,
chronic obstructive pulmonary disease (COPD)or neonates receiving total par-
enteral nutrition, are likely to demonstrate abnormal flora in their oropharynx and
gut [4]. Moreover, previously healthy patients admitted to the ICU and requiring
long-term ventilatory support due to an acute insult, such as (surgical) trauma,
pancreatitis, or acute liver failure, may become carriers of abnormal hospital flora
in their digestive tract. Critical illness is the most important factor in conversion
from normal to abnormal carrier state [5].
2.3.2 Colonization and Infection
Secretions from internal organs, such as the lower airways, sinuses, middle ear,
lachrymal gland, and urinary tract, of healthy individuals are normally sterile.
Colonization of internal organs can occur with the two types of PPMs: normal,
including S. pneumoniae and H. influenzae; and abnormal, such as Klebsiella and
Pseudomonas spp. Three examples illustrate the concept of colonization followed
by infection:
1. Elderly people cared for in a nursing home carry S. pneumoniae and H. influenzae
in their oropharynx. During winter months, elderly people are at high risk of
developing the flu. The flu virus destroys the cilia and causes systemic
immunosuppression. Colonization of the lower airways with S. pneumoniae and
H. influenzae invariably occurs in this population during a flu epidemic. If these
patients do not receive a short course of commonly used antibiotics, colonization
of the lower airways often progresses to pneumonia associated with high
mortality rates. A similar pattern has been described for previously healthy
trauma patients admitted to the ICU [68].
2. COPD patients with a forced expiratory volume in 1 s (FEV1) \50% are
oropharyngeal carriers of both types of flora, including H. influenzae and
AGNB [9]. The severity of their underlying lung disease promotes colonization
of the lower airways with oral flora, including normal and abnormal bacteria.
The presence of bacteria in the lower airways, or colonization, is proinflam-
matory and may result in a range of important effects on the lung, including
activation of host defenses with release of inflammatory cytokines and sub-
sequent neutrophil recruitment, mucus hypersecretion, impaired mucociliary
clearance, and respiratory cell damage [10, 11, 12]. Bacterial colonization
of lower airways in COPD patients modulates the character and frequency
of exacerbations and is associated with greater airway inflammation and
accelerated decline in FEV1 [13]. An acute exacerbation of their underlying
condition may require intubation and ventilation in the ICU. The immediate
administration of an adequate antimicrobial that is active against H. influenzae
and AGNB such as Klebsiella spp. is required in order to prevent infection of

the lower airways.
3. Patients who are transferred from another hospital or ward into the ICU and
require ventilatory support often carry abnormal flora, including MRSA or
Pseudomonas spp., due to the underlying disease. The acute deterioration of
their underlying disease requires intubation, leading to colonization of the
lower airways with abnormal flora. Colonization may develop into infection
depending on the patients level of immunosuppression.
2.4 Defenses Against Carriage, Colonization, and Infection
2.4.1 Defenses Against Carriage
Healthy individuals efficiently clear abnormal AGNB from the oropharyngeal

cavity and digestive tract. This clearing property is called carriage defense [5].
Individuals are continuously exposed to AGNB. Healthy people acquire AGNB in
the oropharynx via food intake, whereas unconscious patients acquire bacteria in
the oropharynx form the environment, e.g., the ICU. The source of AGNB may be
the inanimate or the animate environment, but in general, the other long-term ICU
patients are the main sources. AGNB acquisition from such patients commonly
results in carriage, as the critically ill patient is unable to clear these abnormal
bacteria due to the underlying disease. Seven-mechanisms represent the first line
of defense against carriage of PPMs in the digestive tract (Fig. 2.2) [14]:
1. Intact anatomy of throat and gut. Chewing, swallowing, and intestinal motility
are frequently impaired in critically ill patients [5].
2. Physiology. Any severe underlying disease may reduce the exocrine function of
the stomach and the gastric acid barrier. Moreover, critically ill patients fre-
quently receive proton pump inhibitors or anti-H2 receptor antagonists, which
impair the gastric-acid barrier [12]. Fibronectin is also reduced during critical
illness, resulting in increased AGNB adherence following the increased
availability of AGNB receptor sites on the digestive tract mucosa [15].
Macrophages are thought to release elastase, which denudes the fibronectin
layer from the mucosa, making AGNB receptor sites available [16].
3. Secretions. Bile and mucus secretion is reduced in critically ill patients,
particularly in those who receive parenteral nutrition.
4. Motility. The regular, cyclical, and contractile activity of the gastrointestinal
tract, which flushes out of food remnants, cell debris, and bacteria, is impaired
in critically ill patients, leading to bacterial overgrowth. Depending on the
severity of the underlying disease, gut paralysis is associated with overgrowth
of both normal and abnormal flora [5].
5. Epithelial renewal. Intestinal epithelial cells have a rapid turnover, which
explains why these cells are highly dependent on adequate nutrient and oxygen
supply, which is frequently reduced in ICU patients.
Blood factors aiming at killing:
1. Phagocytic cells
2. Immunity
Internal organs: reduced defense against infection
Six factors:
1. Anatomy 2. Physiology
3. Cilia 4. Mucus
5. IgA 6. Cell turnover
Internal organs: reduced defense against colonization
Seven factors:
1. Anatomy 2. Physiology
3. Cilia 4. Mucus
5. IgA 6. Cell turnover
7. Indigenous flora
Digestive tract: reduced defense against carriage
Fig. 2.2 Three hurdles defending against carriage, colonization, and infection
6. Gut-associated lymphoid tissue (GALT) and IgA. These are important in

eliciting the immune response both locally in the gut and subsequently at a
systemic level. Secretory immunoglobulin A (IgA), released by GALT, is the
most common immunoglobulin in saliva, bile, and mucus, and the protective
effect is due to its ability to inhibit AGNB adherence to the epithelium by
coating the microorganisms [17].
7. Indigenous flora. These flora are primarily anaerobes and the indigenous
E. coli. Indigenous flora helps control carriage of acquired AGNBthe so-called
colonization resistance. In most cases, the beneficial functions of the indigenous
flora outweigh potentially harmful side effects. The microbiota provides diges-
tive functions, modulates host metabolism, and stimulates development of
lymphatic tissue and the mucosal immune system [5, 18]. Moreover, it can
efficiently limit gut infection by pathogenic bacteria. In fact [5, 18]:
normal flora acts as living wallpaper covering the mucosal receptor sites, thus
preventing AGNB adherence to those receptors;
anaerobes in high concentration require a huge quantities of nutrients,
resulting in AGNB starvation;
normal flora produces bacteriocidins that are bactericidal for AGNB, release
volatile fatty acids that create a growth-inhibiting environment, and are an
important source of energy for the gut epithelium (e.g., butyrate produced by
Faecalibacterium prausnitzii [19]);
presence of normal flora stimulates peristalsis;
indigenous flora stimulates host defenses; moreover,
deconjugation of bile salts by the indigenous flora is crucial for enterohepatic
circulation;
anaerobes produce b-lactamases that neutralize b-lactam antibiotics;
indigenous flora releases biopeptides, which play a role in gastroendocrine
metabolism, maintains water balance, promotes digestive tract motility, and
produces vitamins, including vitamin K, biotin, riboflavin, and folate [5].
2.4.2 Defenses Against Colonization of the Internal Organs
Abnormal carriage of AGNB and MRSA inevitably leads to overgrowth of these

bacteria in critically ill patients following deterioration of the underlying disease
[20]. Intestinal overgrowth of abnormal flora has been shown to promote and
maintain systemic immunoparalysis via liver macrophage activation [21] and is
considered an independent risk factor for colonization and infection of internal
organs [2224]. PPMs may migrate from the digestive tract toward lower airways
or bladder (endogenous colonization) or may be introduced directly into the
internal organ from an external source, either animate or inanimate (exogenous
colonization). Six clearing factors are present in these internal organs. In the lower
airways these are [25] (Fig. 2.2):
1. Anatomy integrity. The endotracheal tube damages the mucosa and promotes
microorganism adherence;
2. Physiology integrity. Inhaled particles or microorganisms must survive and
penetrate the aerodynamic filtration system of the tracheobronchial tree. Airflow
is turbulent, causing microorganisms to affect mucosal surfaces. Humidification
also causes hygroscopic organisms to increase in size, thereby aiding trapping.
Mucosal surface adhesins are known to mediate bacterial adherence to host
extracellular matrix components, such as collagen, fibrinogen, and fibronectin
[26, 27]. Fibronectin covers surface-cell receptors and thereby blocks attachment
of many microorganisms. The mucociliary blanket transports the invading
microorganisms out of the lung, and coughing aids this expulsion. In addition,
bronchial secretions contain various antimicrobial substances, such as lysozyme,
and defensins. Once the microorganism reaches the alveoli, the alveolar
macrophages and tissue histiocytes play an important role in protecting the host.
3. Cilia motility. In conjunction with mucus, cilia mechanically remove microor-
ganisms reaching its surface. Airway hygiene depends largely on mucociliary
clearance, which in turn depends upon movement of viscoelastic mucus along the
airway [28]. Aspirated or breathed material sticks to the mucus and is thus cleared
from the respiratory tract. Mucociliary clearance can be impaired by (a) genetic
defects, e.g., primary ciliary dyskinesia, cystic fibrosis; (b) secondary ciliary
dyskinesia due to artificial ventilation or toxins released by microorganisms
producing cytotoxic damage of epithelial cells (in this situation, microorganisms
may remain longer in the airways, causing colonization and infection);
(c) abnormal physicochemical properties of mucus, making it difficult to move it
along the airway. A persistent host inflammatory response driven by cytokines
fails to eliminate microorganisms and maintains the inflammatory process.
4. Secretory IgA. IgA in bronchial secretions coats microorganisms to prevent
adherence to mucosal cell receptors. Secretory IgA is the predominant
immunoglobulin present in the respiratory tract, nasal secretions, saliva, tears,
gastrointestinal fluids, and other mucous secretions. In addition, IgA can
neutralize toxin activity [29].
5. Mucosal cell turnover and desquamation. This process eliminates adherent
bacteria.
Similarly, six mechanisms are present to help prevent fecal PPMs from
colonizing the urinary tract [30] (Fig. 2.2):
1. Anatomical integrity. The bladder mucosa acts as a barrier to invading
microorganisms.
2. Intact physiology. Assists with clearing PPMs migrating from the rectal cavity
into the urethra and finally into the bladder [31]; extreme levels of osmolality,
high urea concentration, and low pH inhibit growth of some bacteria that cause
urinary tract infections;
3. Urinary flow. Mechanically removes PPMs unless they are capable of adhering
to epithelial cells in the urinary tract;
4. Mucus. covers the bladder mucosa.
5. Secretory IgA. Presence in mucus prevents adherence of fecal bacteria;
6. Mucosal cell turnover. Promotes elimination of PPMs already adhering to
bladder mucosal cells.
2.4.3 Defenses Against Infection
Colonizing microorganisms that are not eliminated from internal organs invariably
lead to a high concentration (C105) of PPMs, predisposing to invasion. The host
mobilizes both humoral and cellular defense systems to hinder the invading
microorganisms. However, infection requires both invasion and critical illness,
which jeopardize immunocompetence (Fig. 2.2).
2.5 Mechanisms of Colonization and Infection

in ICU Patients
There are two basic mechanisms of colonization and infection in ICU patients:
migration and translocation. Migration is the movement of live PPMs from one
place, e.g., throat and gut, where they are present in overgrowth, to another site, in
particular, normally sterile internal organs. Migration is the main mechanism
by which microorganisms may cause colonization/infection in ICU patients.
Migration of microorganisms in contaminated secretions from the oropharynx into
the lower airways within a few days of mechanical ventilation is considered to be
the most common route by which PPMs may enter the lung and cause colonization
and infection [22, 32, 33]. The severity of the underlying disease, which impairs
PPM clearance, is the main factor promoting colonization of the lower airways.
The presence of the plastic endotracheal tube is invariably associated with mucosal
lesions, which further enhances colonization. Finally, progression toward infection
depends on the patients immune status or defense capacity.
Potential pathogens may also cause colonization and subsequent infection,
bypassing the stage of carriage and overgrowth, i.e., exogenous colonization/
infection. An example is a lower respiratory tract colonization/infection in a
tracheotomized patient due to microorganisms not previously carried in throat
and/or gut but directly introduced following breaches of hygiene [34].
Translocation (or transmural migration) was originally defined by Berg and
Owens [35] as the passage of viable bacteria from the gut through the epithelium
to the lamina propria and hence to mesenteric lymph nodes and possibly other
organs. This was subsequently modified by Alexander et al. [36] to refer to the
movement of viable and nonviable microorganisms or their toxic products across
an intact intestinal barrier. Tsujimoto et al. [37] recently proposed a radical
revision of the definition, which includes translocation of pathogen-associated
molecular patterns. In normally healthy people, GALT macrophages are generally
effective in killing intestinal microorganisms translocating from the gut. When gut
function is impairedas in the critically ill patienteither in the anatomically
intact gastrointestinal tract or in altered intestinal mucosa, bacterial translocation
can spread into the systemic bloodstream, leading to sepsis and multiple organ
failure [38]. Gut overgrowth of PPMs, in particular, in the terminal ileum, is
required for translocation [39]. The phenomenon of translocation has been
described in surgical patients [40], patients with pancreatitis [23] and neutropenia
[41], in surgical neonates and infants receiving parenteral nutrition [42], and in
patients requiring intensive care, including mechanical ventilation [43]. Critical
illness impacts three elements of the gut: (1) it alters cellular proliferation and
death in the epithelium [44, 45]; it has a profound effect on the number of cells in
the mucosal immune system [46, 47]; (3) it changes the normal carrier state into
abnormal carriage, defined as the persistent presence of abnormal potential
pathogens in the oropharynx and/or gut [22, 32, 33].
2.6 Conclusions
Only a general well-being guarantees the efficacy of carriage defenses, which are
based on seven innate host factors that facilitate clearing abnormal AGNB from the
gut, maintain normal flora, and subsequently prevent colonization and infection of
internal organs. Most importantly, the shift from normal to abnormal flora in
individuals with an underlying disease is thought to depend on the severity of the
illness. The use of antimicrobials, which impair the microbial factor of the carriage
defense system, further promotes gut carriage and overgrowth of abnormal flora.
Oropharyngeal and gastrointestinal eradication of abnormal flora using enteral,
nonabsorbable antimicrobials polymyxin B/tobramycin and amphotericin B is the
most logical approach by which to control or minimize the risk of PPM overgrowth
in the digestive and control colonization and infection of internal organs [48].
References
1. Kerver AJH, Rommes JH, Mevissen-Verhage EAE et al (1987) Colonization and infection in
surgical intensive care patients: a prospective study. Intensive Care Med 13:347351
2. Sarginson RE, Taylor N, van Saene HKF (2001) Glossary of terms and definitions.
Curr Anaesth Crit Care 12:25
3. Pittet D, Monod M, Suter PM et al (1994) Candida colonization and subsequent infections in
critically ill surgical patients. Ann Surg 220:751758
4. Mobbs KJ, van Saene HKF, Sunderland D, Davies PDO (1999) Oropharyngeal Gram-
negative bacillary carriage. A survey of 120 healthy individuals. Chest 115:15701575
5. Rosseneu S, Rios G, Spronk PE, van Saene JJM (2005) Carriage. In: van Saene HKF,
Silvestri L, de la Cal MA (eds) Infection control in the intensive care unit, 2nd edn. Springer,
Milan, pp 1536
6. Sirvent JM, Torres A, Vidaur L et al (2000) Tracheal colonisation within 24 h of intubation
in patients with head trauma: risk factor for developing early-onset ventilator-associated
pneumonia. Intensive Care Med 26:13691372
7. Ewig S, Torres A, El-Ebiary M et al (1999) Bacterial colonization patterns in mechanically
ventilated patients with traumatic and medical injury. Incidence, risk factors and association
with ventilator-associated pneumonia. Am J Respir Crit Care Med 159:188198
8. Acquarolo A, Urli T, Perone G et al (2005) Antibiotic prophylaxis of early onset pneumonia
in critically ill comatose patients. A randomized study. Intensive Care Med 31:510516
negative bacillary carriage in chronic obstructive pulmonary disease: relation to severity of
disease. Respir Med 93:540545
10. Yamamoto C, Yoneda T, Yoshikawa M et al (1997) Airway inflammation in COPD patients
assessed by sputum levels of interleukin-8. Chest 112:505510
11. Sethi S, Murphy TF (2001) Bacterial infection in chronic obstructive pulmonary disease in
2000: state of the art. Clin Microbiol Rev 14:336363
12. Hillman KM, Riordan T, OFarrel SM, Tabacqchali S (1982) Colonization of the gastric
content in critically ill patients. Crit Care Med 10:444447
13. Patel IS, Seemungal TA, Wilks M et al (2002) Relationship between bacterial colonisation
and the frequency, character, and severity of COPD exacerbations. Thorax 57:753754
14. Silvestri L, Lenhart FP, Fox MA (2001) Prevention of intensive care unit infections.
Curr Anaesth Crit Care 12:3440
15. Proctor RA (1987) Fibronectin: a brief overview of its structure function and physiology.
Rev Infect Dis 9:S317S312
16. Dal Nogare AR, Toews GB, Pierce AK (1987) Increased salivary elastase precedes Gram-
negative bacillary colonization in post-operative patients. Am Rev Respir Dis 135:671675
17. Mestesky J, Russel M, Elson CO (1999) Intestinal IgA, novel views on its function in the
defence of the largest mucosal surface. Gut 44:25
18. Barber S, Wolf-Dietrich H (2011) Mechanisms controlling pathogen colonization of the gut.
Curr Opin Microbiol 14:8291
19. Sokol H, Pigneur B, Watterlot L et al (2008) Faecalibacterium prausnitzii is an anti-
inflammatory commensal bacterium identified by gut microbiota analysis of Crohn disease
patients. Proc Natl Acad Sci U S A 105:1673116736
21. Marshall JC, Christou NV, Meakins JL (1988) Small-bowel bacterial overgrowth and
systemic immuno-suppression in experimental peritonitis. Surgery 104:404411
22. van Uffelen R, van Saene HKF, Fidler V et al (1984) Oropharyngeal flora as a source of
colonizing the lower airways in patients on artificial ventilation. Intensive Care Med 10:
233237
23. Luiten EJT, Hop WCJ, Endtz HP et al (1988) Prognostic importance of Gram-negative
intestinal colonization preceding pancreatic infection in severe acute pancreatitis. Intensive
Care Med 24:438445
24. Oodstijk EAN, de Smet AMGA, Kesecioglu J, Bonten MJM, on behalf of the Dutch SOD-SDD
trialists group (2011) The role of intestinal colonization with Gram-negative bacteria as a source
for intensive care unit-acquired bacteremia. Crit Care Med 39:961966
25. Manson CM, Summer WR, Nelson S (1992) Pathophysiology of pulmonary defence
mechanisms. J Crit Care 7:4256
26. Peacock SJ, Foster TJ, Cameron BJ, Berend R (1999) Bacterial fibronectin-binding proteins
and endothelial cell surface fibronectin mediate adherence of Staphylococcus aureus to
resting human endothelial cells. Microbiology 145:34773486
27. Mongodin E, Bajolet O, Cutrona J et al (2002) Fibronectin-binding proteins of
Staphylococcus aureus are involved in adherence to human airway epithelium. Infect
Immun 70:620630
28. Cole P (2001) Pathophysiology and treatment of airway mucociliary clearance. Minerva
Anestesiol 67:206209
29. Hienzel FP (2000) Antibodies. In: Mandell GL, Bennett JE, Dolin R (eds) Mandell, Douglas
and Bennetts principles and practice of infectious diseases. Churchill Livingstone,
Philadelphia, pp 4567
30. Kass EH, Schneiderman LJ (1957) Entry of bacteria into the urinary tract of patients with
implying catheters. N Engl J Med 256:556557
31. Kunin CM, Evans C, Bartholomew D, Bates DG (2002) The antimicrobial defense
mechanism of the female urethra: a reassessment. J Urol 168:413419
32. Johanson WG Jr, Pierce AK, Sandford JP et al (1972) Nosocomial respiratory tract infections
with Gram-negative bacilli: the significance of colonization of the respiratory tract. Ann
33. Estes RJ, Meduri GU (1995) The pathogenesis of ventilator associated pneumonia: I.
Mechanisms of bacterial trans-colonization and airway inoculation. Intensive Care Med
21:365383
34. Morar P, Makura Z, Jones A et al (2000) Topical antibiotics on tracheostoma prevent
exogenous colonization and infection of lower airways in children. Chest 117:513518
35. Berg RD, Owens WE (1979) Inhibition of translocation of viable Escherichia coli from the
gastrointestinal tract of mice by bacterial antagonism. Infect Immun 25:820827
36. Alexander JW, Boyce ST, Babcock GF et al (1990) The process of microbial translocation.
Ann Surg 212:496510
37. Tsujimoto H, Ono S, Mochizuki H (2009) Role of translocation of pathogen-associated
molecular patterns in sepsis. Dig Surg 26:100109
38. Sganga G, van Saene HKF, Brisinda G, Castagneto M (2001) Bacterial translocation.
In: van Saene HKF, Sganga G, Silvestri L (eds) Infection in the critically ill: an ongoing
challenge. Springer, Milan, pp 3545
39. Husebye E (1995) Gastro-intestinal motility disorders and bacterial overgrowth. J Intern Med
237:419427
40. Kane TD, Wesley Alexander J, Johannigman JA (1998) The detection of microbial DNA in
the blood. A sensitive method for diagnosing bacteremia and/or bacterial translocation in
surgical patients. Ann Surg 227:19
41. Tancrede CH, Andremont AO (1985) Bacterial translocation and Gram-negative bacteremia
in patients with hematological malignancies. J Infect Dis 152:99103
42. van Saene HKF, Taylor N, Donnell SC et al (2003) Gut overgrowth with abnormal flora: the
missing link in parenteral nutrition-related sepsis in surgical neonates. Eur J Clin Nutr
57:548553
43. Feltis BA, Wells CL (2000) Does microbial translocation play a role in critical illness?
Curr Opin Crit Care 6:117122
44. Coopersmith CM, Stromberg PE, Davis CG et al (2003) Sepsis from Pseudomonas
aeruginosa pneumonia decreases intestinal proliferation and reduces gut epithelial cell cycle
arrest. Crit Care Med 39:16301637
45. Husain KD, Stromberg PE, Woolsey CA et al (2005) Mechanisms of decreased intestinal
epithelial proliferation and increased apoptosis in murine acute lung injury. Crit Care Med
33:23502357
46. Fukatsu K, Sakamoto S, Hara E et al (2005) Gut ischemiareperfusion affects gut mucosal
immunity: a possible mechanism for infectious complications after severe surgical insults.
Crit Care Med 34:182187
47. Osterberg J, Ljungdahl M, Haglund U (2006) Influence of cyclooxygenase inhibitors on gut
immune cell distribution and apoptosis rate in experimental sepsis. Shock 25:147154
48. van Saene HKF (2008) The history of SDD. In: an der Voort PHJ, van Saene HKF (eds)
Selective digestive decontamination in intensive care medicine. Springer, Milan, pp 135
Classification of Microorganisms
According to Their Pathogenicity 3
M. A. de la Cal, E. Cerda, A. Abella
and P. Garcia-Hierro
3.1 Introduction
Isenberg wrote in 1988: A modern clinical microbiologist who asks what is a

pathogen and what is meant by virulence will meet with derision and probably be
declared heretic, bereft of his or her senses [1]. He expressed the difficulties in
defining the concepts of pathogenicity and virulence, which have changed and are
still changing with the growing number of infectious diseases in the hospital
setting and in the immunocompromised host.
It is accepted that infection is the result of the interaction between the host, the
microorganism, and the environment. Pathogenicity (Table 3.1) [24] is not only
an intrinsic quality of microorganisms but the consequence of some properties of
the microorganisms and the host. For example, coagulase-negative staphylococcus
has been considered an avirulent, opportunistic organism and not a true pathogen
[5], but the increasing number of bacteremias due to this organism in recent
decades has emphasized its pathogenicity and virulence. The ability of coagulase-
negative staphylococcus to induce disease increases when a patients defense
mechanisms are altered. Freeman et al. [6] emphasized that the fivefold increase of
coagulase-negative staphylococcus bacteremia found in a neonatal care unit from
1975 to 1982 was mainly attributable to an increase in the number of children with
a birth weight \1,000 g. Thus, the separation of pathogenicity, defense mecha-
nisms, and type of infection is only justified for didactic reasons. Some terms
routinely used by physicians working in intensive care units (ICUs) describe many
aspects involved in pathogenicity:
M. A. de la Cal (&)
Department of Intensive Care Medicine,
Hospital Universitario de Getafe, Getafe, Spain
30 M. A. de la Cal et al.
Table 3.1 Glossary of terms
Term Definition
Pathogenicity The ability of microorganisms to induce disease, which may be assessed by
disease-carriage ratios
Virulencea The severity of the disease induced by microorganisms. In epidemiological
studies virulence may be assessed by mortality or morbidity rates and the
degree of communicability
Reservoir The place where the organism maintains its presence, metabolizes, and
replicates
Source The place from which the infectious agent passes to the host. In some cases,
the reservoir and the source are the same, but not always
Infection A microbiologically proven clinical diagnosis of inflammation
Carriageb Permanent (minimally 1 week) presence of the same strain in any
concentration in body sites normally not sterile (oropharynx, external nares,
gut, vagina, skin)
Abnormal carrier The abnormal carrier state exists when the isolated microorganism is not a
state constituent of normal flora (i.e., aerobic Gram-negative bacilli and
methicillin-resistant Staphylococcus aureus) [3]
Colonizationb The presence of microorganisms in an internal organ that is normally sterile
(e.g., lower airways, bladder). The diagnostic sample yields less than a
predetermined level of CFU/ml of diagnostic sample [3]
CFU colony-forming units
a
Some authors [4] consider virulence a synonym of pathogenicity
b
Some authors [2] define colonization as the permanent presence of a micro-organism in or on a
host without clinical expression. Carrier state is the condition of an individual colonized with a
specific organism. These definitions do not take the sterility of colonized sites in normal indi-
viduals into consideration
1. Intrinsic characteristics of bacteria, i.e., Grams stain; aerobicanaerobic

requirements for growth; antibiotic sensitivityresistance patterns.
2. Quantitative criteria for defining some infections, i.e., pneumonia associated
with mechanical ventilation or urinary tract infection. For instance, the prob-
ability of having pneumonia is higher if the quantitative culture of a protected
brush catheter sample yields 104 colony forming units (CFU) instead of 103
CFU. Even the significance of the quantitative culture is different if the patient
is neutropenic. These criteria are related to the classic concept of infective dose,
which is the estimated dose of an agent necessary to cause infection.
3. Sites of isolation when evaluating the clinical significance of a culture. For
instance, Staphylococcus aureus may colonize the external nares without any
evidence of disease, but its presence in a fresh surgical wound may indicate
colonization or infection.
3 Classification of Microorganisms According to Their Pathogenicity 31
4. Communityversus hospitalversus ICU-acquired flora, which recognizes

that the relationship between the different species, the host, and the environ-
ment induces changes in the microbial habitat.
5. Carriage, colonization, and infection, which defines some possible host states
according to the significance of the presence of microorganisms in different organs.
6. Exogenous, primary endogenous, and secondary endogenous infections, which
describe a pathogenetic model based on some epidemiological criteria. In this
chapter we address the concept of pathogenicity and the epidemiological
aspects of microorganisms in clinical practice.
3.2 Magnitude of the Problem
Surveillance of microorganisms in the oropharynx and respiratory and digestive

tracts in ICU patients has provided an essential basis for our understanding of
infectious diseases in the ICU:
patients flora change after hospital admission, and this process is time
dependent [79];
in a high percentage of cases (70100%) [7, 8, 10], infections were preceded by
oropharyngeal or gut carriage with the same potentially pathogenic microor-
ganism (PPM);
digestive tract is usually the reservoir of antibiotic-resistant strains;
different microorganisms found in the same patient show different abilities to
induce infection, i.e., they have a different pathogenicity. Those observations
imply two questions: Where is the flora? Which types of flora can be differ-
entiated on an epidemiological basis?
3.2.1 Habitat
It is estimated that the human body consists of approximately 1013 cells, and hosts
10141015 individual microorganisms [1]. These microorganisms can be divided into
two groups: those that usually remain constant in their normal habitat (indigenous
flora), and those that are accidentally acquired and that, after adherence to epithelial
or mucosal surfaces, have to compete with other microorganisms and host defenses.
The final outcome could be clearance or colonization of the new organisms.
Body areas that usually harbor microorganisms (Tables 3.2 and 3.3) [11] are
skin, mouth, nasopharynx, oropharynx and tonsils, large intestine and lower ileum,
external genitalia, anterior urethra, vagina, skin, and external ear. Nevertheless, the
various anatomical sites suitable for microbiological habitats display overlapping
boundaries and are subject to variation. Temporary habitats include larynx, tra-
chea, bronchi, accessory nasal sinuses, esophagus, stomach and upper portions of
the small intestine, and distal areas of the male and female genital organs [12].
Permanent colonization is often found in patients with some risks factors, i.e.,
chronic bronchitis [13].
Table 3.2 Microorganisms commonly found in healthy human body surfaces
Surface Microorganism Frequency of

isolation
Skin Staphylococcus 4+
epidermidis
Diphtheroids 3+
Staphylococcus aureus 2+
Streptococcus spp. +
Acinetobacter spp.
Enterobacteriaceae
Mouth and throat Anaerobic Gram- 4+
negative spp.
Anaerobic cocci +
Streptococcus viridans 4+
Streptococcus 2+
pneumoniae
Streptococcus pyogenes +
Staphylococcus 4+
epidermidis
Neisseria meningitidis +
Haemophilus spp. +
Enterobacteriaceae
Candida spp. 2+
Nose Staphylococcus 4+
epidermidis
Staphylococcus aureus 2+
Streptococcus +
pneumoniae
Streptococcus pyogenes +
Haemophilus spp. +
Large intestine (95% or more of species are Anaerobic Gram- 4+
obligate anaerobes) negative spp.
Anaerobic Gram- 4+
positive spp.
Escherichia coli 4+
Klebsiella spp. 3+
Proteus spp. 3+
Enterococcus spp. 3+
Group B streptococci +
Clostridium spp. 3+
Pseudomonas spp. +
Acinetobacter spp. +
Staphylococcus +
epidermidis
Staphylococcus aureus +
Candida spp. +
(continued)
Table 3.2 (continued)

Surface Microorganism Frequency of
isolation
External genitalia and anterior urethra Skin flora 4+
Gram-negative +
anaerobic spp.
Enterococcus spp. +
Enterobacteriaceae
Vagina Lactobacillus spp. 4+
Gram-negative 2+
anaerobic spp.
Enterococcus spp. 3+
Enterobacteriaceae 1+
Acinetobacter spp.
Staphylococcus 1+
epidermidis
Candida spp. 1+
Relative frequency of isolation: 4+ almost always present, 3+ usually present, 2+ frequently
present, + occasionally present, rarely present
3.2.2 Flora
Indigenous flora is very dynamic and reflects changes induced by environmental

settings, medical treatments, and host and microbial characteristics. It is well
accepted that normal individuals have normal indigenous flora (Tables 3.2
and 3.3) that represent the equilibrium reached between the normal hosts and the
organisms. Quantitative estimation of different microorganisms found in human
body surfaces helps us to understand pathogenicity, because it is often obvious that
predominantly isolated microorganisms rarely induce disease. Anaerobes are a
good example of organisms of low pathogenicity. They represent the highest
percentage of microorganisms isolated per surface area but are only involved in a
small proportion of infections.
Surveillance cultures performed in patients after hospital or ICU admission
[79] demonstrate that flora changes over time. Carriage of aerobic Gram-negative
bacilli (AGNB) Pseudomonas spp., Klebsiella spp., Enterobacter spp., Acinetobacter
spp., Serratia and Morganella spp., and yeasts Candida spp. increases. This process of
new organism acquisition is time dependent. For example, Kerver et al. [7] studied 39
intubated patients admitted to the ICU for more than 5 days, obtaining samples three
times a week from the oropharynx, tracheal aspirate, urine, and feces. The prevalence
of AGNB in the oropharyngeal cavity on admission was 23% and increased to 80%
after 10 days. Similar figures were found for yeasts. In feces, the prevalence of AGNB
other than E. coli was 20% and reached 79% on day 15. Yeasts were found in 13% of
rectal swabs on admission and in 61% of samples on day 15. In 75.6% of infections, the
same PPM was found in previous surveillance cultures [10]. This new flora comes into
Table 3.3 Quantitative cultures of healthy human surfaces
Surface Microorganism Percent CFU

carriers
Skin (per cm) Staphylococcus epidermidis 100 105
Anaerobes (Propionibacterium 100 103
acnes)
Mouth and throat (per ml Anaerobic microorganisms 100 108
of saliva) Streptococcus viridans 100 106
Streptococcus pneumoniae 3060 103 105
Haemophilus influenzae 3080 103 105
Moraxella catarrhalis 5 103 105
Staphylococcus aureus 30 103
Large intestine (per gram Anaerobic spp. 100 1012
of feces) Escherichia coli 100 103 106
Enterococcus spp. 100 103 106
Staphylococcus aureus 30 103 105
Candida spp. 30 103 105
Vagina (per ml of vaginal Aerobic spp. 100 108
fluid) Anaerobic spp. 100 107
the human body from different animate (mostly patients anduncommonly

health personnel) or inanimate (e.g., food; furniture) sources through different
vehicles (e.g., hands; respiratory equipment). After being introduced, the new
organisms adhere to surfaces and have to compete with preexisting flora and host
defense barriers before a permanent carriage state is achieved. Apart from the
severity of the underlying disease, parenteral antibiotic administration is the main
mechanism that favors acquisition of hospital flora through selective pressure
exerted against indigenous flora [14].
The boundaries between normal, community, and abnormal hospital flora are
not always strict. Some groups of patients admitted to the hospital carry
organisms, usually constituents of abnormal hospital flora. Alcoholism, diabetes,
and chronic bronchitis are regarded as risk factors for carrying aerobic
Gram-negative bacilli [13, 15] in the oropharynx and tracheobronchial tree,
respectively.
3.3 Microorganism Classification According to Pathogenicity
Leonard et al. [16] attempted to quantitatively estimate intrinsic pathogenicity in a

population of 40 infants admitted for at least 5 days to a neonatal surgical unit.
The intrinsic pathogenicity index (IPI) for a y species was defined as:
Number of patients infected by y

IPIy
Number of patients carrying y in throat=rectum
The range of this index is 01. The highest IPI found was for Pseudomonas
spp. (0.38). Other potential pathogens isolated had an IPI \0.1 (Enterobacter spp.
0.08; S. aureus 0.06; Klebsiella spp. 0.05; E. coli 0.05; S. epidermidis 0.03;
Enterococcus spp. 0). This index provides useful information about the relative
pathogenicity of different microorganisms in a specific population and could be
used to design antibiotic policies, both prophylactic and therapeutic, in selected
groups of patients in whom microbiological surveillance could be indicated
(e.g., burn, severe trauma patients).
There are inherent limitations related to the small number of studied patients
and the small number of infections (i.e., S. aureus: one infection over 17 colo-
nizations), which can give an unreliable estimation of the IPI. Furthermore, the
authors of the study suggested that the results should be interpreted taking into
account technical aspects (definitions used; sites chosen for surveillance; micro-
biological techniques; result interpretation) and population characteristics, because
IPI does not differentiate between the organisms intrinsic pathogenicity and other
factors (host and environment) that allow their expression. The extreme alterations
in host defense mechanisms in immunosuppressed patients is a good example of
the different pathogenicity of microorganisms depending on the specific type of
systemic immunosuppression, i.e., neutropenia or cellular (T lymphocyte) immune
defect [17].
In general, the classification of microorganisms according to their pathogenicity
is based on scales with few categories. Isenberg and DAmato [12] classify
organisms as commonly involved, occasionally involved, and rarely involved in
disease production. Murray et al. [3] classify the pathogenicity of organisms as
high, potential, and low. Categories in both classifications are not always equiv-
alent. We found that the Murray et al. [3] classification (Table 3.4) is useful in ICU
practice because it is best adapted to flora isolated in ICU patients and is more
discriminatory between organisms of interest in the ICU. Another possible
advantage is that this classification integrates other concepts of clinical epidemi-
ology, such as community, or normal; and hospital, or abnormal, flora.
3.4 Antimicrobial Resistance as a Virulence Factor
Evaluation of the influence of antimicrobial resistance on mortality is difficult

because of the requirement to adjust for underlying disease and illness severity.
Recent literature suggests that the factor of immediate, adequate treatment plays
an important role in evaluating the contribution of antimicrobial resistance to
mortality [1821]. Fagon et al. [22] found that in patients suspected of having
ventilator-associated pneumonia, an invasive strategy based on the use of fiber-
optic bronchoscopy improved the survival rate at day 14 (p = 0.02); 16.2% died in
Table 3.4 Classification of microorganisms based on their intrinsic pathogenicity
Intrinsic pathogenicity Flora Indigenous

flora
Oropharynx: Peptostreptococcus spp., Veillonella spp.,
Streptococcus viridans
Gut: Bacteroides spp., Clostridium spp., Enterococcus spp., Low Normal
Escherichia coli pathogenic
Vagina: Peptostreptococcus spp., Bacteroides spp., Lactobacillus
spp.
Skin: Propionibacterium acnes, coagulase-negative staphylococci,
Community or normal microorganisms
Oropharynx: Streptococcus pneumoniae, Haemophilus influenzae, Potentially Normal
Moraxella catarrhalis pathogenic
Gut: Escherichia coli
Oropharynx and gut: Staphylococcus aureus, Candida spp.
Hospital or abnormal microorganisms: Klebsiella spp., Proteus Potentially Abnormal
spp., Enterobacter spp., Morganella spp., Citrobacter spp., pathogenic
Serratia spp., Pseudomonas spp., Acinetobacter spp., MRSA
Epidemic microorganisms: Neisseria meningitidis, Salmonella Highly Abnormal
spp. pathogenic
MRSA methicillin-resistant Staphylococcus aureus
the invasive management group and 25.8% in the clinical management group, i.e.,
a difference of 9.6%. This survival benefit can be explained by the fact that
significantly more patients who did not undergo bronchoscopy received early,
inadequate antimicrobial therapy (one patient died in the invasive group versus 24
in the control group; p \ 0.001). The survival benefit was only transient, as the
difference in mortality was no longer significant at day 28 (p = 0.10). Only a few
of the evaluable studies adjusted mortality data for appropriate antimicrobial
treatment, underlying disease, and illness severity.
3.4.1 Methicillin-Resistant S. aureus Versus

Methicillin-Sensitive S. aureus
Of 31 bacteremia studies comparing mortality due to methicillin-resistant

S. aureus (MRSA) and methicillin-sensitive S. aureus (MSSA) [23], only six
adjusted for confounding factors, including adequate antibiotic therapy [2429].
Three studies involving 401 patients found a significantly increased mortality rate,
with odds ratios (OR) varying between 3 and 5.6. The other three studies, with a
total of 1,385 patients, showed no significant difference. Only one study compared
mortality due to MRSA in 86 bacteremic pneumonia patients and found no
significant difference [30].
3.4.2 Vancomycin-Resistant Enterococci Versus

Vancomycin-Sensitive Enterococci
Three studies in patients with positive blood cultures and that adjusted for
appropriate antibiotic treatment are available [3133]. Only one study in 106
patients reports a significantly higher mortality rate due to vancomycin-resistant
enterococci (VRE), with an OR of 4.0 (1.213.3). The other two studies in a total
of 467 patients failed to show a mortality rate difference.
3.4.3 Aerobic Gram-Negative Bacilli, Including

Acinetobacter spp. and Pseudomonas aeruginosa
One study in 135 patients compared mortality rates in patients with infections
due to piperacillin-resistant P. aeruginosa and patients with infections due to
piperacillin-sensitive P. aeruginosa [34]. Mortality data were not adjusted for
immediate, adequate antimicrobial therapy, as there was no difference in crude
mortality. There are no data on Acinetobacter mortality, whether sensitive or
resistant.
These data show that the association of antimicrobial resistance with mortality
rate has not yet been appropriately evaluated. Evidence supports the concept that
antibiotic resistance does not contribute to mortality.
3.5 Conclusions
Although the properties of the different microorganisms fail to explain com-

pletely their pathogenicity, it is clear that some characteristics do determine the
different pathogenicity; for example, encapsulated pneumococci are more viru-
lent than nonencapsulated pneumococci. Advances in molecular biology make it
possible to characterize some virulence factors that allow microorganisms to
overcome the set of obstacles to accomplish infection. These include selection of
the niche and adherence to human body surfaces or medical devices (i.e., adhe-
sins), competition with preexisting organisms in some cases, impairment of the
host defense mechanism (i.e., antiphagocytic capsules; toxins), and production of
tissue damage (i.e., toxins; enzymes). One report discusses the impact of critical
illness on the expression of virulence in gut flora [35]. It has been hypothesized
that gut bacteria change and become more virulent as the microorganisms sense
that the hosts capacity to control them is severely impaired. Identification of the
virulent genes also provides a better understanding of the relationship between
virulence factors and their clinical expression [36] and can be helpful in epi-
demiological studies, such as in determining transmission, carriage, colonization,
and infection routes with specific microorganisms and the investigation of
outbreaks [3741].
References
1. Isenberg HD (1988) Pathogenicity and virulence: another view. Clin Microbiol Rev 1:4053
2. Brachman PS (1992) Epidemiology of nosocomial infections. In: Bennett JV, Brachman PS
(eds) Hospital infections, 3rd edn. Little Brown, Boston, pp 320
3. Murray AE, Mostafa SM, van Saene HKF (1991) Essentials in clinical microbiology. In:
Stoutenbeek CP, van Saene HKF (eds) Infection and the anaesthetist, vol 5. Bailliere Tindall,
London, pp 126
4. McCloskey RV (1979) Microbial virulence factors. In: Mandell GL, Douglas RG, Bennett IE
(eds) Principles and practice of infectious diseases, vol 1, 1st edn. Wiley, New York, pp 311
5. Pfaller MA, Herwald LA (1988) Laboratory, clinical, and epidemiological aspects of
coagulase-negative staphylococci. Clin Microbiol Rev 1:281299
6. Freeman I, Platt R, Sidebottom DG et al (1987) Coagulase-negative staphylococcal
bacteremia in the changing neonatal intensive care unit population. JAMA 258:25482552
7. Kerver AIH, Rommes IH, Mevissen-Verhage EAE et al (1987) Colonization and infection in
surgical intensive care patients. Intensive Care Med 13:347351
8. Leonard EM, van Saene HKF, Shears P, Walker I, Tam PKH (1990) Pathogenesis of
colonization and infection in a neonatal surgical unit. Crit Care Med 18:264269
9. van Saene HKF, Stoutenbeek CP, Zandstra DF, Gilberston A, Murray A, Hart CA (1987)
Nosocomial infections in severely traumatized patient: magnitude of problem, pathogenesis,
prevention and therapy. Acta Anaesthesiol Belg 38:347356
10. van Saene HKF, Damjanovic V, Murray AE, de la Cal MA (1996) How to classify infections
in intensive care unitsthe carrier state, a criterion whose time has come? J Hosp Infect
33:112
11. Tramont EC (1979) General or nonspecific host defense mechanisms. In: Mandell GL,
Douglas RG, Bennett IE (eds) Principles and practice of infectious diseases, vol 1, 1st edn.
Wiley, NewYork, pp 1321
12. Isenberg HD, DAmato RF (1990) Indigenous and pathogenic micro-organisms of humans.
In: Mandell GL, Douglas RG, Bennett IE (eds) Principles and practice of infectious diseases,
vol 1, 3rd edn. Churchill Livingstone, New York, pp 214
13. Jordan GW, Wong GA, Hoeprich PB (1976) Bacteriology of the lower respiratory tract as
determined by fiberoptic bronchoscopy and transtracheal aspiration. J Infect Dis 134:428435
14. van der Waaij D (1992) Selective gastrointestinal decontamination: history of recognition
and measurement of colonization of the digestive tract as an introduction to selective
gastrointestinal decontamination. Epidemiol Infect 109:315326
15. Mackowiak PA, Martin RM, Smith LW (1979) The role of bacterial interference in the
increased prevalence of oropharyngeal Gram-negative bacilli among alcoholics and diabetics.
Am Rev Respir Dis 120:289593
16. Leonard EM, van Saene HKF, Stoutenbeek CP, Walker I, Tam PKH (1990) An intrinsic
pathogenicity index for micro-organisms causing infection in a neonatal surgical unit. Microb
Ecol Health Dis 3:151157
17. Shelhamer IH, Toews GB, Masur H et al (1992) Respiratory disease in the immuno-
suppressed patient. Ann Intern Med 117:415443
18. Alvarez-Lerma F (1996) Modification of empiric antibiotic treatment in patients with
pneumonia acquired in the intensive care unit. ICU-acquired pneumonia study group.
Intensive Care Med 22:387394
19. Luna CM, Vujacich P, Niederman MS et al (1997) Impact of BAL data on the therapy and
outcome of ventilator-associated pneumonia. Chest 111:676685
20. Iregui M, Ward S, Sherman G, Fraser VJ, Kollef MH (2002) Clinical importance of delays in
the initiation of appropriate antibiotic treatment for ventilator associated pneumonia. Chest
122:262268
21. Valles J, Rello J, Ochagavia A, Garnacho J, Alcala MA (2003) Community-acquired
bloodstream infection in critically ill adult patients: impact of shock and inappropriate
antibiotic therapy on survival. Chest 123:16151624
22. Fagon JY, Chastre J, Wolff M et al (2000) Invasive and noninvasive strategies for
management of suspected ventilator associated pneumonia. A randomized trial. Ann Intern
Med 132:621630
23. Cosgrove SE, Sakoulas G, Perencevich EN, Schwaber MJ, Karchmer AW, Carmeli Y (2003)
Comparison of mortality associated with methicillin-resistant and methicillin-susceptible
Staphylococcus aureus bacteremia: a meta-analysis. Clin Infect Dis 36:5359
24. Conterno LO, Wey SB, Castelo A (1998) Risk factors for mortality in Staphylococcus aureus
bacteremia. Infect Control Hosp Epidemiol 19:3237
25. Romero-Vivas J, Rubio M, Fernndez C, Picazo JJ (1995) Mortality associated with
nosocomial bacteremia due to methicillin-resistant Staphylococcus aureus. Clin Infect Dis
21:14171423
26. Harbarth S, Rutschmann O, Sudre P, Pittet D (1998) Impact of methicillin resistance on the
outcome of patients with bacteremia caused by Staphylococcus aureus. Arch Intern Med
158:182189
27. Soriano A, Martnez JA, Mensa J et al (2000) Pathogenic significance of methicillin
resistance for patients with Staphylococcus aureus bacteremia. Clin Infect Dis 30:368373
28. Mylotte JM, Tayara A (2000) Staphylococcus aureus bacteremia: predictors of 30-day
mortality in a large cohort. Clin Infect Dis 31:11701174
29. Topeli A, Unal S, Akalin HE (2000) Risk factors influencing clinical outcome in
Staphylococcus aureus bacteraemia in a Turkish University Hospital. Int J Antimicrob
Agents 141:5763
30. Gonzlez C, Rubio M, Romero-Vivas J, Gonzlez M, Picazo JJ (1999) Bacteremic
pneumonia due to Staphylococcus aureus: a comparison of disease caused by methicillin-
resistant and methicillin-susceptible organisms. Clin Infect Dis 29:11711177
31. Garbutt JM, Ventrapragada M, Littenberg B, Mundy LM (2000) Association between
resistance to vancomycin and death in cases of Enterococcus faecium bacteremia. Clin Infect
Dis 30:466472
32. Vergis EN, Hayden MK, Chow JW et al (2001) Determinants of vancomycin resistance and
mortality rates in enterococcal bacteremia. A prospective multicenter study. Ann Intern Med
135:484492
33. Lodise TP, McKinnon PS, Tam VH, Rybak MJ (2002) Clinical outcomes for patients with
bacteremia caused by vancomycin-resistant enterococcus in a level 1 trauma center. Clin
Infect Dis 34:922929
34. Trouillet JL, Vuagnat A, Combes A, Kassis N, Chastre J, Gibert C (2002) Pseudomonas
aeruginosa ventilator-associated pneumonia: comparison of episodes due to piperacillin-
resistant versus piperacillin-susceptible organisms. Clin Infect Dis 34:10471054
35. Alverdy JC, Laughlin RS, Wu L (2003) Influence of the critically ill state on host-pathogen
interactions within the intestine: gut-derived sepsis redefined. Crit Care Med 31:598607
36. Relman DA, Falkow S (1990) A molecular perspective of microbial pathogenicicty. In:
Mandell GL, Douglas RG, Bennett IE (eds) Principles and practice of infectious diseases, vol
1, 3rd edn. Churchill Livingstone, New York, pp 2532
37. Emori TG, Gaynes RG (1993) An overview of nosocomial infections, including the role of
the microbiology laboratory. Clin Microbiol Rev 6:428442
38. Neish AS (2002) The gut microflora and intestinal epithelial cells: a continuing dialogue.
Microbes Infect 4:309317
39. Guarner F, Malagelada JR (2003) Gutflora in health and disease. Lancet 361:512519
40. DAgata EMC, Venkataran L, De Girolami P, Burke P, Eliopoulos GM, Karchmer AW,
Samore MH (1999) Colonisation with broad-spectrum cephalosporin-resistant Gram-negative
bacilli in intensive care units during a non-outbreak period: prevalence, risk factors and rate
of infection. Crit Care Med 27:10901095
41. Safdar N, Maki DG (2002) The commonality of risk factors for nosocomial colonization and
infection with antimicrobial-resistant Staphylococcus aureus, Enterococcus Gram-negative
bacilli, Clostridium difficile and Candida. Ann Intern Med 136:834844
Classification of ICU Infections
4
and A. J. Petros
4.1 Introduction
Classifying infections is crucial in any infection surveillance program,

particularly in the intensive care unit (ICU). Time cutoffs, generally 48 h, have
been accepted to distinguish community- from hospital-acquired infections,
including ICU-acquired infections [1]. However, many clinicians appreciate that
an infection due to a microorganism carried by the patient on admission to the
ICU and that develops 48 h of ICU stay cannot be considered as being a
true ICU-acquired infection. Obviously, this infection is nosocomial, i.e., the
infection occurs in the ICU, but the causative microorganism does not belong
to the ICU microbial ecology, as the patient imported the microorganism in
her/his admission flora. A different classification of ICU infections is based
on knowledge of the patients carrier state [2]. This approach allows the
distinction between primary (imported) and secondary carriage of potentially
pathogenic microorganisms (PPMs) as well as between endogenous and exog-
enous infections.
This chapter compares the traditional approach with the concept of carrier state
for classifying infections, mainly pneumonia, in critically ill patients.
L. Silvestri (&)
4.2 Traditional Approach and Carrier-State

Classification of ICU Infection
4.2.1 Traditional Approach
There are two standard means of classifying infections: the Gram-staining

technique, which groups both microorganisms and infections into Gram-negative and
Gram-positive categories; and incubation time, which distinguishes community from
nosocomial infections. The in vitro staining method is still used to distinguish
Gram-positive from Gram-negative bacteria and to classify microorganisms and
ICU infections. However, there is no correlation between the Gram reaction of a
particular microorganism and its pathogenicity (Chap. 3). For example, among
Gram-positive cocci present in the oropharynx of critically ill ICU patients,
such as enterococci, coagulase-negative staphylococci (CNS), viridans strep-
tococci, methicillin-sensitive Staphylococcus aureus (MSSA) and Streptococcus
pneumoniae, only MSSA and S. pneumoniae generally cause lower airway
infections, suggesting a different pathogenicity among Gram-positive bacteria.
Similarly, mortality rates are higher in ICU patients with a lower airway
infection due to Pseudomonas aeruginosa than with Haemophilus influenzae,
both of which are aerobic Gram-negative bacilli (AGNB) [3].
Studies on the prevalence of infections in the ICU include the usual definition
of hospital-acquired infections (i.e., nosocomial) issued by the Centers for Disease
Control and Prevention (CDC) [4]: for an infection to be defined as nosocomial
there must be no evidence that the infection was present or incubating at the time
of hospital admission. Based on this definition, in the early 1990s, the European
Prevalence of Infection in Intensive Care (EPIC) study distinguished each infec-
tion as one of the following [5]:
community acquired: an infection occurring in the community and manifest on
admission to the ICU;
hospital acquired: an infection manifest on admission to the ICU and deemed to
be related to the present hospital admission;
ICU acquired: an infection originating in the ICU but not clinically manifest at
the time of ICU admission.
The EPIC study showed that approximately 45% of enrolled patients were
infected; nearly half (46%) of them acquired their infection in the ICU. The
vagueness of this classification prompted investigators to adopt incubation times
in order to improve the distinction between infections acquired in the community
and/or hospital/ward from those acquired during the patients ICU stay. Arbitrary
time cutoffs varying between 24 h and 7 days [614] were chosen to distinguish
ICU from non-ICU infections. Italian authors introduced early-onset and late-
onset concepts in accordance with a cutoff of 4 days [13, 15] and emphasized
that infections developing in the first 4 days could not be considered as ICU acquired
[16]. Interestingly, 15 years after the EPIC study [5], the authors of the EPIC II study
[17] did not focus their analysis on infection classification because they were
4 Classification of ICU Infections 43
concerned that it might be difficult to distinguish between community-, hospital-,

and ICU-acquired infections. This concept was also embraced by the CDC, which
recently dismissed the previous classification as obsolete and replaced it with the
term healthcare-associated infection rather than nosocomial [18]. CDC defined
healthcare-associated infection as a localized or systemic condition resulting
from an adverse reaction to the presence of an infectious agent(s) or its toxin(s),
without any evidence that the infection was present or incubating at the time of
admission to the acute care setting. Notwithstanding, the time cutoff method has
not been completely withdrawn, and the following classification of pneumonia in
the ICU is still in use [1921]:
community-acquired pneumonia (CAP): pneumonia present at hospital/ICU
admission in patients who do not meet the criteria for healthcare-associated
pneumonia (HCAP);
HCAP: pneumonia present at hospital/ICU admission in patients with risk
factors of multidrug-resistant pathogens because of prior contact with a health
care environment (at least one of the following risk factors: hospitalization for
C2 days in an acute care facility within 180 days of infection, residence in a
nursing home or long-term care facility, antibiotic therapy, chemotherapy,
wound care within 30 days of current infection, hemodialysis treatment
at a hospital or clinic, home infusion therapy or home wound care, family
member with infection due to multidrug-resistant bacteria, significant immune
suppression);
hospital-acquired pneumonia (HAP): pneumonia occurring typically C48 h
after hospital admission in a nonintubated patient;
ventilator-associated pneumonia (VAP): pneumonia occurring typically C48 h
after hospital admission and endotracheal intubation and/or mechanical venti-
lation that was not present before intubation.
This alphabet soup of pneumonia, i.e., CAP, HCAP, HAP and VAP, is
increasingly complex [22]. The definitions reflect the assumption that all pneu-
monias occurring after 48 h of ICU stay are acquired in the unit and are due to
microorganisms transmitted via careers hands, and this substantially magnifies the
problem of defining ICU-acquired infections.
4.2.2 Carrier-State Approach
The traditional approach has been challenged by the carrier state concept [2].
Carriage or carrier state exists when the same strain is isolated from at least two
consecutive surveillance samples (e.g., throat and rectal swabs) from an ICU
patient, at any concentration, over a period of at least 1 week [23]. A surveillance
set comprises throat and rectal swabs taken on admission and twice weekly
afterward (e.g., Monday and Thursday). Diagnostic or clinical samples are samples
from internal organs that are normally sterile, such as lower airways, blood,
Table 4.1 Classification of infections occurring in the ICU
Infection PPMs Timing Frequency (%) Preventing maneuver

Primary endogenous 6 normal \1 week 55 Parenteral antibiotics
9 abnormal
Secondary endogenous 9 abnormal [1 week 30 Enteral antibiotics in
throat and gut; hygiene
Exogenous 9 abnormal Any time during 15 Topical antibiotics
ICU stay and hygiene
PPMs potentially pathogenic microorganisms
Normal PPMs: S. pneumoniae, H. influenzae, Moraxella catarrhalis, Candida albicans, S. aureus,
Escherichia coli; abnormal PPMs: Klebsiella, Proteus, Morganella, Enterobacter, Citrobacter,
Serratia, Acinetobacter, Pseudomonas species and methicillin-resistant S. aureus
bladder and skin lesions and are only taken on clinical indication with the aim of
microbiologically proving a diagnosis of inflammation, either generalized or local
[23]. Knowledge of the carrier state, together with diagnostic cultures, allows the
distinction between the three types of infection occurring in the ICU [2, 24]
(Table 4.1):
1. Primary endogenous infections are the most frequent infections in the ICU;
the incidence varies between 50% and 85%, depending on the population
studied and their degree of immunosuppression. They are caused by both
normal and abnormal PPMs imported into the ICU by the patient in the
admission flora. These episodes of infection generally occur early, during
the first week of ICU stay. S. pneumoniae, H. influenzae and S. aureus are
the etiological agents in previously healthy individuals requiring intensive
care following an acute event, such as (surgical) trauma, pancreatitis, acute
hepatic failure and burns. Abnormal AGNB can cause primary endogenous
infections in patients with previous chronic underlying disease, such as
severe chronic obstructive pulmonary disease, following acute deterioration
of the underlying disease. Adequate parenteral antibiotics given immediately
on admission to the ICU reduce the incidence of primary endogenous
infection.
2. Secondary endogenous infections are invariably caused by one or more of eight
abnormal AGNB as well as methicillin-resistant S. aureus (MRSA), accounting
for one-third of all ICU infections. This type of infection, in general, occurs
after 1 week in the ICU. These PPMs are first acquired in the oropharynx and
subsequently in the gut. The topical application of nonabsorbable antimicro-
bials polymyxin E/tobramycin/amphotericin B has been shown to control
secondary endogenous infection.
3. Exogenous infections are caused by abnormal hospital PPMs (15%) and may
occur at any time during the patients stay in the ICU. Typical examples are
Acinetobacter lower airway infection following the use of contaminated
Detect the patients carrier Confirm the microorganisms

state on admission and twice causing the infection by
weekly by surveillance diagnostic samples
cultures
Is the PPM causing infection

Yes carried by the patient in No
throat and/or gut?
Endogenous Exogenous
infection
infection
Is the PPM carried

in the patients
admission flora?
Yes No
PPM acquired during treatment in the

Primary carriage ICU, leading to secondary or
supercarriage
Primary
endogenous
Secondary
infection
endogenous
infection
Fig. 4.1 Flowchart for classifying infections in the ICU using knowledge of the carrier state
(PPM potentially pathogenic microorganism)
ventilation equipment, cystitis caused by Serratia associated with urinometers,

or a MRSA tracheobronchitis in a tracheostomized patient. Surveillance sam-
ples are negative for microorganisms that readily appear in diagnostic samples.
High levels of hygiene and topical antibiotics are required to control these
infections.
According to this criterion, only secondary endogenous and exogenous infec-
tions are labeled ICU-acquired infections, whereas primary endogenous infections
are considered to be imported infections. Figure 4.1 depicts infection classification
based on knowledge of the carrier state.
4.3 Evidence Behind Time and Carriage Classification

Systems of ICU Infections
4.3.1 Time Classification
Evidence that infections occurring on or at a specific time after ICU admission are
attributable solely to microorganisms transmitted via careers handsand hence
acquired during the ICU stayis limited. Classifications based on time have been
developed following the common experience of specific incubation times associated
with highly pathogenic microorganisms due to their high intrinsic pathogenicity
(or virulence) [25]. However, patients requiring intensive care develop infections with
PPMs, including MRSA and AGNB and with low-level pathogens such as CNS and
enterococci, due to the severity of their illness and associated immunosuppression
[26]. The severity of illness rather than microorganism virulence is the most important
factor for the conversion of the normal into the abnormal carrier state and will
determine the time at which a potential or a low-level pathogen will cause infection.
According to the pathogenesis of ICU-acquired infections, PPM acquisition is
followed by carriage and overgrowth of that microorganism before colonization
and infection of an internal organ occurs. Undoubtedly, this process takes more
than 24 days. Accordingly, a lower respiratory tract infection due to a PPM
already carried in the throat and/or gut on admission and that develops in a
ventilated trauma patient after 3, 4 or even 10 days from ICU admission, cannot
be considered as ICU acquired. The concept of early- and late-onset infection,
mainly pneumonia, remains accepted in general [21, 27]. Moreover, early-onset
nosocomial pneumonia is believed to be due primarily to normal flora, such as
H. influenzae, MSSA and S. pneumoniae, whereas late-onset nosocomial pneu-
monia is mainly caused by higher-level antibiotic-resistant AGNBs, such as
P. aeruginosa, Acinetobacter species, or MRSA [28]. These statements are not
generalizable to all ICU populations; patients with chronic underlying diseases, such
as diabetes, alcoholism, chronic obstructive pulmonary disease and liver disease,
may carry abnormal flora on ICU admission, including AGNB and/or MRSA, in the
throat and gut. This is why some authors found no difference in microorganisms
between early- and late-onset infections [28]. Moreover, it is unknown what is the
best cut-off time by which to separate early- from late-onset pneumonia, as it is
unknown how long it takes to develop pneumonia after aspiration of microorgan-
isms [27]. Therefore, many clinicians, in prolonging the cutoff time, implicitly
recognized the inaccuracy of the time classification method to determine whether an
infection was imported or was really nosocomial [13, 14, 28].
4.3.2 Carriage Classification
In contrast, knowledge of the carrier state at the time of admission and throughout
the ICU stay is indispensable in distinguishing infections due to imported PPMs
(i.e., primary endogenous) from infections due to bacteria acquired on the unit
(i.e., secondary endogenous and exogenous). Only secondary endogenous and

exogenous infections are true ICU-acquired infections, as the origin of the
causative bacteria is outside the ICU patientin the ICU environment. In the
case of the secondary endogenous infections, the microorganism acquired in
the unit goes through a digestive tract phase, but this does not apply to exog-
enous infections.
Six studies prospectively evaluated the accuracy of the 48-h time cutoff using
carriage as the gold standard (Table 4.2) [2934]. More than 2,000 patients with
1,000 infection episodes were evaluated; 71% of all infection episodes were
classified as nosocomial, as they occurred after 48 h. In contrast, using the carrier
state criterion, of the 1,000 infection episodes, 612 (61%) were due to bacteria
imported into the ICU, whereas 388 (39%) of all infections were caused by
microorganisms transmitted in the unit via careers hands. This figure was similar,
or even higher in the pediatric population [31, 33, 34]. Moreover, two cohort
studies assessed the time cutoff that was most in line with the carrier-state concept
[30, 33]: a period ranging from 7 to 10 days (depending on the population studied)
identified more accurately ICU-acquired infections than did the 48-h cutoff,
although time was a less reliable method of identifying imported and ICU-acquired
infections.
4.4 Impact of Time and Carrier-State Classification

of ICU Infections
Although the CDC introduced a new classification of infections in which the

concept that infections may originate from endogenous or exogenous sources [18],
the 48-h time cutoff is still used to distinguish different types of infections, namely,
pneumonia [19]. This approach implies that most pneumonias occurring in the
ICU are nosocomial due to microorganisms transmitted via careers hands, except
pneumonias established in the first 2 days. The 48-h time cutoff is also responsible
for blaming staff for almost all infections occurring in the ICU, for initiating
expensive transmission investigations, and for reinforcing handwashing to control
transmission. However, community-acquired and healthcare-associated infections,
which are present or incubating at the time of ICU admission, cannot be prevented
by handwashing [35].
These concepts are in contrast to data from the six previous studies [2934].
Table 4.2 shows that 61% of all ICU infections are primary endogenous, i.e., due
to microorganisms not related to the ICU ecology, and develop during the first
week of ICU stay. The remaining 39% are true ICU-acquired infections and
develop after 1 week in the unit. Handwashing cannot be expected to control
primary endogenous infections, as it fails to clear oropharyngeal and gastroin-
testinal carriage of PPMs present in the patient on arrival at the ICU. Being
inherently active solely on transmission, handwashing cannot reduce the major
infection problem of primary endogenous infection, as transmission is not involved
in this type of infection [35]. Additionally, handwashing does not influence the
48
Table 4.2 Classification of ICU infections using the 48-h time cutoff compared with carrier-state criteria
Author No. of No. of infection Classification of infection episodes

patients episodes
Carriage Nosocomial
(48-h cutoff)
Primary Secondary Exogenous [n (%)] Total secondary [n (%)]
endogenous endogenous endogenous and
[n (%)] [n (%)] exogenous
[n (%)]
Murray et al. [29] 21 12 6 (50) 6 (50) 0 6 (50) 9 (75)
Silvestri et al. [30] 117 74 44 (60) 17 (23) 13 (17) 30 (40) 59 (80)
Petros et al. [31] 52 18 15 (85) 2 (10) 1 (5) 3 (15) 15 (83)
de la Cal et al. [32] 56 37 21 (57) 14 (38) 2 (5) 16 (43) 30 (81)
Silvestri et al. [33]
Adult 130 27 14 (52) 10 (37) 3 (11) 13 (48) 19 (70)
Pediatric 400 40 32 (80) 4 (10) 4 (10) 8 (20) 26 (65)
Sarginson et al. [34] 1,241 792 480 (61) 42 (5) 270 (34) 312 (39) 547 (69)
Total 2,017 1,000 612 (61) 95 (9) 293 (30) 388 (39) 705 (71)
For each study, the same episodes of infection are classified both with the carrier-state concept and with the traditional 48-h cutoff. ICU-acquired, secondary
endogenous and exogenous infections based on knowledge of the carrier state; nosocomial, infection episodes of the previous columns reclassified using the
traditional 48-h time cutoff. The two columns on the right show comparison between the two types of classifications
L. Silvestri et al.
patients immune status. Detecting primary endogenous infection as the major

infection problem in the ICU avoids blaming health-care workers for most ICU
infection, for which they are not responsible, and prevents unnecessary expensive
cross-infection investigations. Finally, in strictly identifying the primary endoge-
nous and the nosocomial problem of secondary endogenous and exogenous
infections, surveillance of both infection and carriage allows the intensivist to start
with the appropriate prevention measures, including selective decontamination of
the digestive tract (SDD) [24].
4.5 Conclusions
Most ICU infections are due to microorganisms carried by the patient on admis-
sion to the unit. The difference in philosophy between the traditionalists and those
who advocate the carriage-state method for classifying ICU infections is that the
former focus on preventing transmission of all microorganisms via careers hands
in order to control all Gram-positive and Gram-negative infections occurring after
2 days of ICU stay. However, we believe that ICU patients may benefit from an
infection control program that includes surveillance of both carriage and infection.
In detecting abnormal carriage and overgrowth, surveillance cultures are indis-
pensable for identifying a subset of patients at high risk of infection. Awareness of
carriage in long-stay patients can provide more insights into the epidemiology of
infection. The true nosocomial infection problem (i.e., secondary endogenous and
exogenous infections) is detected easily and early. A regular audit of patients with
nosocomial infections only may be useful, as the combination of secondary
endogenous and exogenous infections may highlight a transmission problem in
the ICU.
References
1. Spencer RC (1996) Definitions of nosocomial infections: surveillance of nosocomial
infections. Baillieres Clin Infect Dis 3:237252
2. van Saene HKF, Damjanovic V, Murray AE et al (1996) How to classify infections in
intensive care unitsthe carrier state, a criterion whose time has come? J Hosp Infect
33:112
3. Park DR (2005) The microbiology of ventilator associated pneumonia. Respir Care
50:742763
4. Garner JS, Jarvis WR, Emori TG et al (1988) CDC definitions for nosocomial infections,
1988. Am J Infect Control 16:128140
5. Vincent J-L, Bihari DJ, Suter PM et al (1995) The prevalence of nosocomial infection in
intensive care units in Europe. JAMA 274:639644
6. McGowan JE, Barnes MW, Finland M (1975) Bacteremia at Boston city hospital: occurrence
and mortality during 12 selected years (19651972), with special reference to hospital-
acquired cases. J Infect Dis 132:316335
7. Potgieter PD, Hammond JMJ (1992) Etiology and diagnosis of pneumonia requiring
ICU-admission. Chest 101:199203
8. Estes RJ, Meduri GV (1995) The pathogenesis of ventilator-associated pneumonia.

I. Mechanisms of bacterial transcolonization and airway inoculation. Intensive Care Med
21:365383
9. Pugin J, Auckenthaler R, Lew DP, Suter PM (1991) Oropharyngeal decontamination
decreases incidence of ventilator-associated pneumonia: a randomized, placebo-controlled,
double-blind clinical trial. JAMA 265:27042710
10. Chevret S, Hemmer M, Carlet J, Langer M (1993) Incidence and risk factors of pneumonia
acquired in intensive care units: results from a multicenter prospective study on 996 patients.
11. Rolando N, Gimson A, Wade J et al (1993) Prospective controlled trial of selective parenteral
and enteral antimicrobial regimen in fulminant liver failure. Hepatology 17:196201
12. Korinek AM, Laisne MJ, Nicolas MH et al (1993) Selective decontamination of the digestive
tract in neurosurgical intensive care units patientsa double-blind, randomized, placebo-
controlled study. Crit Care Med 21:14661473
13. Langer M, Cigada M, Mandelli M et al (1987) Early onset pneumonia: a multicenter study in
intensive care units. Intensive Care Med 13:342346
14. Trouillet JL, Chastre J, Vuagnat A et al (1998) Ventilator-associated pneumonia caused by
potentially drug-resistant bacteria. Am J Respir Crit Care Med 157:531539
15. Antonelli M, Moro ML, DErrico RR et al (1996) Early and late onset bacteremia have
different risk factors in trauma patients. Intensive Care Med 22:735741
16. Mandelli M, Mosconi P, Langer M et al (1986) Is pneumonia developing in patients in
intensive care always a typical nosocomial infection? Lancet 2:10941095
17. Vincent JL, Rello J, Marshall J et al (2009) International study of the prevalence and
outcomes of infection in intensive care units. JAMA 302:23232329
18. Horan TC, Andrus M, Dudeck MA (2008) CDC/NHSN surveillance definition of health care-
associated infection and criteria for specific types of infections in the acute care setting. Am J
Infect Control 36:309332
19. Morrow LE, Kollef MH (2010) Recognition and prevention of nosocomial pneumonia in the
intensive care unit and infection control in mechanical ventilation. Crit Care Med 38(Suppl
8):S352S362
20. Masterton RG, Galloway A, French G et al (2008) Guidelines for the management of
hospital-acquired pneumonia in the UK: report of the working party on hospital-acquired
pneumonia of the British Society for antimicrobial chemotherapy. J Antimicrob Chemother
62:534
21. Niederman MS, Craven DE, Bonten MJ et al (2005) Guidelines for the management of adults
with hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J
Respir Crit Care Med 171:388416
22. Anand L, Kollef MH (2009) The alphabet soup of pneumonia: CAP, HAP, HCAP, NHAP,
VAP. Semin Respir Crit Care Med 30:39
23. Sarginson RE, Taylor N, van Saene HKF (2001) Glossary of terms and definitions. Curr
Anaesth Crit Care 12:25
24. van Saene HKF, Petros AJ, Ramsay G, Baxby D (2003) All great truths are iconoclastic:
selective decontamination of the digestive tract moves from heresy to level 1 truth. Intensive
Care Med 29:677690
25. Gorbach SL, Bartlett JG, Blacklow NR (1998) Infectious diseases, 2nd edn. Saunders,
Philadelphia
26. Leonard EM, van Saene HKF, Stoutenbeek CP et al (1990) An intrinsic pathogenicity index
for microorganisms causing infection in a neonatal surgical unit. Microbiol Ecol Health Dis
3:151157
27. Torres A, Ewig S, Lode H et al (2009) Defining, treating and preventing hospital acquired
pneumonia: European perspective. Intensive Care Med 35:929
28. Gastmeier P, Sohr D, Geffers C et al (2009) Early- and late-onset pneumonia: is this still a
useful classification? Antimicrob Agents Chemother 53:27142718
29. Murray AE, Chambers JJ, van Saene HKF (1998) Infections in patients requiring ventilation
in intensive care: application of a new classification. Clin Microb Infect 4:94102
30. Silvestri L, Monti-Bragadin C, Milanese M et al (1999) Are most ICU infections really
nosocomial? a prospective observational cohort study in mechanically ventilated patients.
J Hosp Infect 42:125133
31. Petros AJ, OConnell M, Roberts C et al (2001) Systemic antibiotics fail to clear
multiresistant Klebsiella from a pediatric intensive care unit. Chest 119:862866
32. de la Cal MA, Cerda E, Garcia-Hierro P et al (2001) Pneumonia in severe burns: a
classification according to the concept of the carrier state. Chest 119:11601165
33. Silvestri L, Sarginson RE, Hughes J et al (2002) Most nosocomial pneumonias are not due to
nosocomial bacteria in ventilated patients: prospective evaluation of the accuracy of the 48 h
time cut-off using carriage as the gold standard. Anaesth Intensive Care 30:275282
34. Sarginson R, Taylor N, Reilly N et al (2004) Infection in prolonged pediatric critical illness: a
prospective four year study based on knowledge of the carrier state. Crit Care Med
32:839847
35. Silvestri L, Petros AJ, Sarginson RE et al (2005) Handwashing in the intensive care unit: a
big measure-with modest effects. J Hosp Infect 59:175179
Gut Microbiology: Surveillance
Samples for Detecting the Abnormal 5
Carrier State in Overgrowth
H. K. F. van Saene, G. Riepi, P. Garcia-Hierro,

B. Ramos and A. Budimir
5.1 Introduction
Critical illness impacts all organ systems, such as lungs, heart and gut. The gut also
includes the vast living microbial tissue of the indigenous, mainly anaerobic, flora.
This enormous bacterial tissue is embedded in the mucous layer and covers the
inner wall of the gut. Amongst the aerobic Gram-negative bacilli (AGNB), only the
indigenous Escherichia coli is carried by healthy people in the gut. Critical illness
converts the normal carrier state of E. coli into carriage of abnormal AGNB,
including Klebsiella, Enterobacter and Pseudomonas species [1], and methicillin-
resistant Staphylococcus aureus (MRSA) [2]. It is hypothesised that receptors for
AGNB and MRSA are constitutively expressed on the mucosal lining but are
covered by a protective layer of fibronectin in the healthy mucosa. Significantly
increased levels of salivary elastase have been shown to precede AGNB carriage in
the oropharynx in post-operative patients and the elderly [3, 4]. It is probable that in
individuals with both acute and chronic underlying illness, activated macrophages
release elastase into mucosal secretions, thereby denuding the protective fibronectin
layer. It is thought that this possible mechanism is a deleterious consequence of
the inflammatory response encountered during and after illness. Critical illness
profoundly impacts body flora in two ways: it induces qualitative changes from
normal to abnormal flora [1, 2], as well as quantitative changes from low- to
high-grade carriage or gut overgrowth defined as C2+ or C105 potential pathogens
per millilitre of saliva and/or gram of faeces [5]. Overgrowth concentrations of both
normal and abnormal flora in surveillance samples are frequently found on
H. K. F. van Saene (&)

Institute of Ageing and Chronic Diseases, University of Liverpool,
Duncan Building, Liverpool, UK
54 H. K. F. van Saene et al.
admission to the intensive care unit (ICU), accounting for the large percentage of
primary endogenous infections [6, 7] (Table 4.1). Abnormal flora is often acquired
during treatment in the ICU. Acquisition invariably leads to abnormal carriage due
to critical illness. Most iatrogenic interventions in the patient requiring intensive
care, including mechanical ventilation, promote quantitative changes from
low- to high-grade carriage or overgrowth. Gut protection using H2 antagonists and
antimicrobials are commonly applied in the critically ill. H2 antagonists increase
gastric pH, thereby impairing the gastric acidity barrier [8]. Antimicrobials that are
active against the indigenous, mainly anaerobic, flora, which are excreted via
bile into the gut, may disturb gut ecology [9]. Integrity of both physiology and flora
is essential for the individuals defence against AGNB carriage. Impairment of
these two factors promotes overgrowth of abnormal, potentially pathogenic,
microorganisms (PPM), such as AGNB in concentrations of C2+ or C105 colony
forming units (CFU) per millilitre or gram of faeces.
Gut overgrowth of abnormal flora is not only a marker of critical illness, but it
harms the patient, as it is a disease in itself. In addition, gut overgrowth of
abnormal flora has a major epidemiological impact on the other patients in the ICU
as well as on the ICU environment.
5.2 Clinical Impact of Gut Overgrowth
Gut overgrowth harms the critically ill in four ways:

1. Immunosuppression. Overgrowth of abnormal AGNB (and associated endo-
toxin) impairs systemic immunity due to generalised inflammation following
absorption of AGNB and/or endotoxin [10];
2. Inflammation. Overgrowth of abnormal AGNB and/or endotoxin has been
shown to lead to cytokinaemia and inflammation of major organ systems [11];
3. Infection. There is a quantitative relationship between surveillance and diag-
nostic samples; as soon as there is overgrowth in surveillance samples, the
diagnostic samples become positive, which is the first stage in the development
of infection [12];
4. Antimicrobial resistance. The abnormal carrier state in overgrowth concen-
trations guarantees increased spontaneous mutation, leading to polyclonality
and antibiotic resistance [13].
5.3 Epidemiological Impact of Gut Overgrowth
The higher the salivary and faecal concentrations of AGNB and MRSA, the higher
the possibility of PPM transmission via carers hands [14, 15]. PPM acquisition
invariably leads to carriage, as the critically ill are unable to clear the acquired
AGNB and MRSA. Carriers of abnormal bacteria in overgrowth shed these
microorganisms into the environment and determine the contamination level of the
inanimate environment, including beds, tables, telephones and floors [16, 17].
5 Gut Microbiology 55
5.4 Definitions
5.4.1 Surveillance Samples
Surveillance samples are defined as samples obtained from body sites where PPM
may potentially be carried, i.e. the digestive tract, comprising oropharyngeal and
rectal cavities [18]. Surveillance cultures should be distinguished from surface and
diagnostic samples.
5.4.2 Surface Samples
Surface samples are taken from the skin, such as axilla, groin and umbilicus, and from
the nose, eye and ear. They do not belong to a surveillance sampling protocol because
positive surface swabs merely reflect the oropharyngeal and rectal carrier states.
5.4.3 Diagnostic Samples
Diagnostic samples are from internal organs that are normally sterile, such as lower
airways, blood, bladder, and skin lesions. They are only taken on clinical indication.
The endpoint of diagnostic samples is clinical, as they aim to prove microbiologically
a clinical diagnosis of inflammation, both generalised and/or local.
5.5 Endpoints of Surveillance Samples
The aim of obtaining surveillance cultures is to determine the microbiological

endpoint of the abnormalin overgrowth concentrationscarrier state. Carriage
or a carrier state exists when the same bacterial strain is isolated from at least two
consecutive surveillance samples of the ICU patient in any concentration over a
period of at least 1 week. Carriage implies persistent presence of a PPM and is
distinguished from acquisition or transient presence. Persistent presence is the
forerunner and virtually always precedes overgrowth (C2+ or C105) [57].
Surveillance samples are not useful for diagnosing infection of lungs, blood,
bladder, or wounds. Diagnostic samples are required for this purpose.
5.6 Sampling for Surveillance Purposes
5.6.1 Which Patients?
Only the most critically ill patients require intensive microbiological monitoring
using surveillance samples to detect the abnormal carrier state of AGNB and
MRSA in overgrowth concentrations. Due to the severity of their illness, these
patients require intensive care, including mechanical ventilation, for a minimum of
3 days. Critical illness related carriage in overgrowth [CIRCO] is often present on

admission. In general, they have impaired gut motility and, hence, are at high risk
of developing throat and gut overgrowth during treatment on ICU.
5.6.2 What Samples?
A surveillance programme for this type of patient includes samples from both
oropharynx and gut. Potential pathogens carried in the throat and gut cause pneu-
monia and septicaemia, respectively [19]. These two serious infections are respon-
sible for a high rate of mortality. Potential pathogens present in overgrowth in the
throat and gut are implicated in transmission via the hands of carers, in particular, in
outbreak situations. A throat and rectal swab are taken to detect oropharyngeal
and gut carriage of AGNB and MRSA. Rectal swabs must be coated with stool.
As MRSA has an affinity for the skin, skin is sampled only if lesions are present.
5.6.3 When?
Surveillance sets are obtained on admission and thereafter twice weekly

(e.g. Monday, Thursday) throughout the ICU treatment to distinguish carriage due
to PPM imported in the admission flora (import; primary carriage) from carriage
due to ICU-associated PPM in the oropharynx or gut acquired during treatment in
ICU (nosocomial; secondary; super carriage).
5.7 Microbiological Procedures
Throat and rectal swabs are processed qualitatively and semiquantitatively to

detect the level of carriage of the two types of target microorganisms: AGNB and
MRSA [20]. Two solid-media, MacConkey (AGNB) and a staphylococcal
(MRSA) agar, are inoculated using the four-quadrant method (Fig. 5.1). Each
swab is streaked onto the two solid media. All cultures are incubated aerobically at
37C. The MacConkey plate is examined after one night, the plate for MRSA after
2 nights. A semiquantitative estimation is made by grading growth density on a
scale of 1+ to 4+, as follows (Table 5.1): growth in the first quadrant of the solid
plate 1+ ([103 CFU/ml), in the second quadrant 2+ ([105 CFU/ml), in the third
quadrant 3+ ([107 CFU/ml) and on the entire plate 4+ ([109 CFU/ml). Macro-
scopically distinct colonies are isolated in pure culture. Standard methods for
identification, typing, and sensitivity patterns are used for all microorganisms.
5.8 Interpretation of Surveillance Samples
Surveillance cultures allow the intensive care specialist to distinguish normal from
abnormal carrier state, overgrowth from low-level carriage and endogenous from
exogenous infections in combination with diagnostic samples.
1*
1st quadrant
+1: > 103 CFU/g or ml
2*
4th quadrant 2nd quadrant

+4: > 109 CFU/g or ml +2: > 105 CFU/g or m
3rd quadrant
+3: > 107 CFU/g or ml
1* Inoculation of solid medium (1st quadrant)

2* Diluting using different loops
Fig. 5.1 Processing surveillance swabs using the four quadrant method
Table 5.1 Comparison of

Four-quadrant method: Growth density Dilution series:
surveillance (throat/rectal)
semiquantitative quantitative
swabs and (salivary/faecal)
swab method specimen method
specimens for detecting
the carriage level (growth 1+ Low 103
density) of aerobic
2+ Moderate 105
Gram-negative bacilli and
methicillin-resistant 3+ High 107
Staphylococcus aureus
4+ Very high 109
(MRSA) 5
Moderate growth density, i.e. C2 or C10 colony forming units,
reflects overgrowth
5.8.1 Normal Versus Abnormal Carriage
Surveillance swabs processed for one group of target microorganisms, AGNB,

using an inexpensive MacConkey agar plate, yield a positive or negative
result after 18 h of incubation. AGNB, including E. coli, are uncommon in the
oropharynx, whereas healthy people carry their own indigenous E. coli in
the intestine in concentrations varying between 103 and 106 CFU/ml or gram of
faeces. There are no other AGNB, including Klebsiella, Proteus, Morganella,
Enterobacter, Citrobacter, Serratia, Acinetobacter and Pseudomonas species, in
either the throat or gut. Interpreting the staphylococcal plate requires 2 nights of
incubation. About one-third of the healthy population carries methicillin-sensitive
S. aureus. MRSA isolation is always abnormal.
5.8.2 Low-Level Carriage Versus Overgrowth
Oropharyngeal and intestinal overgrowth is defined as C2+ or C105 micro-

organisms per millilitre of saliva and/or gram of faeces and is distinguished from
low-grade carriage of \2+ or \105 microorganisms [57]. Individuals with a
chronic disease, such as chronic obstructive pulmonary disease, generally carry
abnormal flora in low concentrations once the forced expiratory volume in 1 s
(FEV1) is\50% [21]. The low-level carrier status is mainly due to the presence of
clearing mechanisms, such as swallowing, chewing and peristalsis. However,
patients who require mechanical ventilation for a minimum of 3 days generally
have CIRCO on admission and often develop it during treatment in ICU [20].
Gut overgrowth harms the critically ill, as it causes immunosuppression [10],
inflammation [11], infection [12] and resistance [13].
5.8.3 Primary Versus Secondary (Super) Carriage
Knowledge of the carrier state at the time of admission (primary carriage) and
subsequently during treatment in the ICU (secondary or super carriage) is crucial
for managing infection in the unit.
5.8.3.1 Primary Carriage

Primary carriage of potential pathogens is defined as the carrier state of potential
pathogens present in the admission flora, i.e. imported into the ICU. The potential
pathogens include both normal and abnormal potential pathogens.
5.8.3.2 Secondary (Super) Carriage

Secondary carriage of potential pathogens is defined as the carrier state of potential
pathogens not present in the admission flora but acquired later and subsequently
carried during treatment in the ICU. Potential pathogens are invariably abnormal
bacteria, including AGNB and MRSA.
Primary and secondary (super) infections are preceded by primary endogenous
and secondary (super) carriage in overgrowth concentrations, respectively [57].
Exogenous infections are caused by potential pathogens not carried at all.
In general, exogenous infections are due to abnormal bacteria (Table 5.2).
Hygiene measures will only have an impact on infections due to externally
transmitted microorganisms, e.g. secondary endogenous (super) and exogenous
infection [22]. A primary endogenous infection caused by a potential pathogen
imported by the patient into the ICU in the admission flora can only be managed
effectively with knowledge of the carrier state. It is obvious that hand hygiene
fails to eradicate carriage of potential pathogens in throat and gut detected by
surveillance samples on admission.
To detect resistance at an early stage, surveillance cultures are more sensitive than
diagnostic cultures. Eighteen percent of critically ill patients carried AGNB resistant
to ceftazidime in their admission flora. Diagnostic cultures detected these resistant
Table 5.2 Carriage classification of severe infections of lower airways and blood
Infection PPM Timing Frequency(%) Manoeuvre

Primary 6 normal; 9 \1 week 55 Parenteral
endogenous abnormal antimicrobials
Secondary 9 abnormal [1 week 30 Hygiene and enteral
endogenous antimicrobials
Exogenous 9 abnormal Anytime during ICU 15 Hygiene and topical
treatment antimicrobials
PPM, potentially pathogenic microorganism; 6 normal PPM, Streptococcus pneumoniae,
Haemophilus influenzae, Moraxella catarrhalis, Candida albicans, Staphylococcus aureus,
Escherichia coli; 9 abnormal PPM, Klebsiella, Enterobacter, Citrobacter, Proteus, Morganella,
Serratia, Acinetobacter, Pseudomonas species and methicillin-resistant Staphylococcus aureus
(MRSA)
bacteria in only 5% of these patients [23]. A recent study from Trieste [7] reports
resistance figures of 8% in diagnostic cultures compared with 22% in surveillance
cultures. This difference is highly likely to be due to the observation that overgrowth
in the throat and gut is necessary to make diagnostic cultures positive [24].
However, information provided by surveillance cultures of throat and rectum
enables the intensivist to implement isolation and to reinforce hygiene measures
as soon as possible following admission. Two recent studies show that MRSA
and ceftazidime-resistant AGNB were identified in 23.8% and 52.1% of patients,
respectively, within the first 72 h of admission to the ICU [6, 25].
5.9 Role of Surveillance Samples in Infection

Control in the ICU Patient
Recent studies using surveillance cultures of throat and rectum to detect carrier
state demonstrate that only infections occurring after 1 week of ICU stay are
due to microbes transmitted via the hands of health-care workers [2630]. The
incidence varies between 15% and 45% depending on illness severity. Microor-
ganisms related to the ICU environment are first acquired in the oropharynx. In the
critically ill, oropharyngeal acquisition invariably leads to secondary carriage.
The subsequent build up to digestive tract overgrowth, which can then result in
colonisation of normally sterile internal organs, takes a few days. Finally, it is
the degree of immunosuppression of the ICU patient that determines the day of
colonization, leading to an established secondary endogenous or super infection.
The other type of ICU infection is the exogenous infection [3133] due to breaches
of hygiene. Causative bacteria are also acquired in the unit but are never present in
patient throat and/or gut flora. For example, long-stay patients, particularly those
who receive a tracheostomy in respiratory units, are at high risk of exogenous
lower-airway infections. Purulent lower-airway secretions yield a microorganism
that has never been previously carried by the patient in the digestive tract flora, or
indeed in their oropharynx. Although both the tracheostomy and the oropharynx
are equally accessible for bacterial entry, the tracheotomy tends to be the entry site
for bacteria that colonise/infect the lower airways.
However, primary endogenous infections cause the major infection problems,
and the microorganisms involved do not bear any relation to ICU ecology
[34, 35]. A recent study compared the traditional 48 h cutoff and the criterion
of the carrier state and found that the time cutoff significantly overestimated the
magnitude of the nosocomial problem [30]. This approach to the carrier state
may be more useful for interhospital comparison, as only infections due to
microorganisms acquired in the different units are compared, independent of
illness severity.
In identifying the right population with primary endogenous infections,
classification using the carrier state avoids blaming staff for all infections
occurring after 48 h for which they are not responsible. Knowledge of carrier
status thus prevents fruitless investigation of apparent cross-infection episodes.
Secondly, without surveillance samples, exogenous infections are impossible to
recognise, at least at an early stage when only diagnostic samples such as
tracheal aspirate, urine and blood have been tested. Finally, knowledge of the
carrier state using surveillance cultures on admission and twice weekly is an
effective strategy for early identification of carriers of multidrug-resistant
microorganismsincluding AGNB such as A. baumannii [7, 36], MRSA [6, 24]
and vancomycin-resistant enterococci [37]both on admission and during ICU
stay. Surveillance cultures, in particular of the oropharynx, that become positive
for a PPM during ICU stay reveal ongoing transmission and an impending outbreak
long before the diagnostic samples yield the outbreak strain [38]. This surveillance
strategy optimises targeted infection control interventionsincluding (1) hand
hygiene, (2) isolation, (3) personal protective equipment and (4) care of patient
equipmentto control transmission from one patient-carrier to another patient via
carers hands.
5.10 Future Lines of Research on Surveillance

Samples in the ICU Patient
Most infection surveillance programmes include all patients admitted to the ICU,
whether they stay a few days or 2 weeks [39, 40]. Including a large number of
relatively short-stay patients with a low risk of infection tends to dilute total rates
of infection by increasing the size of the denominator. However, whereas low
percentages look good to the hospital manager, they do not allow room for
improvement, i.e. detecting a significant reduction in infection rate following the
introduction of an intervention [39]. We believe that critically ill patients benefit
from a surveillance programme of both infection and of carriage [41, 42], in
particular, in combination with selective decontamination of the digestive tract
(SDD) [4345].
References
hospitalized patients. Emergence of Gram-negative bacilli. New Engl J Med 281:11371140
2. Chang FY, Singh N, Gayowski T et al (1998) Staphylococcus aureus nasal colonization in
3. Dal Nogare AR, Toews GB, Pierce AK (1987) Increased salivary elastase precedes Gram-
negative bacillary colonization in post-operative patients. Am Rev Respir Dis 135:671675
4. Palmer LB, Albulak K, Fields S et al (2001) Oral clearance and pathogenic oropharyngeal
colonization in the elderly. Am J Respir Crit Care Med 164:464468
5. van Saene HKF, Damjanovic V, Murray AE, de la Cal MA (1996) How to classify infections in
intensive care unitsthe carrier state, a criterion whose time has come? J Hosp Infect 33:112
6. Viviani M, van Saene HKF, Dezzoni R et al (2005) Control of imported and acquired
methicillin-resistant Staphylococcus aureus [MRSA] in mechanically ventilated patients: a
dose response study of oral vancomycin to reduce absolute carriage and infection. Anaesth
Intensive Care 33:361372
7. Viviani M, van Saene HK, Pisa F et al (2010) The role of admission surveillance cultures in
patients requiring prolonged mechanical ventilation in the intensive care unit. Anaesth
8. Hillman KM, Riordan T, OFarrell SM, Tabaqchali S (1982) Colonization of the gastric
contents in critically ill patients. Crit Care Med 10:444447
9. Vollaard EJ, Clasener HAL (1994) Colonization resistance. Antimicrob Agents Chemother
38:409414
11. Baue AE (1993) The role of the gut in the development of multiple organ dysfunction in
cardiothoracic patients. Ann Thoracic Surg 55:822829
12. van Uffelen R, van Saene HKF, Fidler V et al (1984) Oropharyngeal flora as a source of
bacteria colonizing the lower airways in patients on artificial ventilation. Intensive Care Med
10:233237
13. van Saene HKF, Taylor N, Damjanovic V, Sarginson RE (2008) Microbial gut overgrowth
guarantees increased spontaneous mutation leading to polyclonality and antibiotic resistance
in the critically ill. Curr Drug Targ 9:419421
14. Riley TV, Webb SAR, Cadwallader H et al (1996) Outbreak of gentamicin-resistant
Acinetobacter baumanii in an intensive care unit: clinical, epidemiological and micro-
biological features. Pathology 28:359363
15. Lin MY, Hayden MK (2010) Methicillin-resistant Staphylococcus aureus and vancomycin-
resistant enterococcus: recognition and prevention in intensive care units. Crit Care Med
38(suppl 8):S335S344
16. Go ES, Urban C, Burns J et al (1994) Clinical and molecular epidemiology of Acinetobacter
infections sensitive only to polymyxin B and sulbactam. Lancet 344:13291332
17. Bocher S, Skov RL, Knudsen MA et al (2010) The search and destroy strategy prevents
spread and long-term carriage of MRSA: results from follow-up screening of a large ST22
(E-MRSA) 15 outbreak in Denmark. Clin Microb Infect 16:14271434
18. Damjanovic V, van Saene HKF, Weindling AM (1994) The multiple value of surveillance
cultures: an alternative view. J Hosp Infect 28:7178
negative bacillary carriage in chronic obstructive pulmonary disease: relation to severity of
disease. Respir Med 93:540545
big measure with modest effects. J Hosp Infect 59:172179
23. DAgata EM, Venkataraman L, DeGirolami P et al (1999) Colonization with broad-spectrum
cephalosporin-resistant gram-negative bacilli in intensive care units during a nonoutbreak
period: prevalence, risk factors, and rate of infection. Crit Care Med 27:10901095
24. de la Cal MA, Cerda E, van Saene HKF et al (2004) Effectiveness and safety of enteral
25. Toltzis P, Yamashita T, Vilt L et al (1997) Colonization with antibiotic-resistant Gram-
negative organisms in a pediatric intensive care unit. Crit Care Med 25:538544
26. Murray AE, Chambers JJ, van Saene HKF (1998) Infections in patients requiring ventilation
in intensive care: application of a new classification. Clin Microbiol Infect 4:94102
27. Silvestri L, Monti Bragadin C, Milanese M et al (1999) Are most ICU-infections really
nosocomial? A prospective observational cohort study in mechanically ventilated patients.
28. Petros AJ, OConnell M, Roberts C et al (2001) Systemic antibiotics fail to clear multi-
drug-resistant Klebsiella from a pediatric ICU. Chest 119:862866
29. de la Cal MA, Cerda E, Garcia-Hierro P et al (2001) Pneumonia in patients with severe burns.
A classification according to the concept of the carrier state. Chest 119:11601165
30. Silvestri L, Sarginson RE, Hughes J et al (2002) Most nosocomial pneumonias are not due to
nosocomial bacteria in ventilated patients. Evaluation of the accuracy of the 48 h time cut-off
using carriage as the gold standard. Anaesth Intensive Care 30:275282
31. Hammond JMJ, Potgieter PD, Saunders GL et al (1992) Double blind study of selective
decontamination of the digestive tract in intensive care. Lancet 340:59
32. Morar P, Singh V, Makura Z et al (2002) Differing pathways of lower airway colonization
and infection according to mode of ventilation (endotracheal versus tracheostomy). Arch
Otolaryngol Head Neck Surg 128:10611066
33. Morar P, Makura Z, Jones AS et al (2000) Topical antibiotics on tracheostoma prevents
34. Stoutenbeek CP (1989) The role of systemic antibiotic prophylaxis in infection prevention in
intensive care by SDD. Infection 17:418421
35. Sirvent JM, Torres A, El-Ebiary M et al (1997) Protective effect of intravenously
administered cefuroxime against nosocomial pneumonia in patients with structural coma.
Am J Respir Crit Care Med 155:17291734
36. Corbella X, Pujol M, Ayats J et al (1996) Relevance of digestive tract colonization in
the epidemiology of nosocomial infections due to multiresistant Acinetobacter baumannii.
Clin Infect Dis 23:329334
37. Hendrix CW, Hammond JMJ, Swoboda SM et al (2001) Surveillance strategies and impact of
vancomycin-resistant enterococcal colonization and infection in critically ill patients. Ann
Surg 233:259265
38. Chetchotisakd P, Phelps CL, Hartstein AI (1994) Assessment of bacterial cross-transmission
as a cause of infections in patients in intensive care units. Clin Infect Dis 18:929937
39. Kollef MH, Sherman G, Ward S, Fraser VJ (1999) Inadequate antimicrobial treatment of
infections. Chest 115:462474
40. Richards MJ, Edwards JR, Culver DH et al (1999) Nosocomial infections in medical
intensive care units in the United States. Crit Care Med 27:887892
41. Langer M, Carretto E, Haeusler EA (2001) Infection control in ICU: back (forward) to
surveillance samples? Intensive Care Med 27:15611563
42. Silvestri L, van Saene HKF (2002) Surveillance of carriage. Minerva Anestesiol 68(Suppl 1):
S179S182
43. de Jonge E, Schultz MJ, Spanjaard L et al (2003) Effects of selective decontamination of the
digestive tract on mortality and acquisition of resistant bacteria in intensive care: a
randomised controlled trial. Lancet 362:10111016
45. Liberati A, DAmico R, Pifferi S et al (2009) Antibiotic prophylaxis to reduce respiratory
tract infections and mortality in adults receiving intensive care. Cochrane Database Syst Rev
CD000022
Part II
Antimicrobials
Systemic Antibiotics
6
A. R. De Gaudio, S. Rinaldi and C. Adembri
6.1 Introduction
Systemic antibiotics remain the main causative therapy for critically ill patients
with infection. This chapter aims to provide a clinical review of the antibiotics
available for systemic administration in the intensive care unit (ICU). Pharma-
cological and microbiological factors that affect antimicrobial administration
regimen in the critical patient are also discussed.
As new drugs have been developed to overcome antimicrobial resistance,
bacteria acquired new kinds of resistance in the endless war for survival. The
opportunity of using appropriate pharmacokinetic/pharmacodynamic (PK/PD)
parameters to optimize dosing in order not only to cure patients but also to reduce
the spreading of resistance is therefore becoming imperative, as there are now
fewer new therapeutic options than in the past.
Bacteria have the capacity to adapt to a wide range of conditions. Any
strategy aimed at destroying bacterial flora has resulted in a dramatic failure. It is
likely that in the future, the definitive answer to infectious diseases will stand not
only on the development of new antibiotics but also in immunomodulating
strategies. The goal of the intensivist challenged with an infectious disease will
include turning the relationship between bacterial flora and the host from
infection back to symbiosis.
C. Adembri (&)
Department of Medical and Surgical Critical Care,
Section of Anesthesiology and Intensive Care, University of Florence,
Azienda Ospedaliero-Universitaria Careggi, Florence, Italy
68 A. R. De Gaudio et al.
6.2 The Puzzle of Antimicrobial Activity
Antimicrobial activity is the result of many variables that fit together in the clinical
setting just as in a puzzle (Table 6.1), as it depends on the PD properties of the
antibiotic, the specific bacterial strain, the infection site, and the setting in which
the antimicrobial drug challenges the microorganism [13]. In the ICU, infections
are often caused by multiresistant strains and develop in patients who have
coexistent multiple organ dysfunction and impaired immune function, a clinical
setting in which optimal antimicrobial therapy (in terms of both activity spectrum
and administration dose and modality) is essential to improve outcome.
One of the main goals of antimicrobial therapy is that the antimicrobial reaches
and remains in the site of infection in a sufficient concentration and for a sufficient
time. Regarding the pharmacological action of antimicrobials (that is, killing
activity), this may be concentration dependent or time dependent [4]. Concen-
tration-dependent antibiotics kill bacteria at a greater rate and to a greater extent
with increasing antibiotic concentrations, whereas time-dependent antibiotics kill
bacteria at the same rate and to the same extent once an appropriate concentration
threshold has been achieved. Aminoglycosides, fluoroquinolones, clarithromycin,
azalides, ketolides, and metronidazole are considered concentration-dependent
antimicrobial drugs, whereas glycopeptides, clindamycin, natural macrolides,
b-lactams, linezolid, and quinupristin-dalfopristin are considered time-dependent
drugs. These differences in PD activity should result in different dosing regimens
for time- and concentration-dependent antibiotics.
Besides these PD considerations, the bactericidal activity of antimicrobial agent
is influenced by several factors correlated with infection site, including antibiotic
diffusion, local pH, bacterial load, phase of bacterial growth, and oxygen tension.
In recent years, antibiotic penetration into the infection site has become one of the
main factors that should be taken into account when choosing an antibiotic regi-
men. Linezolid and fluoroquinolones penetrate well into the lungs, and their use in
nosocomial pneumonia is associated with a significant clinical success rate.
Vancomycin has a poor lung penetration, and many clinical failures of this drug in
pneumonia due to strains sensitive to this antimicrobial may derive from its poor
lung disposition.
Several studies regarding the inoculum effect considered the effect of the bacterial
load on the minimal inhibitory concentration (MIC) of several antibiotics. Standard
laboratory inoculum for MIC determination is around 105 CFU/ml. The presence of
the inoculum effect is defined as an eightfold or greater increase in MIC on testing,
with an inoculum as high as 107108 CFU/ml, with higher inoculum better reflecting
most clinical settings. Aztreonam, piperacillin, cefotaxime, and cefoxitin show a
significant inoculum effect against susceptible strains. A significant inoculum effect
is associated with b-lactams with prevalent inhibition of penicillin-binding protein
3(PBP3) because, in the setting of a high inoculum, bacteria proliferation slows and
they produce less PBP3. Moreover, the inoculum effect is present when a bacterium
produces an enzyme able to destroy the tested antibiotic, because the enzymes
6 Systemic Antibiotics 69
Table 6.1 Pieces of the antimicrobial activity puzzle
Variables influencing antimicrobial effectiveness

Concentration dependency
Pharmacodynamic properties
Time dependency
Antibiotic penetration into the infection site
pH and oxygen tension at the infection site
Vascularization at the infection site
Phase of bacterial growth
Inoculum effect
Endotoxin release
Postantibiotic effect
released from the dead cells inactivate the antibiotic, which is why the combination
with b-lactamase inhibitors may prevent the inoculum effect observed with many
b-lactams. Among the clinical conditions mimicking a high inoculum are endocar-
ditis, meningitis, septic arthritis, osteomyelitis, abscesses, and deep-seated infec-
tions. In these conditions, antibiotics without an inoculum effect have been
demonstrated to be more effective.
Regarding the bactericidal activity of antibiotics, another important factor is the
presence or absence of a postantibiotic effect (PAE). PAE is the ability of an
antimicrobial drug to exert a persistent inhibitory effect on microorganism growth
after the drug has been completely removed. Usually, antibiotics with a main time-
dependent activity lack a significant PAE, although differences among the different
classes do exist. b-lactams, for example, show a significant PAE only against
Gram-positive organisms, although carbapenems may have a sustained PAE
against aerobic Gram-negative bacilli (AGNB). Agents that interfere with protein
or DNA synthesis, such as fluoroquinolones and aminoglycosides, usually show a
sustained PAE, mainly against AGNB. The PAE is unique to the pathogen and is
generally longer in vivo than in vitro. For example, PAE duration for aminogly-
cosides ranges from 0.5 to 8 h depending on the bacterial strain, the MIC,
the duration of exposure, and the relative concentration of the aminoglycoside.
The PAE may decrease with multiple dosing [13].
6.3 The Frame of the Puzzle: Critical Illness
Assuming antimicrobial activity is like a puzzle with many important pieces,

critical illness is like the frame of this puzzle, representing the context in which
antimicrobial activity challenges microorganisms. But it is not a passive frame,
because it continuously interacts with antimicrobial activity in many ways. In fact,
pathophysiological changes in critical illness affect antibiotic PK variables, with

possible effects on antibiotic efficacy and clinical outcomes. Critical illness often
changes the volume of distribution (Vd) and clearance (CL) of antibiotics,
parameters usually correlated to the hydrophilicity and lipophilicity of the anti-
biotic molecule. During sepsis, vascular endothelium derangement results in
maldistribution of blood flow and increased capillary permeability, with fluid shifts
from the intravascular compartment to the interstitial space [5, 6]. This usually
increases the Vd of hydrophilic drugs, decreasing their plasma concentration.
A similar effect on Vd of hydrophilic drugs is caused by mechanical ventilation,
hypoalbuminemia, extracorporeal circuits, postsurgical drains, and burns. Lipo-
philic drugs typically have a large Vd because of their partitioning into adipose
tissue or into cells. Therefore, the increase in Vd resulting from third spacing is
considered insignificant [7, 8].
Drug-elimination half-life (T1/2) is indirectly related to antibiotic CL and
directly related to its Vd and is therefore reduced with an increased drug CL
and increased with an increased Vd. In critically ill patients, cardiovascular
support with intravenously administered fluids, vasopressors, and inotropes
sustains a hyperdynamic state that, in the absence of significant organ dysfunction,
increases renal perfusion and creatinine clearance (CrCl) of hydrophilic
antibiotics. It follows that dose adjustment for hydrophilic antibiotics can be guided
by CrCl [9].
Protein binding is a factor that may influence the Vd and CL of many
antibiotics with high protein-binding affinity (such as ceftriaxone, oxacillin,
teicoplanin). In hypoalbuminemic states, as commonly occur in critically ill
patients, these drugs may have a 100% increased CL and Vd. With further
deterioration in the patients health status, significant myocardial depression can
occur, which leads to a decrease in organ perfusion and failure of the micro-
vascular circulation until multiple-organ dysfunction syndrome occurs, with renal
and/or hepatic dysfunction. In this phase of critical illness, there is usually a
decreased antibiotic CL and prolonged T1/2. Finally, antibiotic penetration at the
target site is impaired in patients with septic shock, being 510 times lower than
in healthy volunteers [7].
The above-mentioned causes of altered PKs in the ICU septic patients require
specific dosing strategies in this special patient population. Ongoing evaluations of
sickness severity can facilitate timely adjustment of antibiotic dosing [9]. Specific
consideration regarding each class of antibiotics is necessary when they are used in
critically ill patients.
6.3.1 Aminoglycosides
The kill characteristic of aminoglycosides is concentration dependent. Amino-

glycosides often have increased Vd in critically ill patients, which can result in a
decreased maximum concentration (Cmax). Vd has been shown to increase pro-
portionally with increasing sickness severity. Because of their narrow therapeutic
index, maximal weight-based once-day dosing is required to achieve adequate

Cmax:MIC ratios. PK variability and potential for adverse effects mandates mon-
itoring plasma aminoglycoside concentrations. Once-daily administration appears
to be as effective as multiple doses per day but with reduced risks of toxicity.
Therefore, multiple doses per day should only be considered for treating endo-
carditis or in neutropenic patients [7, 8].
6.3.2 b-Lactams
The b-lactam antibiotics are generally hydrophilic molecules, renally cleared,

with moderate-to-low protein binding (although ceftriaxone has high protein
binding). An improved PD profile in critically ill patients is obtained with either
more frequent dosing or extended or continuous infusions. This mode of
administration is especially useful in critically ill patients with high glomerular
filtration rate or increased Vd. The superiority of continuous infusions of
b-lactam antibiotics in comparison with intermittent administration is supported
by the fact that the former administration modality theoretically favors the
achievement of the optimal PD target. Reduced mortality from extended infu-
sions (4 h infusion every 8 h) of piperacillintazobactam has been shown in
critically ill patients with Pseudomonas aeruginosa infection [10]. In critical
illness, carbapenems develop increased Vd and higher CL. Extended infusion of
these drugs seems to be appropriate in critical patients because optimal activity
requires time above MIC [40%.
6.3.3 Glycopeptides
Glycopeptides are relatively hydrophilic drugs, with PD properties not completely

elucidated. It remains unclear whether AUC/MIC or time [MIC is the target
parameter to optimize activity. The benefit of continuous infusion, especially
against methicillin-resistant Staphylococcus aureus (MRSA), seems significant,
although data require further confirmation. Nonrenal CL of vancomycin occurs in
critical patients with acute renal failure. Therefore, empirical dosing based on CrCl
data with subsequent therapeutic drug monitoring of Cmin plasma concentrations
(1520 mg/L) are recommended because of the risk of nephrotoxicity with higher
Cmin ([20 mg/L) concentrations.
6.3.4 Fluoroquinolones
Fluoroquinolones are lipophilic antibiotics with extensive distribution and excel-

lent penetration into neutrophils and lymphocytes. Their Vd is minimally affected
in the critically ill patient. Fluoroquinolones display concentration-dependent kill
characteristics but also some time-dependent effects. A Cmax/MIC ratio of 10

seems to be the critical variable in predicting bacterial eradication, but in critically
ill patients, an area under the concentrationtime curve (AUC)/MIC [125 in case
of Gram-negative infection and [30 in case of Gram-positive infection is also
fundamental to achieve successful clinical outcome and to avoid the emergence of
resistant bacterial strains.
6.3.5 Linezolid
Linezolid is quite lipophilic, and it distributes widely into tissues and is mostly
metabolized hepatically before being cleared renally. No dose adjustment seems to
be necessary in renal or hepatic dysfunction. The killing characteristic is time
dependent, being optimal when the time above MIC is [4080%. In critical
patients, linezolid shows an increased Vd and CL, requiring appropriate dose
adjustment.
6.3.6 Tigecycline
Tigecycline is a lipophilic glycylcycline eliminated by biliary excretion. Its killing

characteristic is time dependent, but the AUC/MIC ratio is the target parameter for
efficacy because of its PAE. Critical illness does not seem to affect its PKs sig-
nificantly, but further clinical studies specifically performed in this patient popu-
lation are needed to confirm this.
6.3.7 Daptomycin
Daptomycin is a concentration-dependent antibiotic, and PK parameters that best

predict its efficacy are Cmax/MIC and AUC/MIC. Daptomycin is eliminated pri-
marily by glomerular filtration; people receiving daptomycin should be monitored
for skeletal muscle damage. Doses higher than those initially suggested (4 mg/kg
once daily) are necessary in severely ill or ICU patients.
6.3.8 Colistin
The polymyxin antibiotics are hydrophilic molecules with predominantly

concentration-dependent bacterial killing activity. Patient dosing should follow
patient weight and renal dysfunction considerations because of the risk of
nephrotoxicity [7].
6.4 Time-Dependent Antibiotics
Time-dependent antibiotic activity is correlated to T [ MIC; that is, the duration

of time that antibiotic concentrations exceed the minimum inhibitory concentra-
tion [1]. Larger doses increase the efficacy of these antibiotics, extending the time
during which the drug remains above the effective concentration rather than
increasing its absolute concentration. PD studies led to a shift of dosage regimens
toward more fragmented doses or even continuous infusion. The results of many
clinical studies recommend exceeding the MIC by 25 times for between 40 and
100% of the dosage interval when using time-dependent antibiotics. This time
should probably be further extended when using antibiotics that are highly protein
bound ([90%) [2].
Strategies to keep drug concentrations above the MIC for a longer period
include: multiple doses given frequently, administration in a continuous infusion,
use of agents with long serum half-lives, use of agents with active metabolites,
choice of agents with the lowest MIC toward the causative microorganism, and
concomitant administration of antibiotic elimination inhibitors (i.e., probenecid).
6.5 Concentration-Dependent Antibiotics
PD parameters predicting the performance of concentration-dependent antibiotics

are: AUC024/MIC (ratio of AUC during a 24-h dosing period to MIC) and
Cmax/MIC (ratio of maximum serum antibiotic concentration to MIC). In experi-
mental studies, a Cmax/MIC ratio above at least 810 was required to optimize
bactericidal activity of concentration-dependent antibiotics [1].
Clinical studies demonstrated a correlation between higher aminoglycoside
peak levels and improved clinical outcome. The dosage of aminoglycosides in case
of a once-daily strategy should be 57 mg/kg for gentamicin and tobramycin and
1530 mg/kg for amikacin. These dosages have been selected to reach a Cmax/MIC
ratio of 10 considering the usual MIC for susceptible P. aeruginosa strains. With
this causative pathogen, tobramycin is preferred because it usually has a lower
MIC. In patients with decreased renal function, the interval between doses should
be lengthened, but doses should not be decreased, so that a high Cmax/MIC ratio is
maintained and an appropriate drug-free period thus allowed [1, 2].
Also for fluoroquinolones, many studies confirmed the importance of a
Cmax/MIC ratio [10 as an important predictor of antimicrobial effect and clinical
outcome. Regarding the effect of the AUC024/MIC ratio on the antimicrobial
activity of fluoroquinolones, a ratio [100 seems to optimize fluoroquinolone
activity, whereas a lower AUC024/MIC ratio is associated with an increased risk
of antimicrobial resistance development. Improper dosing regimens of fluor-
oquinolones that do not produce AUC024/MIC ratios [100 or Cmax/MIC ratios
[810 may increase the risk of resistance in AGNB [1]. Also new macrolides
(clarithromycin, azalides, ketolides), colistin, and daptomycin have a concentra-
tion-dependent activity.
6.6 Tissue Penetration of Antibiotics
An important factor in determining a successful therapeutic response to antimi-

crobial drugs is maintaining the level of free antibiotic in extravascular fluid above
the MIC of the infecting organism [1, 3, 8]. Extravascular distribution, tissue
binding, and free drug concentration of an antibiotic can be predicted from serum
kinetics and serum protein binding [11].
Serum protein binding of antimicrobial drugs may have important conse-
quences on their action. In particular, high serum protein binding can reduce
antimicrobial activity, tissue distribution, and elimination of antibiotics. Binding
C80% have the potential to significantly reduce free drug levels and affect
therapeutic efficacy [2]. The extent of protein binding is a major factor
determining active drug concentration in serum and most extravascular fluids.
As a general rule, agents with minimal protein binding penetrate tissues better
than those that are highly protein bound, although the latter might be excreted
much faster. Among drugs that are \8085% protein bound, differences appear
to be of slight clinical importance [11]. The concentration of several plasma
proteins can be altered by many factors, including stress, surgery, liver or
kidney dysfunction, and pregnancy. In such circumstances, the free-drug con-
centration may increase if the antimicrobial drug has high protein binding.
Antibiotic penetration into tissues is related to the amount of antibiotic not
protein bound.
The ability of antibiotics to penetrate can be evaluated using the ratio of the
AUC for antibiotic in the peripheral site to that of serum. Penetration into fibrin
and lymph follows Ficks law of diffusion. The antibiotic present in the serum
penetrates into peripheral tissues after lag times [1, 3, 10]. Drug concentration in
tissues is generally lower than in serum; peaks may occur simultaneously or
shortly after the maximal levels are reached in serum, and for most antibiotics,
elimination from the extravascular component is slower than from serum.
The relationship between protein binding, tissue penetration, and excretory
mechanisms has been widely studied, especially for third-generation cephalospo-
rins. These derivatives seem to penetrate adequately into the cerebrospinal fluid
and appear to be appropriate agents for treating meningitis [11]. Generally, the
fluoroquinolones have a low degree of plasma protein binding, with excellent
tissue penetration and high concentrations, as reflected by their particularly large
apparent Vd [11].
6.7 Inactivation of Antibiotics
This mechanism of resistance is common for many antibiotics, including b-lactams

and aminoglycosides [12, 13]. Inactivation of b-lactam antibiotics requires specific
bacterial enzymes. Currently, more than 300 b-lactamase enzymes have been
described, and they can be divided in four different functional groups (14). Group
1 includes b-lactamases that confer resistance to all classes of b-lactams, except
carbapenems, and that are not inhibited by clavulanic acid. They can be either
chromosomal or plasmid encoded. Group 2 includes b-lactamases susceptible to
inhibition due to clavulanic acid, and they are divided in six subgroups. Subgroup
2a includes staphylococcal and enterococcal penicillinases. Subgroup 2b includes
broad spectrum b-lactamases of AGNB strains but also extended spectrum
b-lactamases. Subgroup 2c includes carbenicillin hydrolyzing b-lactamases.
Subgroup 2d includes cloxacillin hydrolyzing enzymes. Subgroup 2e includes
cephalosporinases. Subgroup 2f includes b-lactamases active on carbapenems.
Group 3 includes metallo-beta-lactamases with activity against carbapenems and
all b-lactam classes except monobactams. Group 4 includes the other b-lactamases
not included in the previous groups [12].
Extended spectrum b-lactamases (ESBL) are b-lactamase able to hydrolyze
the extended-spectrum cephalosporins, including third-generation compounds.
Antibiotics susceptible to ESBL include cefotaxime, ceftazidime, aztreonam,
and other expanded-spectrum cephalosporins, whereas imipenem is usually
resistant. These enzymes are encoded also in plasmids containing genes for
aminoglycoside resistance, trimethoprimsulfamethoxazole, and often fluoro-
quinolones. ESBL are often present in Escherichia coli or Klebsiella
pneumoniae but can be transferred to Proteus mirabilis, Citrobacter, Serratia,
and other AGNB [12]. Every E. coli and K. pneumoniae with reduced sus-
ceptibility to these drugs or to aztreonam should be considered at risk of
possessing these enzymes. Treating infections caused by ESBL-producing
strains is difficult due to the high risk of concomitant resistance to aminogly-
cosides, trimethoprimsulfamethoxazole, and fluoroquinolones. Carbapenems
are the agents of choice against ESBL-producing bacteria because they are
highly stable against b-lactamase hydrolysis [12].
Inactivation is by far the most important mechanism of acquired microbial
resistance toward aminoglycosides in clinical practice. Aminoglycosides may
become the substrate of several microbial enzymes that may phosphorylate,
adenylate, or acetylate specific hydroxyl or amino groups. Phosphorylation is the
main mechanism of inactivation for aminoglycosides, it is determined by either
Gram-positive or AGNB strains, and it results in complete inactivation of
aminoglycosides. Some aminoglycosides are resistant to phosphorylation due to
the presence of side chains, such as amikacin, or the absence of specific hydroxyl
groups, such as tobramycin and gentamicin. Aminoglycoside-inactivating enzymes
are often encoded in plasmids that are especially common in hospital environ-
ments. Pharmacological strategies to inhibit aminoglycoside-inactivating enzymes
are under evaluation.
6.8 Relationship Between Antibiotics and Endotoxins
Bacterial endotoxins are thought to play a fundamental role in the pathogenesis of

sepsis and septic shock. Bacterial growth and death is associated with the release
of variable amounts of endotoxins outside the bacteria. Antibiotic therapy
may increase the release of endotoxins that trigger the septic response by
binding to specific receptors upon mononuclear phagocytes, endothelial cells,
and polymorphonuclear leucocytes [1417]. This effect has been studied for
several agents, including b-lactams, glycopeptides, aminoglycosides, and
fluoroquinolones.
Regarding b-lactams, different patterns of endotoxin release seem to be cor-
related with the inhibition of different PBPs. In particular, inhibition of PBP1a and
1b results in a massive bactericidal activity against AGNB associated with
bacterial wall molecule degradation. These changes lead to production of
spheroplasts that undergo immediate bacteriolysis. In AGNB, PBP2 inhibition
results in spheroid elements osmotically stable but unable to proliferate, and thus
increases bacterial mass. Conversely, inhibition of PBP3 located in bacterial septa
prevents separation of the proliferating bacteria, resulting in long filaments that
are vital syncytia resistant to lysis but able to produce and release endotoxins.
Carbapenems, ceftriaxone, cefepime, and b-lactams in combination with
b-lactamase inhibitors inhibit PBP2 and 3 in AGNB. Therefore, they have a rapid
bactericidal activity without increments of bacterial mass and are associated with
poor endotoxin release. Ceftazidime, ticarcillin, and cefoxitin inhibit PBP3 and, at
high concentrations, PBP1; they have a slower bactericidal activity associated with
a slight increase in bacterial mass and a moderate release of endotoxins. Pipera-
cillin, monobactams, cefuroxime, and cefotaxime inhibit mainly PBP3, resulting in
slow bactericidal activity associated with an increase in bacterial mass and
endotoxin release [17].
Glycopeptides induce bacteriolysis and release endotoxins, but the lipid part
of the endotoxin is inactivated by these antibiotics. Fluoroquinolones result in a
slightly greater endotoxin release than do carbapenems, as they cause DNA
damage associated with the inhibition of cellular division that produces bac-
terial filaments able to release endotoxins. Besides the role of antibiotic therapy
in endotoxin release, they may have also endotoxin-neutralizing properties that
may inhibit the effects of endotoxin [18]. Antibiotics with endotoxin-neutral-
izing properties are polycationic molecules that can link the polyanionic moi-
eties of lipopolysaccharides, including polymyxins, teicoplanin, and
aminoglycosides [14, 15]. In experimental studies, polymyxin B resulted in
reduced endotoxin shock, pyrogenicity, Shwartzman reaction, and endotoxin
lethality [15]. Teicoplanin shows an endotoxin-neutralizing property that results
in reduced levels of tumor necrosis factor (TNF)-a and interleukin (IL)-1 after
endotoxin stimulation.
Antibiotics inhibiting protein synthesis result in a lower release of endotoxins
than the agents acting on the cell wall. Aminoglycosides inhibit endotoxin syn-
thesis, and although they favor endotoxin release, they neutralize them chemically
by binding their anionic moieties. In this regard, variable effects have been
observed for different compounds of this class of antibiotics: tobramycin seems to
be the most effective in neutralizing endotoxins.
6.9 Respect for Ecology: A Prerequisite

for a Safe Antibiotic Therapy
The relationship between the microbes and host is not just a fight between two
contenders [19]. Ecological studies show symbiosis in the microbial communities
of the skin and mucous membranes. In these ecosystems, the resident microor-
ganisms play an important physiological role with mutualistic relationships,
protection from infections included [20, 21]. The normal microflora acts as a
barrier against carriage of potentially pathogenic microorganisms and against
overgrowth of opportunistic microorganisms (colonization resistance) [21, 22].
For this protective effect to occur, great stability of these ecosystems is funda-
mental [22]. Administration of antimicrobial agents interferes with the ecological
balance between host and normal microflora.
The impact of antibiotics on normal intestinal flora has been widely studied.
The gastrointestinal tract is a complex ecosystem that may be altered by
antibiotic administration, resulting clinically in diarrhea and fungal infections
that usually cease after treatment. Pseudomembranous colitis is an example of
the deleterious effects of antimicrobial therapy that disturbs the normal ecologic
balance of the bowel flora, leading to an abnormal proliferation of Clostridium
difficile. The effects of antibiotics on oropharyngeal and skin microflora may be
important as well. Antibiotic therapy may cause alterations in skin microbial
flora with a decline in the number of bacteriamostly anaerobic bacteria [23].
Moreover, exposure to antibiotics often results in selection of antibiotic-
resistant organisms that can transfer resistance to other microorganisms. By
using antimicrobial agents that do not alter the level of resistance of colonizing
microorganisms, the risk of emergence and spread of resistant strains is reduced
[20, 21].
Respect for the individuals ecology should be a prerequisite for every antibi-
otic policy, and the use of antibiotics with little or no impact on the normal flora
should always be encouraged. Antibiotics that are not active against the indigenous
bacteria in the mouth and intestine and are not excreted to a significant degree via
the intestine, saliva, or skin are therefore preferred [20, 21]. First generation
b-lactams, aminoglycosides, and polymyxins are favorable from an ecological
point of view.
6.10 De-Escalation Therapy
In addition to treating infections, antibiotic use contributes to the emergence of

resistance among potentially pathogenic microorganisms. Therefore, avoiding
unnecessary antibiotic use and optimizing antimicrobial agent administration will
help improve patient outcomes while minimizing further pressures for resistance
[24]. Antibiotic administration guidelines/protocols developed locally or by
national societies suggest avoiding unnecessary administration [25, 26].
Severe infections, including pneumonia and bacteremia, remain a major

cause of morbidity and mortality among hospitalized patients. Clinical evidence
suggests that failure to quickly treat these infections with an adequate initial
antibiotic regimen is associated with greater patient morbidity and mortality.
Delays of 2448 h in initiating adequate treatment are associated with increased
mortality rates. Appropriate therapy should be instituted as soon as sepsis is
suspected in critically ill patients [27]. During recent decades, the types of
pathogens have significantly changed, with shifts from AGNB to Gram-positive
multi-drug-resistant organisms requiring appropriate empirical regimens.
Although early and appropriate therapy are the best predictors of positive out-
come, the microorganism causing the infection is frequently not known at the
time antimicrobial therapy is initiated [28]. This has led to the development of a
novel paradigm guiding empirical antimicrobial therapy administration for
patients with serious infections, called antibacterial de-escalation therapy. This is
an approach to antibacterial use that attempts to balance the need to provide
appropriate, initially broad antibacterial treatment while limiting the emergence
of antibacterial resistance. Resistance is minimized by narrowing the antibacte-
rial regimen once the pathogens and their susceptibility profiles are known and
by employing the shortest course of therapy [29]. De-escalation antimicrobial
therapy should be tailored to critically ill patients according to their clinical
status, illness severity, and suspicion of sepsis or nosocomial pneumonia [25].
Risk stratification should be employed to identify patients at high risk of
infection with antibiotic-resistant bacteria (prior treatment with antibiotics during
hospitalization; prolonged hospital stay; and presence of invasive devices, e.g.,
central venous catheters, endotracheal tubes, urinary catheters). Patients at high
risk for infection with resistant bacteria should be treated initially with an
appropriate combination of antibiotics. This approach must be modified if spe-
cific microorganisms are excluded based on examination of appropriate clinical
specimens. Delays in de-escalating unnecessary antibiotic treatment are harmful,
as they promote emergence of resistant bacterial pathogens. Decreasing the
overall duration of empirical and potentially unnecessary antibiotic use reduces
the incidence of hospital-acquired superinfections attributed to antibiotic-resistant
microorganisms [24].
6.11 Main Classes of Systemic Antibiotics
Systemic antibiotics include drugs commonly used in the ICU, such as b-lactams;
drugs to be used under specific indications, such as aminoglycosides, glycopeptides,
fluoroquinolones, and macrolides; and drugs used only for multi-drug-resistant
microorganisms, including streptogramins, linezolid, colistin, advanced-generation
cephalosporins, glycylcycline, and daptomycin (Table 6.2).
6
Table 6.2 Main systemic antibiotic used in ICU and their spectrum of antimicrobial activity
Class Drug S. spp. E. spp. SA MRSA HA NM AGNB PA Anaerobes

G+ G
Penicillin Penicillin G S S R R R S R R S R
Penicillinase-resistant Oxacillin S R S R R R R R R R
Penicillin Nafcillin S R S R R R R R R R
Systemic Antibiotics
Aminopenicillin Amoxicillin S S R R S/R S S/R R S R

Ampicillin S S R R S/R S S/R R S R
Carboxy-penicillin Ticarcillin S S R R S/R S S S S R
Carbenicillin S S R R S/R S S S S R
Ureidopenicillins Piperacillin S S R R S/R S S S S R
Mezlocillin S S R R S/R S S S S R
Cephalosporin Cefazolin S R S R S R R R R R
I 1st generation Cephalothin
Cephalosporin Cefoxitin S R S R S S S/R R S S
II 2nd generation Cefuroxime
Cephalosporin Ceftriaxone S R S R S S S S/R S R
III 3rd generation Ceftazidime S R S R S S S S S R
Cephalosporin Cefepime S R S R S S S S R R
IV 4th generation
Advanced-generation Ceftobiprole S S S S S S S S S R
Cephalosporin Ceftaroline S S S S S S S S S R
Carbapenem Imipenem S S/R S R S S S S S S
Monobactam Aztreonam R R R R S S S S R R
(continued)
79
80
b-lactam ? b lactam inhibitors Amoxicillin + S S S R S S S R S S

clavulanate
Piperacillin + S S S R S S S S S S
tazobactam
Aminoglycosides Tobramycin R R/S S RR S S S S R R
Amikacin R R/S S S S S S R R
Fluoroquinolone Ciprofloxacin R R S R S S S S R R
II 2nd generation Ofloxacin R R S R S S S S R R
Fluoroquinolone Levofloxacin S R/S S R S S S S S R
III 3rd generation Moxifloxacin S R/S S R S S S S S R
Macrolides I 1st generation Erythromycin S S S R R S R R R R
Macrolides Clarithromycin. S S S R S S S S S R
III 3rd generation Azithromycin S S S S S S S S R
Macrolides IV 4th generation Telithromycin S S S R S S S S S R
Glycopeptides Vancomycin S S S S R R R R S R
Teicoplanin S S S S R R R R S R
Polymyxin Colistin R R R R S S S S R S
Oxazolidones Linezolid S S S S R R R R S R
Streptogramins Dalfopristin S S S S R R R R S R
Quinupristin S S S S R R R R S R
Lipopeptides Daptomycin S S S S R R R R S R
Glycylcycline Tigecycline S S S S S S S R S S
S. spp. Streptococcus spp., E. spp. Enterococcus spp., SA Staphylococcus aureus, MRSA methicillin-resistant S aureus, HI Haemophilus influenzae,
NM Neisseria meningitidis, AGNB aerobic Gram-negative bacilli, PA Pseudomonas aeruginosa, S sensitive, R resistant
A. R. De Gaudio et al.
6.11.1 b-Lactams
The basic structure of these antibiotics is a b-lactam ring to which is attached a

side chain [17].
6.11.1.1 Mechanism of Action

These drugs act by inhibiting the synthesis of the bacterial cell wall due to
selective acylation of specific membrane proteins (the so-called PBPs). There
are several PBPs with different affinities for different b-lactam antibiotics.
PBP1a and 1b include the transpeptidases involved in peptidoglycan synthesis.
Other PBPs maintain the rodlike shape or are involved in septum formation at
division [17].
6.11.1.2 Mechanism of Resistance

intrinsic low affinity of PBPs;
development of high molecular weight PBPs with decreased affinity for
antibiotics;
enzymatic disruption by b-lactamases;
changes in permeability to the antibiotic or extrusion from the target site.
6.11.1.3 Side Effects

hypersensitivity reactions (maculopapular rash, urticarial rash, fever, bron-
chospasm, vasculitis, serum sickness, exfoliative dermatitis); cross-reactions
with different molecules of the same class do occur;
exfoliative dermatitis and exudative erythema multiforme of either the eryth-
ematopapular or vesiculobullous type constitute the characteristic Stevens
Johnson syndrome;
bone marrow depression with granulocytopenia;
hepatitis.
6.11.2 Penicillin G
6.11.2.1 Spectrum of Activity

Active against most anaerobic microorganisms (except Bacteroides fragilis):
Borrelia burgdorferi, Treponema pallidum, B. anthracis. Syphilis, actinomycosis,
and anthrax can be treated successfully with penicillin G. It may be still also useful
in meningococcal and clostridial infections [17].
6.11.2.2 Pharmacological Properties

The T1/2 of penicillin G is within 30 min, it is rapidly eliminated from the body,
mainly by the kidney (10% by glomerular filtration and 90% by tubular secretion);
a small amount is excreted in the bile.
6.11.3 Penicillinase-Resistant Penicillins (Nafcillin, Oxacillin,

Cloxacillin, Dicloxacillin)

These are resistant to staphylococcal penicillinase and remain useful despite the
increasing incidence of methicillin-resistant (MR) microorganisms, a term that
denotes resistance to all penicillinase-resistant penicillins and cephalosporins.
MRSA contains a high molecular weight PBP (PBP2) with a very low affinity for
b-lactams. Fifty percent of S. epidermidis strains are resistant to the penicillinase-
resistant penicillins by the same mechanism. The daily dose of oxacillin for adults
is 24 g divided into four doses [17].

All these congeners are bound to plasma albumin (approximately 90%); none is
removed from the circulation to a significant degree by hemodialysis. They are
rapidly excreted by the kidney, with half-lives between 30 and 60 min. Intervals
between doses of oxacillin, cloxacillin, and dicloxacillin do not have to be altered
for patients with renal failure.
6.11.4 Aminopenicillins (Ampicillin, Amoxicillin)

Active against Gram-positive and Gram-negative bacteria. Many pneumococcal
isolates are resistant to ampicillin. Haemophilus influenzae, Enterococcus spp.,
and the viridans group of streptococci usually are inhibited by ampicillin. Many
strains of Neisseria gonorrhoeae, E. coli, P. mirabilis, Salmonella, Shigella,
and Enterobacter are now resistant. Most strains of Pseudomonas, Klebsiella,
Serratia, Acinetobacter, and indole-positive Proteus also are resistant to this
group of penicillins. The combination with a b-lactamase inhibitor, such as
clavulanic acid or sulbactam, markedly expands the spectrum of activity of
these drugs. Aminopenicillins are useful in upper respiratory tract infections
(sinusitis, otitis media, acute exacerbations of chronic bronchitis, and epiglot-
titis) caused by Streptococcus pyogenes, S. pneumoniae, and H. influenzae [17].
6.11.4.2 Pharmacologic Properties

Ampicillin and amoxicillin are used at doses of 14 g per day every 6 h. Their
half-time is around 80 min. Dose adjustment is required in case of renal dys-
function. About 20% protein bound in plasma. Ampicillin is excreted in the bile,
amoxicillin in the urine.
6.11.5 Carboxypenicillins (Carbenicillin, Ticarcillin)

Active against some isolates of P. aeruginosa and Proteus that are resistant to
aminopenicillins. They are not active against S. aureus. Carbenicillin use may
cause excessive Na+ intake with congestive heart failure; it may also interfere with
platelet function. Ticarcillin is associated with fewer side effects [17].
6.11.6 Ureidopenicillins (Mezlocillin, Piperacillin)

These are active against P. aeruginosa and Klebsiella spp. [27]. They are neu-
tralized by b-lactamases. Ureidopenicillins are important agents for treating
patients with serious infections, such as bacteremia, pneumonia, and burns, caused
by AGNB.

The usual adult dosage is 618 g per day divided into 46 doses. These drugs are
excreted in bile to a significant degree [27].
6.11.7 Cephalosporins

These drugs inhibit bacterial cell-wall synthesis in a similar manner to that
provoked by b-lactams.
6.11.7.2 Spectrum of Action

Cephalosporins are not active against the following microorganisms: MRSA, MR
coagulase-negative staphylococci (MRCNS), Enterococcus spp., Listeria mono-
cytogenes, Legionella pneumophila, C. difficile, Stenotrophomonas maltophilia,
Campylobacter jejuni, and Acinetobacter spp. [28].
6.11.7.3 Mechanisms of Resistance

inability to reach the sites of action;
alterations in PBPs, especially PBP1a and 2x;
b-lactamasescephalosporins have variable susceptibility to b-lactamase;
cefazolin is more susceptible than cephalothin, cefoxitin, cefuroxime [28].

hypersensitivity reactions (maculopapular rash, anaphylaxis, bronchospasm,
urticaria); patients allergic to penicillins may manifest cross-reactivity to
cephalosporins;
nephrotoxicity, although less than the aminoglycosides or polymyxins;

diarrhea and alcohol intolerance;
serious bleeding related to either hypoprothrombinemia, thrombocytopenia,
and/or platelet dysfunction.
6.11.7.5 Classification
Classification of cephalosporins by generations is based on general features of
their antimicrobial activity [28]:
First-generation cephalosporinsinclude cephalothin, cefazolin, cephalexin,
cephradine, and cefadroxil. They have good activity against Gram-positive
bacteria with the exception of enterococci, MRSA, and MRCNS, and relatively
modest activity against Gram-negative microorganisms. Most oropharyngeal
anaerobes are sensitive, but the B. fragilis group is usually resistant. Cepha-
lothin is the most resistant to staphylococcal b-lactamase and is effective in
severe staphylococcal infections.
Second-generation cephalosporinsincluding cefamandole, cefoxitin, cefaclor,
loracarbef, cefuroxime, cefotetan, and cefprozil, have somewhat increased
activity against AGNB (Enterobacter, Proteus, and Klebsiella spp.), but are less
active than the third-generation agents. A subset of second-generation agents
including cefoxitin and cefotetan are active against B. fragilis.
Third-generation cephalosporinsinclude cefotaxime, ceftizoxime, ceftriaxone,
cefixime, cefpodoxime, cefoperazone, and ceftazidime. They are less active than
first-generation agents against Gram-positive cocci, but they are much more
active against the Enterobacteriaceae, including b-lactamase-producing strains.
A subset of third-generation agents (ceftazidime and cefoperazone) also is active
against P. aeruginosa. Ceftriaxone has a T1/2 of about 8 h, so it can be used once or
twice daily. More than half the drug can be recovered from the urine; the
remainder appears to be eliminated by biliary secretion. Ceftazidime has a better
activity than cefoperazone or piperacillin against Pseudomonas.
Fourth-generation cephalosporinssuch as cefepime, have an extended spec-
trum of activity compared with the third generation, with increased resistance to
b-lactamases. Fourth-generation agents may prove to have particular therapeutic
usefulness for treating infections due to AGNB resistant to third-generation
cephalosporins. Cefepime has greater activity than cefotaxime against Gram-
negative bacteria (H. influenzae, N. gonorrhoeae and N. meningitidis). Regarding
P. aeruginosa, cefepime is as active as ceftazidime and a higher activity against
streptococci and S. aureus. It is not active against MRSA, penicillin-resistant
pneumococci, enterococci, B. fragilis, or L. monocytogenes. Its serum T1/2 is 2 h
and it is renally excreted. Doses should be adjusted for renal failure. Cefepime has
excellent penetration into CSF. Adult dosage is 2 g intravenously every 12 h.
Advanced-generation cephalosporinssuch as ceftobiprole and ceftaroline has
been proposed for the treatment of complicated skin and skin structure infec-
tions and bacterial pneumonia [29]. Ceftobiprole has activity against aerobic
Gram-negative bacilli, which extends to cefepime-sensitive P. aeruginosa, and
activity against Gram-positive organisms, which includes methicillin-resistant
Staphylococcus aureus. Its most common side effects are nausea and dysgeusia.
Ceftaroline is a novel, broad-spectrum, with antimicrobial activity against
MRSA, Streptococcus pneumonia and respiratory Gram-negative pathogens
such as Moraxella catarrhalis and H. influenzae. It is eliminated primarily by
renal excretion, with a T1/2 of approximately 3 h [29].
6.11.8 Carbapenems (Imipenem, Meropenem,

Doripenem, Ertapenem)

Mechanism of action is similar to that of other b-lactams; very resistant to
hydrolysis by most b-lactamases [30].

These are active against aerobic and anaerobic microorganisms: streptococci,
enterococci (excluding Enterococcus faecium and non-b-lactamase-producing
penicillin-resistant strains), staphylococci, Listeria, Enterobacteriaceae, many
strains of Pseudomonas and Acinetobacter, and all anaerobes, including B. fragilis.
On the other hand, MRSA and methicillin-resistant MRCNS, as well as
S. maltophilia, are resistant to carbapenems [30]. Doripenem may offer slightly
more activity against selected pathogens, as it seems to resist attack by carbape-
nem-hydrolyzing b-lactamases such as imipenem-hydrolyzing b-lactamase Sme1.
Therefore, empiric therapy with doripenem may be useful when multidrug resis-
tance is suspected. Ertapenem is a unique member of this class; it lacks activity
against Pseudomonas and Enterococcus, but it has a favorable PK profile.

Imipenem is hydrolyzed rapidly by a dipeptidase of the proximal renal tubule,
which is inhibited by cilastatin. Imipenem has a T1/2 of about 1 h, and 70% is
recovered in the urine. Dosage should be modified for patients with renal insuf-
ficiency. Meropenem does not require coadministration with cilastatin because it is
not sensitive to renal dipeptidase.

nausea and vomiting are the most common;
seizures have been noted, especially with imipenem when high doses are given
to patients with central nervous system lesions and to those with renal
insufficiency.
6.11.9 Monobactams (Aztreonam)

These antibiotics interact with PBP to form long, filamentous bacterial structures.

Gram-positive bacteria and anaerobic organisms are naturally resistant. On the
other hand, activity against Enterobacteriaceae, P. aeruginosa, H. influenzae, and
gonococci is good. Monobactams are resistant to many b-lactamases.

The usual adult dose is 2 g every 68 h. The T1/2 is 1.7 h, and most of the drug is
recovered unaltered in the urine.
6.11.10 b-Lactams Combined with b-Lactamase Inhibitors
b-lactamase inhibitors bind to b-lactamases and inactivate them. They are mostly
active against plasmid-encoded b-lactamases, including the extended-spectrum
ESBL, but are inactive against type I chromosomal b-lactamases of AGNB.
b-lactamase inhibitors include clavulanic acid, sulbactam, and tazobactam [12].
Clavulanic acid has been combined with amoxicillin and ticarcillin with extension
of the antimicrobial activity to b-lactamase-producing strains of staphylococci,
H. influenzae, gonococci, and E. coli. Sulbactam is usually combined with ampi-
cillin; tazobactam has been combined with piperacillin. Although this combination
extends its spectrum of activity, it does not enhance that against P. aeruginosa, as
resistance is due to either chromosomal b-lactamases or decreased permeability of
piperacillin into the periplasmic space due to either porin protein OprD loss or
multidrug efflux system upregulation.
6.11.11 Aminoglycosides (Gentamicin, Tobramycin, Amikacin)
These drugs are used primarily to treat infections caused by AGNB [31].

Aminoglycosides interact with the 30S ribosomal subunit, with inhibition of
protein synthesis. They have a concentration-dependent effect and have a PAE.

The antibacterial activity of aminoglycosides includes AGNB. Kanamycin, as with
streptomycin, has a more limited spectrum compared with other aminoglycosides
and should not be used against Serratia spp. or P. aeruginosa. They have
little activity against anaerobic microorganisms, and their action against most
Gram-positive bacteria is limited. S. pneumoniae and S. pyogenes are resistant.
Both gentamicin and tobramycin are active against S. aureus and S. epidermidis.
However, staphylococci become rapidly gentamicin resistant during exposure to
the drug.
6.11.11.3 Clinical Use

Gentamicin, tobramycin, amikacin, and netilmicin can be used for treating
AGNB infections, including urinary tract infections, bacteremia, infected burns,
osteomyelitis, pneumonia, peritonitis, and otitis caused by P. aeruginosa,
Enterobacter, Klebsiella, Serratia, and other species resistant to less toxic
antibiotics. Penicillins and aminoglycosides must never be mixed in the same
flask because penicillin inactivates the aminoglycoside. An aminoglycoside in
combination with a b-lactam is indicated for empirical therapy of pneumonia
acquired in the ICU. Combination therapy is also recommended for treating
pneumonia caused by P. aeruginosa. Aminoglycosides are ineffective for treating
S. pneumoniae. They are very useful, however, in cases of enterococcal endo-
carditis. In granulocytopenic patients infected by P. aeruginosa, antipseudo-
monal penicillin administration in combination with aminoglycosides is
recommended [31].

mutations affecting proteins in the bacterial ribosome;
impaired transport of the drug into the bacteria;
acquisition of plasmids coding for aminoglycoside-metabolizing enzymes, such
as phosphoryl, adenyl, or acetyl transferases.

There is negligible binding (\5%) of aminoglycosides to plasma albumin. They
have poor penetration in most cells, in the central nervous system, and into
respiratory secretions. Aminoglycosides are excreted almost entirely by glomer-
ular filtration. Their T1/2 in plasma are similar and vary between 2 and 3 h in
patients with normal renal function. A linear relationship exists between plasma
creatinine concentration and T1/2 of aminoglycosides. Gentamicin and tobramycin
are used at the adult dosage of 35 mg/kg per day. Amikacin is used at 15 mg/kg.
Its spectrum of antimicrobial activity is the broadest of the group, and it is resistant
to aminoglycoside-inactivating enzymes. Routine aminoglycoside plasma con-
centration monitoring is strongly recommended to confirm that drug concentra-
tions are in the therapeutic range [31].

Side effects are vestibular, cochlear, and renal toxicity, and rarely, neuromuscular
blockade. Ototoxicity and nephrotoxicity are more frequent in patients with ele-
vated drug concentrations in plasma. Ototoxicity is largely irreversible, resulting
from progressive destruction of vestibular or cochlear sensory cells. Nephrotoxi-
cityusually reversibleis the result of marked accumulation of aminoglycosides
in the proximal tubular cells. Once-daily administration reduces nephrotoxicity
because it can saturate the tubular cells, with no reduction in efficacy [31].
6.11.12 Fluoroquinolones

Fluoroquinolones inhibit two enzymes involved in bacterial DNA synthesis: DNA
gyrase and topoisomerase IV [32, 33].
6.11.12.2 Generational Classification

First-generation drugs comprises nonfluorinated compounds, including nalidixic acid,
pipemidic acid, and cinoxacin; second-generation compounds include norfloxacin,
enoxacin, pefloxacin, ofloxacin, lomefloxacin, and ciprofloxacin; third-generation
compounds include levofloxacin, gatifloxacin, and moxifloxacin [32].

Second-generation quinolones are rapidly bactericidal on E. coli and various
species of Salmonella, Shigella, Enterobacter, Campylobacter, and Neisseria.
Ciprofloxacin and levofloxacin have good activity against non-multi-drug-resistant
strains of P. aeruginosa. Third-generation quinolones are highly active against
pneumococci and enterococci, though they have low activity against P. aeruginosa
strains. Several intracellular bacteria are inhibited by fluoroquinolones
(Chlamydia, Mycoplasma, Legionella, Brucella, and Mycobacterium). Most
anaerobic microorganisms are resistant to quinolones but sensitive to third-gen-
eration compounds [32].

Norfloxacin is useful for urinary tract infections only. The major limitation of
second-generation quinolones for treating community-acquired pneumonia and
bronchitis is their poor activity against S. pneumoniae. For these infections, third-
generation derivatives are highly effective. However, fluoroquinolones have in
vitro activity against other common respiratory pathogens, including H. influenzae,
M. catarrhalis, S. aureus, M. pneumoniae, C. pneumoniae, and L. pneumophila.
Mild to moderate respiratory exacerbations due to P. aeruginosa in patients with
cystic fibrosis have responded to orally administered fluoroquinolone therapy.
Third-generation quinolones are indicated for infections caused by L. pneumophila,
C. pneumoniae, and M. pneumoniae. MRSA is generally resistant to fluoroquin-
olones [33].

alterations in the drug target enzymes (DNA gyrase or topoisomerase mutations);
bacterial expression of membrane-associated efflux pumps [32].

Quinolones are widely distributed in body tissues. Serum T1/2 ranges from 3 to 5 h
for norfloxacin and ciprofloxacin, 78 h for gatifloxacin and levofloxacin,
and 1012 h for pefloxacin and moxifloxacin. Concentrations of quinolones in

urine, kidney, lung and prostate tissue; stool, bile, macrophages, and neutrophils
are higher than serum levels. Concentrations in CSF are lower than in serum.
Elimination routes differ: renal clearance predominates for ofloxacin, lomefloxa-
cin, levofloxacin, and gatifloxacin; pefloxacin and nalidixic acid are predominantly
eliminated nonrenally. Dose adjustments in patients with renal insufficiency are
generally required, except for pefloxacin and moxifloxacin. None of the agents is
efficiently removed by peritoneal or hemodialysis.

nausea, abdominal discomfort;
headache, dizziness;
rarely: hallucinations, delirium, and seizures, especially if associated with
non-steroidal anti-inflammatory drugs;
rashes, including photosensitivity;
arthralgias and joint swelling have developed in children.
6.11.13 Macrolides
6.11.13.1 Generational Classification

First-generation macrolide is erythromycin; second generation is represented by
spiramycin; third generation is clarithromycin and azithromycin, fourth generation
corresponds to ketolides [34, 35].

Macrolides interact with the peptidyl transferase center in the 50S ribosomal
subunit.

Erythromycin is most effective against aerobic Gram-positive cocci and bacilli.
Some staphylococci are sensitive to erythromycin, but resistant strains are
frequently encountered in hospitals. Many other Gram-positive bacilli, including
C. perfringens and Corynebacterium diphtheriae, are sensitive. Erythromycin is
not active against most AGNB. However, it retains activity against other Gram-
negative organisms, including N. meningitidis and N. gonorrhoeae, and is effective
against M. pneumoniae, L. pneumophila, and C. trachomatis. Clarithromycin is
more potent than erythromycin against streptococci and staphylococci. Azithro-
mycin generally is less active than erythromycin against Gram-positive organisms
and is more active than either erythromycin or clarithromycin against H. influenzae.
It is highly active against M. catarrhalis, Chlamydia spp., M. pneumoniae,
L. pneumophila, B. burgdorferi, Fusobacterium spp., and N. gonorrhoeae [24].
Telithromycin is the first ketolide antibiotic; it is used for treating community-
acquired respiratory tract infections and has high activity against many common
and atypical/intracellular respiratory pathogens, including macrolide-lincosamide-

streptogramin (MLSB)-resistant strains. It has more in vitro activity against
Gram-positive aerobes than the other macrolides. It is inactive against Entero-
bacteriaceae, non-fermentative Gram-negative bacilli, and A. baumannii [32].
Macrolides are not active against MRSA [35].

modifies the 50S ribosomal subunit;
produces inactivating enzymes;
decreases bacterial influence of the drug;
increases bacterial efflux of the drug.

Erythromycin protein binding is approximately 7080%. It is concentrated in the
liver and excreted in the bile, with a plasma elimination T1/2 of 1.6 h. Clarithro-
mycin distributes widely throughout the body and achieves high intracellular
concentrations; its protein binding ranges from 40 to 70%. It is metabolized in the
liver and is eliminated by renal and nonrenal mechanisms, with T1/2 of 37 h.
Clarithromycin is usually given at an adult dosage of 250 or 500 mg twice daily.
Azithromycin reaches high concentrations within cells, including phagocytes. Its
protein binding is low (51%), and the antibiotic undergoes some hepatic metab-
olism to inactive metabolites, even if biliary excretion is the major elimination
route, with T1/2 of 68 h that allows a once-daily regimen of 500 mg. Telithromycin
is metabolized hepatically and is eliminated primarily through the feces. It shows
good penetration into respiratory tissues, with concentration above mean MIC 90 s
for common respiratory pathogens in bronchial mucosa and epithelial lining fluid.
Macrolides have been reported to cause clinically significant drug interactions,
potentiating the effects of astemizole, carbamazepine, corticosteroids, cyclospor-
ine, digoxin, ergot alkaloids, terfenadine, theophylline, triazolam, valproate, and
warfarin. Telithromycin should be used cautiously in combination with simva-
statin, midazolam, theophylline, and digoxin [34, 35].

allergic skin eruptions;
cholestatic hepatitis;
epigastric distress with abdominal cramps, nausea, vomiting, and diarrhea;
cardiac arrhythmias, including QT prolongation with ventricular tachycardia.
6.11.14 Glycopeptides (Vancomycin and Teicoplanin)

Glycopeptides bind to the D-alanyl-D-alanine terminus of cell-wall precursor
units, inhibiting transglycosylase and transpeptidase enzymes that are necessary for
bacterial-wall synthesis. They are rapidly bactericidal for dividing bacteria [36, 37].

Gram-positive bacteria such as S. aureus and S. epidermidis, including methi-
cillin-resistant strains, are usually sensitive. Strains with MIC [1 mg/L, which
include glycopeptide intermediate S. aureus (GISA) are now spreading,
requiring other antimicrobial agents. Synergism between vancomycin and
gentamicin or tobramycin has been demonstrated in vitro against S. aureus,
including methicillin-resistant strains. S. pyogenes, S. pneumoniae, and viridans
streptococci are highly susceptible, as are most strains of Enterococcus spp.
Glycopeptides are not generally bactericidal for Enterococcus spp., and the
addition of a synergistic aminoglycoside might be necessary to produce a
bactericidal effect. Corynebacterium, Actinomyces, and Clostridium spp. are
sensitive [38, 39].

These drugs express an enzyme that modifies the cell-wall precursor so that it
no longer binds vancomycin (Enterococci). Three clinically important types of
resistance have been described for vancomycin. The Van-A-inducible pheno-
type confers resistance to both teicoplanin and vancomycin. The Van-B
phenotype tends to confer a lower level of resistance to vancomycin but not to
teicoplanin. The Van-C constitutive phenotype confers resistance to vancomycin
only [40].

Vancomycin has a serum elimination T1/2 of about 6 h, and approximately 55% is
bound to plasma protein. It appears in various body fluids, including CSF when the
meninges are inflamed, bile, and pleural, pericardial, synovial, and ascitic fluids. It
scarcely penetrates the lungs. More than 90% of an injected dose is excreted by
glomerular filtration. The dose for adults is 30 mg/kg per day. The best way to
divide this dose (every 6 or 12 h) and the best administration modality (inter-
mittent administration vs continuous infusion) have not been established. Dosage
adjustments must be made in cases of renal failure. Teicoplanin is a mixture of six
closely related compounds. It is highly bound by plasma proteins (9095%) and
has an extremely long serum elimination T1/2 (up to 100 h). Dosage in adults is
630 mg/kg per day. As with vancomycin, teicoplanin doses must be adjusted in
patients with renal insufficiency [39].

hypersensitivity reactions produced by vancomycin (macular skin rashes, chills,
fever);
nephrotoxicity and ototoxicity (less frequent with teicoplanin);
rapid intravenous infusion may cause erythematous or urticarial reactions,
flushing (red-neck or red-man syndrome), tachycardia, and hypotension [39].
6.11.15 Polymyxins (Polymyxin E or Colistin and Polymyxin B)

Polymyxins interact with phospholipids and disrupt their structure [41, 42].

There is reduced influx of the drug across the bacterial wall into the cell
membrane [39].

AGNB, including Enterobacter, E. coli, Klebsiella, Citrobacter, A. baumannii, and
P. aeruginosa. Proteus, Morganella, Serratia are intrinsically resistant.

Used to treat multi-drug-resistant nosocomial infections from AGNB. Aerosol
administration of the parenteral preparation has also been used as an adjuvant in
patients with severe Pseudomonas and A. pneumonia [43].

nephrotoxicity.
6.11.16 Oxazolidinones
Linezolid is the first member (and at present the only available molecule) of a new
class of antibiotics known as oxazolidinones [44, 45].

Oxazolidinones prevent formation of the fmet-total RNA:messenger RNA:30S
subunit ternary complex with ribosomal protein synthesis inhibition.

Linezolid is active against nosocomial infections involving Gram-positive
organisms, including MRSA, multi-drug-resistant strains of S. pneumoniae, and
vancomycin-resistant E. faecium (VRE) [44].

Linezolid-resistant VRE has been isolated, especially in immunosuppressed
patients [44].

thrombocytopenia and myelosuppression;
elevated liver and pancreatic enzymes;
pseudomembranous colitis;
diarrhea, nausea, and vomiting;

headache and dizziness;
monoamine oxidase inhibition.

Linezolid is widely distributed in the respiratory tract, with plasma protein binding
of 30%. About one-third of the dose is eliminated in the urine, with elimination
T1/2 of about 6.5 h. Dosage for most indications is 600 mg (iv or po) every 12 h.
There appears to be no need for renal dose adjustment. Linezolid has a monoamine
oxidase (MAO) inhibitor action. Therefore, precautions should be followed to
avoid potentially serious fooddrug or drugdrug interactions with other MAO
inhibitors [44].
6.11.17 Streptogramins (Streptogramin A-Dalfopristin

and Streptogramin B-Quinupristin)
These drugs are available in an injectable combination in a ratio of 70:30 [46, 47].

Binding to 23S RNA of the 50S ribosomal subunit, where they cause dissociation
of peptidyl-tRNA from the ribosome [46].

These antibiotics are active against Gram-positive bacteria, including multi-
drug-resistant strains of staphylococci, pneumococci, and enterococci. They are
inactive against E. faecalis but effective against VRE. They are as active as
vancomycin against MRSA [45].

enzymatic modification through virginiamycin acetyltransferases;
efflux pump;
alteration of the 50S ribosomal subunit.

reversible arthralgias, myalgias;
peripheral venous irritation.

These medications are metabolized by liver enzymes, including CYP450. The
postantibiotic effect and their PK characteristics allow dosing at 812-h intervals.
A potential for drug interactions exists because they inhibit the cytochrome
P450-3A4 enzyme system.
6.11.18 Lipopeptides (Daptomycin)

Action is via plasma membrane alteration without penetrating into the cytoplasm.
Inhibiting the production of protein has also been reported [48].

S. aureus, S. pyogenes, S. agalactiae, group C and G b-hemolytic streptococci,
E. faecalis, MRSA and VRE [48] are responsive to lipopeptides

Daptomycin is eliminated primarily by glomerular filtration, with a T1/2 of
0.91.4 h. The level of protein binding is 90%. Daptomycin demonstrates
concentration-dependent killing action and produces in vivo PAEs of 4.810.8 h.
Under conditions of high inoculum, daptomycin is highly effective. The intrave-
nous adult dose is 46 mg/kg once a day.

gastrointestinal disorders;
injection-site reactions;
fever;
dizziness;
rash;
people receiving daptomycin should be monitored for muscle pain or weakness,
and blood tests measuring creatine phosphokinase levels should be monitored.
6.11.19 Glycylcycline (Tigecycline)

Tigecycline inhibits protein translation by binding to the 30S ribosomal subunit
and blocking entry of amino-acyl tRNA molecules into the A site of the ribosome
[49, 50].

Gram-positive pathogens, including methicillin-resistant MRSA and S. epidermi-
dis (MRSE), VRE, Gram-negative pathogens, and anaerobic pathogens. Pseudo-
monas is usually resistant [51].

Tigecycline resistance in some bacteria (e.g., Acinetobacter strains) is associated
with multi-drug-resistant efflux pumps.

This drug is used to treat complicated skin and skin-structure infections, compli-
cated intra-abdominal infections, and bacterial pneumonia.

Tigecycline is not extensively metabolized, and is eliminated by biliary/fecal
excretion (60%) and in urine (33%). Its T1/2 is 27 h and protein binding of 80%.
Both AUC and time [ MIC correlates with efficacy, but the AUC/MIC ratio is
likely the primary PK/PD determinant of efficacy because tigecycline exhibits
time-dependent bacterial killing properties in combination with a moderate-
to-prolonged postantibiotic effect [51].

nausea, vomiting, diarrhea, abdominal pain, and headache are the most
common ([5%);
anaphylactic reactions;
hepatic dysfunction and failure (transaminases should be monitored);
discoloration of the teeth if used during tooth development;
photosensitivity;
antianabolic action, with increased blood urea nitrogen, acidosis, and
hyperphosphatemia.
References
1. Gunderson BW, Ross GH, Ibrahim KH, Rotschafer JC (2001) What do we really know about
antibiotic pharmacodynamics? Pharmacotherapy 21:302s318s
2. Mehrota R, De Gaudio R, Palazzo M (2004) Antibiotic pharmacokinetic and
pharmacodynamic considerations in critical illness. Int Care Med 30:21452156
3. Kollef MH (2001) Optimizing antibiotic therapy in the intensive care unit setting. Crit Care
5:189195
4. Ambrose PG, Bhavnani SM, Rubino CM et al (2007) Pharmacokinetics-pharmacodynamics
of antimicrobial therapy: its not just for mice anymore. Clin Infect Dis 44:7986
5. Pea F, Viale P (2009) Bench-to-bedside review: appropriate antibiotic therapy in severe
sepsis and septic shockdoes the dose matter? Crit Care 13:214
6. Kumar A (2009) Optimizing antimicrobial therapy in sepsis and septic shock. Crit Care Clin
25:733751
7. Roberts JA, Lipman J (2009) Pharmacokinetic issues for antibiotics in the critically ill
patient. Crit Care Med 37:840851
8. Figueiredo Costa S (2008) Impact of antimicrobial resistance on the treatment and outcome
of patients with sepsis. Shock 30:2329
9. Lipman J, Boots R (2009) A new paradigm for treating infections: go hard and go home.
Crit Care Resusc 11:276281
10. Lodise TP Jr, Lomaestro B, Drusano GL (2007) Piperacillin-tazobactam for Pseudomonas
aeruginosa infection: clinical implications of an extended-infusion dosing strategy. Clin
Infect Dis 44(3):357363
11. Barbour A, Scaglione F, Derendorf H (2010) Class-dependent relevance of tissue distribution
in the interpretation of anti-infective pharmacokinetic/pharmacodynamic indices. Int J
Antimicrob Agents 35:431438
12. Drawz SM, Bonomo RA (2010) Three decades of beta-lactamase inhibitors. Clin Microbiol
Rev 23:160201
13. Novelli A, Mini E, Mazzei T (2004) Pharmacological interactions between antibiotics and
other drugs in the treatment of lower respiratory tract infections. Eur Respir Mon 28:126
14. Nagaoka I, Hirota S, Niyonsaba F et al (2001) Cathelicidin family of antibacterial peptides
CAP18 and CAP11 inhibit the expression of TNF-alpha by blocking the binding of LPS to
CD14(+) cells. J Immunol 167:33293338
15. Zhang L, Dhillon P, Yan H et al (2000) Interactions of bacterial cationic peptide antibiotics
with outer and cytoplasmic membranes of Pseudomonas aeruginosa. Antimicrob Agents
Chemother 44:33173321
16. Tsuzuki H, Tani T, Ueyama H, Kodama M (2001) Lipopolysaccharide: neutralization by
polymyxin B shuts down the signaling pathway of nuclear factor kappa B in peripheral blood
mononuclear cells, even during activation. J Surg Res 100:127134
17. Surez C, Gudiol F (2009) Beta-lactam antibiotics. Enferm Infecc Microbiol Clin
27(2):116129 Article in Spanish
18. Augusto LA, Decottignies P, Synguelakis M et al (2003) Histones: a novel class of
lipopolysaccharide-binding molecules. Biochemistry 42:39293938
19. Sullivan A, Edlund C, Nord CE (2001) Effect of antimicrobial agents on the ecological
balance of human microflora. Lancet Infect Dis 1:101114
20. Klein G (2003) Taxonomy, ecology and antibiotic resistance of enterococci from food and
the gastro-intestinal tract. Int J Food Microbiol 88:123131
21. Tannock GW (2001) Molecular assessment of intestinal microflora. Am J Clin Nutr 73(Suppl
2):410S414S
22. Dunne C (2001) Adaptation of bacteria to the intestinal niche: probiotics and gut disorder.
Inflamm Bowel Dis 7:136145
23. Brook I (2000) The effects of amoxicillin therapy on skin flora in infants. Pediatr Dermatol
17:360363
24. Antonelli M, Mercurio G, Di Nunno S, Recchioni G, Deangelis G (2001) De-escalation
antimicrobial chemotherapy in critically ill patients: pros and cons. J Chemother 1:218223
25. Dogan O, Glmez D, Haselik G (2010) Effect of new breakpoints proposed by Clinical and
Laboratory Standards Institute in 2008 for evaluating penicillin resistance of Streptococcus
pneumoniae in a Turkish University Hospital. Microb Drug Resist 16:3941
26. Feldman C (2004) Clinical relevance of antimicrobial resistance in the management of
pneumococcal community-acquired pneumonia. J Lab Clin Med 143:269283
27. Goosen H (2003) Susceptibility of multi-drug-resistant Pseudomonas aeruginosa in intensive
care units: results from the European MYSTIC study group. Clin Microbiol Infect 9:980983
28. Kahlmeter G (2008) Breakpoints for intravenously used cephalosporins in
EnterobacteriaceaeEUCAST and CLSI breakpoints. Clin Microbiol Infect 14:169174
29. Bazan JA, Martin SI, Kaye KM (2009) Newer beta-lactam antibiotics: doripenem,
ceftobiprole, ceftaroline, and cefepime. Infect Dis Clin North Am 23:983996
30. Baughman RP (2009) The use of carbapenems in the treatment of serious infections.
J Intensive Care Med 24:230241
31. Smith CA, Baker EN (2002) Aminoglycoside antibiotic resistance by enzymatic deactivation.
Curr Drug Targets Infect Disord 2:143160
32. Viale P, Pea F (2003) What is the role of fluoroquinolones in intensive care? J Chemother
15(Suppl 3):510
33. Blondeau JM (2004) Fluoroquinolones: mechanism of action, classification, and development
of resistance. Surv Ophthalmol 49(Suppl 2):S73S78
34. Labro MT (2004) Macrolide antibiotics: current and future uses. Expert Opin Pharmacother
5:541550
35. Ackermann G, Rodloff AC (2003) Drugs of the 21st century: telithromycin (HMR 3647)
the first ketolide. J Antimicrob Chemother 51:497511
36. Malabarba A, Ciabatti R (2001) Glycopeptide derivatives. Curr Med Chem 8:17591773
37. Esposito S, Noviello S (2003) What is the role of glycopeptides in intensive care? J
Chemother 15(Suppl 3):1116
38. Parenti F, Schito GC, Courvalin P (2000) Teicoplanin chemistry and microbiology.
J Chemother 12(Suppl 5):514
39. Harding I, Sorgel F (2000) Comparative pharmacokinetics of teicoplanin and vancomycin.
J Chemother 12(Suppl 5):1520
40. Lundstrom TS, Sobel JD (2000) Antibiotics for Gram-positive bacterial infections.
Vancomycin, teicoplanin, quinupristin/dalfopristin, and linezolid. Infect Dis Clin North
Am 14:463474
41. Rocha JL, Kondo W, Baptista MI et al (2002) Uncommon vancomycin-induced side effects.
Braz J Infect Dis 6:196200
42. Beringer P (2001) The clinical use of colistin in patients with cystic fibrosis. Curr Opin Pulm
Med 7:434440
43. Tsubery H, Ofek I, Cohen S et al (2002) Modulation of the hydrophobic domain of
polymyxin B nonapeptide: effect on outer-membrane permeabilization and
lipopolysaccharide neutralization. Mol Pharmacol 62:10361042
44. Diekema DJ, Jones RN (2001) Oxazolidinone antibiotics. Lancet 358:19751982
45. Paradisi F, Corti G, Messeri D (2001) Antistaphylococcal (MSSA, MRSA, MSSE, MRSE)
antibiotics. Med Clin North Am 85:117
46. De Gaudio AR, Di Filippo A (2003) What is the role of streptogramins in intensive care? J
Chemother 15(Suppl 3):1721
47. Hershberger E, Donabedian S, Konstantinou K, Zervos MJ (2004) Quinupristin-dalfopristin
resistance in Gram-positive bacteria: mechanism of resistance and epidemiology. Clin Infect
Dis 38:9298
48. Devasahayam G, Scheld WM, Hoffman PS (2010) Newer antibacterial drugs for a new
century. Expert Opin Investig Drugs 19:215234
49. Zuckerman JM, Qamar F, Bono BR (2009) Macrolides, ketolides, and glycylcyclines:
azithromycin, clarithromycin, telithromycin, tigecycline. Infect Dis Clin North Am
23:9971026
50. Barton E, MacGowan A (2009) Future treatment options for Gram-positive infections
looking ahead. Clin Microbiol Infect 15:1725
51. Bouza E (2009) New therapeutic choices for infections caused by methicillin-resistant
Staphylococcus aureus. Clin Microbiol Infect 15:4452
Systemic Antifungals
7
7.1 Introduction
Since approximately 2005, there have been important developments in antifungal

therapy relevant to managing invasive fungal infections (IFIs) in intensive care
unit (ICU) patients. We now have a new triazoleposaconazoleand two new
echinocandinsanidulafungin and micafunginto add to the previous repertoire
of systemic antifungals for treating IFIs. There has been much debate on how best
to use systemic antifungals in patients at high risk of candidaemia/invasive can-
didiasis, with studies and reviews addressing the issues of antifungal prophylaxis,
empirical and pre-emptive therapy, and treating documented infection. Unfortu-
nately, the one area that has not progressed significantly is diagnosis. There are
developments with nonculture diagnostic methods, such as detecting circulating
beta-D-glucan, and polymerase chain reaction (PCR) for detecting candidiasis
and some other mycoses, but none has yet become firmly incorporated into
established care pathways. Surveillance studies of candidaemia show that Candida
albicans continues to be the most common cause, followed by C. glabrata and
C. parapsilosis. A worrying development is resistance to echinocandins, especially
in C. glabrata, but the impact of this in clinical practice is not clear. Invasive
aspergillosis (IA) has been reported in ICU patients outside of the traditional
immunocompromised risk groups. The incidence appears quite variable between
units, but it is not so common that it needs to be factored in when considering
empirical therapy unless the ICU involved has experienced an increase in cases or
the patient is in a recognised risk group. The older agents have been addressed in
previous editions of this book [1]; we therefore concentrate on the newer
antifungals in this edition.
C. J. Collins (&)
Clinical Microbiology, Trinity College Dublin, Dublin,
Leinster, Republic of Ireland
100 C. J. Collins and Th. R. Rogers
7.2 Antifungal Drugs
7.2.1 Posaconazole
Posaconazole (Noxafil) is a structural analogue of itraconazole. It is licensed in

Europe for first-line treatment of oropharyngeal candidiasis in patients who have
severe disease or are immunocompromised. Unlike under the US Food and Drug
Administration (FDA) license, it may be used for those with refractory IFIs or with
IFIs and who are intolerant of first-line antifungals. It is also approved for IFI
prophylaxis in severely immunocompromised patients, e.g., haematopoietic stem-
cell transplantation (HSCT) recipients with graft versus host disease (GVHD) who
are at high risk for IFIs and in patients with acute myelogenous leukaemia (AML)
or myelodysplastic syndrome (MDS) who are at high risk for IFIs.

Posaconazole is active against all organisms covered by the other triazoles,
including fluconazole-resistant Candida spp., but it is also active against members
of the Zygomycetes [2]. It is fungistatic against many of the common Candida
spp., including C. albicans, but is generally fungicidal against Aspergillus spp. [2].

Posaconazole is available only as an oral suspension. The bioavailability is vari-
able but is considerably increased when taken with fatty meals. The dosing varies
according to indication: 200 mg three times daily for prophylaxis against IFIs;
100 mg twice daily for treating oropharyngeal candidiasis; 400 mg twice daily for
treating oropharyngeal candidiasis refractory to other azoles. There are no dose
adjustments required for renal or liver impairment. It is highly protein bound
([/=99%). Although it is not metabolised by CYP enzymes, it has been found to
significantly decrease the activity of the CYP3A4 isoenzyme, resulting in inter-
actions with drugs that are metabolised via this pathway. Also, coadministration
with phenytoin, rifabutin (increased clearance of both) or cimetidine (may
decrease absorption) is not recommended unless the benefits outweigh the risks.
It is eliminated primarily unchanged and has no microbiologically active metab-
olites. Posaconazole is generally well tolerated, with gastrointestinal symptoms
such as nausea, abdominal pain and diarrhoea being the most common adverse
effects.
7.2.1.3 Clinical Experience

Posaconazole is mainly used as a prophylactic agent against IFI in severely
immunocompromised patients. Based on two large antifungal prophylaxis studies
[3, 4] it has received an A-I grade recommendation from the Infectious Diseases
Society of America (IDSA) for prophylaxis in HSCT recipients with GVHD who
are at high risk for IA and in patients with AML or MDS who are at high risk for
IA [5]. In a study of salvage therapy for IA in which controls were historical and
7 Systemic Antifungals 101
had mostly received amphotericin B, the response rates overall were superior for
posaconazole, irrespective of the Aspergillus spp. [6]. In two studies and a case
report including a review of the literature, posaconazole as salvage therapy for
zygomycosis was evaluated with encouraging results [79]. However, there are no
data yet supporting its use as first-line therapy for IA, and this may have to await
the development of an intravenously administered formulation. Posaconazole has
also been investigated as treatment for other mould infections in immunocom-
promised patients, with similarly encouraging results [2, 10].
So far, clinical experience of using posaconazole in ICU patients appears to be
restricted to small case report series [11]. Due to the lack of an intravenously
administered formulation, therapeutic drug monitoring may be required in patients
with invasive fungal disease. Studies to date are limited on its value, but one small
series in critical care patients has shown that serum concentrations [0.5 mg/1 are
more likely to achieve therapeutic success [12].
7.2.2 Echinocandins and Pneumocandins
Caspofungin, anidulafungin and micafungin are now approved for treating inva-
sive Candida infections. In the 2009 IDSA guidelines for managing candidiasis,
echinocandins along with fluconazole received a grade A-I recommendation for
treating candidaemia in non-neutropenic patients, with the expert panel favouring
echinocandins in those with moderate to severe illness or with recent azole
exposure [13]. They also received a grade A-II recommendation for the initial
treatment of candidaemia in neutropenic patients and a grade B-III recommen-
dation for the treatment of C. glabrata. Although head-to-head studies among
the three approved echinocandins are limited [14], overall, there have been no
significant differences found in their clinical efficacy [15], although there are
pharmacokinetic and pharmacodynamic differences [16].
7.2.3 Caspofungin
Caspofungin (Cancidas) was reviewed in the last edition of this series [1]. It was
the first echinocandin to be approved for clinical use in Europe and is the first and
only member of this group to be licensed by the European Medicines Agency
(EMEA) and the FDA for treating patients with invasive Aspergillus infections
refractory to other therapy. It has also been licensed as empirical therapy for
presumed fungal infections in febrile, neutropenic patients and for treating inva-
sive candidiasis. It is also the only echinocandin with both an FDA and EMEA
license for therapeutic use in the paediatric population.
Standard dosing is a 70-mg loading dose followed by 50 mg daily. Dosing in
the paediatric population is based on the patients body surface area (BSA).
High-dose therapy with caspofungin was recently evaluated primarily for safety.
Betts et al. [17] undertook a double-blind comparison of caspofungin 150 mg daily

with standard dosing in patients diagnosed with invasive candidiasis at a sterile
site. This higher dose was well tolerated. There were insufficient patients in the
study to draw any conclusion on the possibility of superior efficacy of the higher
dose regimen. Liver function tests should be monitored during caspofungin
treatment, especially when it is prescribed concomitantly with cyclosporine.
7.2.4 Micafungin
Micafungin (Mycamine) was synthesised by chemical modification of an envi-

ronmental mould Coleophoma empedri. It is approved by both the EMEA and
FDA for treating invasive and oesophageal candidiasis. It is also approved by both
for the prophylaxis of Candida infections in patients undergoing HSCT. It has not
yet been approved for use in Aspergillus infections. Only the EMEA has licensed
its use in the paediatric population.

Micafungin appears to be similar to caspofungin in terms of spectrum of activity.

Micafungin, like the other members of this class, is available only as an intrave-
nously administered formulation. Dosing varies according to indication: 100 mg
once daily for treating invasive Candida infections; 150 mg once daily for treating
oesophageal candidiasis; 50 mg once daily for prophylaxis against Candida
infections. A loading dose is not required, and no dosage adjustments dependent
on renal or liver indices are required, including for patients on dialysis. However,
its use in severe hepatic impairment is not recommended, as no data are available.
Micafungin is generally well tolerated, with the most common side effects being
fever, headache, gastrointestinal symptoms and electrolyte and haematological
disturbances. The liver function tests should be monitored during treatment.
It is highly protein bound. Monitoring for toxicity due to increased serum
concentrations of sirolimus, nifedipine and itraconazole should be performed when
these are co-prescribed with micafungin. Preclinical data in rats raised some
concerns regarding micafungin and the development of hepatic neoplasms,
although the relevance of this to human use is not clear. This has prompted the
EMEA to refer to this potential risk in its product information sheet and advises
that micafungin should only be used if other antifungals are not appropriate.

In two separate double-blind, randomised non-inferiority studies, activity of
micafungin (100 mg/day) versus liposomal amphotericin B (3 mg/kg/day) and
micafungin (100 or 150 mg/day) versus caspofungin (70 mg on day 1 followed by
50 mg/day) was compared for treating candidaemia and invasive candidiasis in
predominantly non-neutropenic patients [18, 19]. Micafungin, at the studied doses,

was found to be as effective as the two other agents but was associated with fewer
adverse events when compared with liposomal amphotericin B. In a post-hoc
analysis of data from the study comparing micafungin with liposomal amphoter-
icin B, the therapeutic efficacy of micafungin was lower in ICU patients when
compared with non-ICU patients. However, multivariate regression analysis
suggested that this was related only to Acute Physiology and Chronic Health
Evaluation II (APACHE II) score, and mortality at either days 8 or 30 was lower in
patients with low APACHE II scores [20].
Micafungin has been compared in a randomised, double-blind trial with
liposomal amphotericin B for treating candidaemia and other forms of invasive
candidiasis in a young paediatric population. Rates of treatment success were similar
with a lower overall rate of adverse events associated with micafungin [21]. Clinical
trials have shown that, at the appropriate dosage (i.e., 100 and 150 mg/day), mica-
fungin efficacy is comparable with that of fluconazole (200 mg/day) for treating
endoscopically and culture-proven oesophageal candidiasis, with no significant
differences in safety and tolerability [22, 23]. In a randomised controlled study of
antifungal prophylaxis in an adult and paediatric population following HSCT,
patients were randomised to receive either micafungin (50 mg or 1 mg/kg/day) or
fluconazole (400 mg or 8 mg/kg/day) [24]. Where success was defined as the
absence of IFI through the end of therapy and at the end of the 4-week period after
treatment, overall micafungin efficacy was superior to that of fluconazole (80.0%
in the micafungin arm vs. 73.5% in the fluconazole arm, a statistically significant
finding). So far, there are relatively limited clinical data on micafungin therapy for
aspergillosis. In a prospective, open-label, noncomparative study, micafungin was
evaluated in patients with probable or proven IA and found to be efficacious and
safe [25]. However, the numbers treated with micafungin only were small.
7.2.5 Anidulafungin
Anidulafungin (Eraxis, Ecalta) is also a semisynthetic cyclic lipopeptide syn-

thesised from a fermentation product of A. nidulans. It is approved by the EMEA
for treating invasive candidiasis in non-neutropenic adult patients. It has not yet
been approved for use in Aspergillus infections, nor has it received a license for
use in the paediatric population either in the USA or Europe. It has been approved
by the FDA for treating candidaemia and other forms of Candida infections
(intra-abdominal abscess and peritonitis) and oesophageal candidiasis.

Anidulafungin appears to be similar to caspofungin and micafungin in terms of
spectrum of activity.

This has recently been reviewed [26]. Anidulafungin, as with the other members of
this class, is available only for intravenous administration. For treating candida-
emia, intra-abdominal or peritoneal candidiasis, a loading dose of 200 mg on day 1
is required, followed by a maintenance dose of 100 mg once daily thereafter.
For treating oesophageal candidiasis, loading and maintenance doses are 100 and
50 mg once daily, respectively. It is not metabolised in the body but is slowly
chemically degraded and excreted into the intestine in an inactive form. No dose
adjustment is required in either hepatic or renal failure, including for patients on
dialysis. There are no serious drug interactions to date. Overall, it has a good safety
profile, but liver function should be monitored during treatment.

Anidulafungin (200 mg on day 1, then 100 mg/day) was compared to intrave-
nously administered fluconazole (400 mg/day) in a double-blind, randomised trial
of invasive candidiasis in mostly non-neutropenic patients [27]. In the modified
intention-to-treat group analysis, the global response rate (clinical and microbio-
logical) at the end of intravenous therapy significantly favoured anidulafungin
(however, the study had a number of limitations). The frequency and types of
adverse events were similar in the two groups. Anidulafungin (100 mg on day 1,
50 mg once daily maintenance) was compared with fluconazole (200 mg/day) in a
randomised double-blind trial in patients with oesophageal candidiasis [28]. In an
intent-to-treat analysis, the rate of endoscopic success for anidulafungin was found
not to be inferior to that for fluconazole. The safety profiles were similar. How-
ever, recurrence rates at the 2-week follow-up visit were higher for anidulafungin.
7.2.6 Combination Antifungal Therapy
Until recently, the mechanism of action of antifungal drugs was based mostly on
activity on the fungal cell membrane, such as for amphotericin B and triazoles.
Therefore, the benefit of a combination of antifungals that both acted on the cell
membrane was debatable. However, the echinocandins act at a different site, i.e.,
the fungal cell wall, so arguments for combination therapy, such as an echino-
candin and a polyene or an echinocandin and an azole, make more theoretical
sense. Unfortunately, high-quality clinical data on combination antifungal therapy
is limited. Recommendations for their use are often based on in vitro data, animal
studies and limited clinical experience. The IDSA published guidelines on treating
various fungal infections and recommended combination therapy in some
instances [5, 13, 2933]. For Candida spp. infections, combination therapy is
recommended for treating some forms of invasive infection but not for
candidaemia. A combination of amphotericin B and flucytosine is recommended
by the IDSA as first-line therapy for central nervous system infections, endocar-
ditis, endophthalmitis, urinary fungal balls, infected pacemakers, intracardiac
devices and ventricular assist devices and as second-line therapy for pyelonephritis
[13]. Combinations such as amphotericin B and flucytosine, amphotericin B and
fluconazole, and fluconazole and flucytosine may be used for treating cryptococcal
infection [30].
For initial primary IA therapy, combination therapy is not routinely recom-
mended. However, addition of a second agent may be considered as part of salvage
therapy when a primary agent is failing [5]. For endemic mycoses, combination
therapy is again not routinely recommended. However, combination therapy with
amphotericin B and an azole has been used to treat coccidioidomycosis, especially
when infection is widespread or there is single-agent failure [32].
Consensus guidelines for treating rare mould infections, such as zygomycosis,
are not available and are much needed. Combination therapy has been used to treat
some of these infections, but there are very limited clinical data to support definite
strategies of management.
7.3 Antifungal Susceptibility Testing
Data on antifungal susceptibility is important for patient management andas

there is only a limited number of classes of antifungal agents availablefor
surveillance of drug resistance. Epidemiological data and guidelines, such as those
provided by the IDSA for treating invasive candidiasis, may be used to guide
initial therapy. However, antifungal susceptibility testing provides additional data
on local epidemiology and may provide valuable information for decisions
regarding ongoing therapy in terms of the safest, most efficacious and cost-
effective agent. There have been major developments in antifungal susceptibility
testing in recent years, with several automated or semiautomated commercial
systems and methods based on the disk-diffusion technique now available, pro-
viding simple, flexible and affordable procedures in the clinical laboratory [34].
Some of these commercial methods (e.g., E test) correlate well with reference
procedures.
Minimum inhibitory concentration (MIC) breakpoints have traditionally divi-
ded an organism into three categories of susceptibility to a drug (susceptible,
intermediate and resistant), with these breakpoints determined using epidemio-
logical, in vitro, pharmacokinetic/pharmacodynamic and clinical data. For some
fungi, the Clinical Laboratories Standard Institute (CLSI) has replaced the inter-
mediate category with a category labelled susceptibledose dependent (S-DD),
whereby an antifungal would be expected to work with maximisation of dose and
bioavailability. Committees in Europe [e.g., The European Committee on Anti-
microbial Susceptibility Testing (EUCAST)] and the US (CLSI) are responsible
for developing reproducible methods of antimicrobial susceptibility testing and for
setting interpretative breakpoints, with efforts being made to align these for a
consistent approach. These are continuously reviewed and are subject to change.
Unlike breakpoints for antibacterial antibiotics, where much more clinical and in
vitro data are available to set these breakpoints, there is still much work to be done
in setting interpretative breakpoints for all antifungal agents against all clinically
relevant fungi. For example, whereas there are standard guidelines with inter-
pretative breakpoints for several antifungal agents versus Candida spp. [35], there
are still no interpretative criteria for any antifungal agent versus Cryptococcus
neoformans. However, based on current in vitro and clinical data, the activities of
the antifungal agents currently in use are known for many of the common IFIs
(Table 7.1).
7.4 Impact of Changing Epidemiology of Invasive Mycoses

in Intensive Care
Since the last edition of this book [1], there have been further epidemiological
studies to determine what the predominant fungal infections are, and in particular,
the frequency of candidaemia/candidiasis is in intensive care practice [36].
Differences in the reported incidence of candidaemia between European countries
have been noted, although there is no clear explanation for this [37]. C. albicans
continues to account for about half of all documented cases, although improved
microbiological investigation has meant that other Candida spp. are being more
reliably identified. In a recent Italian study [38], the incidence of candidaemia and
distribution of the causative species were recorded for 19992007 in an 18-bed
mixed medical and surgical ICU. The incidence rate was 1.42 episodes/10,000
patient days/year. Overall, C. albicans accounted for 46% of isolates and other
Candida spp. 54%, mainly C. parapsilosis (22%) and C. glabrata (13%).
Data from the 20082009 SENTRY Antimicrobial Surveillance Program on
nosocomial Candida bloodstream infections showed that C. glabrata was the
second most commonly isolated Candida spp. after C. albicans and was associated
with the highest frequency of resistance to both fluconazole and other triazoles
(range 5.17.7%) and the echinocandins (range 3.25.1%) [39].
7.5 Relationship Between Treating Invasive Candidiasis/

Candidaemia and Outcome
Several recent studies assessed the relationship between patient survival, time of
starting antifungal therapy and its appropriateness following candidaemia docu-
mentation. In a Canadian study, administering adequate empirical antifungal
therapydefined as giving an adequate dose of antifungal drug to which the
identified pathogen was shown to be susceptible in vitrowas associated with a
significant decrease in mortality rate [40]. When assessing the appropriateness
of empirical antibiotic therapy in septic patients admitted to the ICU,
Garnacho-Montero et al. [41] found that fungal infection was an independent
variable related to inappropriate therapy, which in turn was a predictor of
7
Table 7.1 Comparison of the major antifungal agents used to treat fungal infections in the intensive care unit
Class Polyenes
Generic name Amphotericin B Amphotericn B colloidal dispersion Liposomal amphotericin B Amphotericin B lipid complex
deoxycholate
(conventional)
Trade name(s) Fungizone Amphocil (Eur) AmBisome Abelcet
Fungilin Amphotec (US)
Main spectrum of activity (relative clinical and in vitro activity)

Candida spp.
C. albicans ++
C. glabrata ++
C. tropicalis ++
C. parapsilosis ++
C. krusei +
Cryptococcus neoformans ++
Trichosporon spp. +
Aspergillus spp.
A. fumigatus ++
A. flavus ++
A. terreus
A. niger ++
A. nidulans
Zygomycetes
Fusarium spp. +
S. prolificans
S. apiospermum
107
(continued)
108
Class Polyenes
deoxycholate
(conventional)
Indication Fungizone: Amphocil: AmBisome: Abelcet:
(Please refer to specific Systemic fungal Severe systemic or deep mycoses 1. Severe systemic or deep 1. Severe invasive candidiasis
licensed indications and full infections where toxicity and renal failure mycoses where toxicity 2. Severe systemic fungal infections
prescribing information Fungilin: preclude use of conventional (especially nephrotoxicity) in patients not responding to
according to jurisdiction; Oral and perioral amphotericin precludes use of conventional amphotericin or to
adults only unless specified) fungal infections Amphotec conventional amphotericin other antifungal drugs or where
Invasive aspergillosis in patients not 2. Suspected or proven toxicity or renal impairment
responding to conventional infection in febrile precludes conventional
amphotericin or where toxicity or neutropenic patients amphotericin, including invasive
renal impairment precludes unresponsive to broad- aspergillosis, cryptococcal
conventional amphotericin spectrum antibacterials meningitis and disseminated
cryptococcosis in HIV patients
Routes of administration Fungizone: IV, neb IV, neb IV, neb
(labelled and unlabelled IV, neb, IT,
routes) bladder
irrigation
Fungilin: topical
PO dose NA NA NA NA
(Adult only)
(continued)
7
Class Polyenes
deoxycholate
(conventional)
IV dose 0.71 mg/kg OD 37.5 mg/kg OD 35 mg/kg OD 5 mg/kg OD

(Adult only)
Dose adjustment if renal No
impairment
Dose adjustment if liver No
impairment
TDM recommended No
Main side effects Infusional toxicity and nephrotoxiciy more common with the conventional form. Hypokalaemia disturbance headache,
fever, rigors, hypotension, tachypnoea
Drug interactions Renal toxicity may indirectly affect the action of othr drugs (continued)
109
110
Class Azoles
Generic name Fluconazole Itraconazole Voriconazole Posaconazole
Trade name(s) Diflucan Trican Sporanox Vfend Noxafil
Candida spp.
C. albicans ++ ++ ++ ++
C. glabrata + + +
C. tropicalis ++ ++ ++ ++
C. parapsilosis ++ ++ ++ ++
C. krusei + + +
Cryptococcus neoformans + ++ ++
Trichosporon spp. ++ N N N
Aspergillus spp.
A. fumigatus ++ ++ ++
A. flavus ++ ++ ++
A. terreus ++ ++ ++
A. niger ++ ++ ++
A. nidulans ++ ++ ++
Zygomycetes +
Fusarium spp. +
S. prolificans N N N
S. apiospermum + N
(continued)
7
Class Azoles

Indication Generic: Generic: Generic: Noxafil (EMEA)
(Please refer to specific 1. Vaginal candidiasis 1. Oropharyngeal candidiasis 1. Invasive aspergillosis 1. Prophylaxis of invasive fungal
licensed indications and full and candidal balanitis 2. Vulvovaginal candidiasis 2. Serious infections caused infections in:
prescribing information 2. Mucosal candidiasis 3. Fungal skin infections by S. apiospermum, - those receiving remission-
according to jurisdiction; 3. Fungal skin 4. Onychomycosis Fusarium spp., or invasive induction chemotherapy AML
adults only unless specified) infections 5. Histoplasmosis fluconazole-resistant or MDS expected to result in
4. Invasive candidal 6. Systemic aspergillosis, Candida spp. (including C. prolonged neutropenia and who
infections including candidiasis and cryptococcosis krusei) are at high risk of developing
candidaemia and including cryptococcal invasive fungal infections
disseminated meningitis where other antifungal - HSCT recipients who are
candidiasis drugs are inappropriate or undergoing high-dose
5. Cryptococcal ineffective immunosuppressive therapy for
infections including 7. Maintenance in AIDS patients GVHD and who are at high risk of
meningitis to prevent relapse of underlying developing invasive fungal
6. Prevention of fungal infection infections
relapse of 8. Prophylaxis in haematological 2. Severe oropharyngeal
cryptococcal malignancy or if undergoing candidiasis or if immuno-
meningitis in AIDS BMT compromised
patients 3. Salvage therapy for invasive
7. Prevention of fungal aspergillosis, fusariosis,
infections in immuno- chromoblastomycosis, mycetoma
compromised patients and coccidioidomycosis in patients
refractory to or intolerant of first-
line therapy with other antifungal
agents
FDA: not licensed for indication
number 3
111
(continued)
112
Class Azoles

Routes of administration PO, IV PO, IV, topical mouth PO, IV PO
(labelled and unlabelled
routes)
PO dosea 1. 150 mg 1. 100200 mg OD [40kg: 1. 200 mg TDS
(Adult only) single dose 2. 200 mg BD for 1 day 400 mg BD/day, then 200300 2. 200 mg/day, then 100 mg OD
2. 50100 mg 3. 100 mg OD to 200 mg BD mg BD 3. 400 mg BD with food; 200 mg QDS
OD 4. 200 mg OD \40kg without food
3. 50 mg OD 5. 200 mg OD/BD 200 mg BD/day, then 100150
4. 400800 mg 6. 100200 mg OD/BD mg BD
OD 7. 200 mg OD/BD
5. 400800 mg 8. 5 mg/kg daily in 2 divided doses
OD
6. 200 mg OD
7. 50400 mg
OD
IV dosea 4. 400800 mg 5. and 6. 200 mg BD for 2 days, then 6 mg/kg BD for 24hrs, then 34 NA
(Adult only) OD 200 mg OD mg/kg BD
5. 400800 mg
OD
6. 200 mg OD
7. 50400 mg
OD
Dose adjustment if renal Yes Yes Oral formulation if possible No
impairment
(continued)
7
Class Azoles

Dose adjustment if No No Yes No
liver impairment
TDM recommended No Yesb Yesb Yesb

Main side effects GI disturbance, GI disturbance especially Hallucinations, Transient visual disturbance, GI disturbance, abnormal LFTs,
abnormal LFTs nausea and diarrhoea raised creatinine, photosensitivity headache, fever, rigors
Drug interactions Few Numerous Numerous Some
Inhibits CYP3A4 Inhibits CYP3A4 Inhibits CYP3A4
Also interacts with CYP 2C19
(continued)
113
114
Class Echinocandins
Generic name Caspofungin Micafungin Anidulafungin
Trade name(s) Cancidas Mycamine Ecalta (Eur)

Eraxis (US)

Candida spp.
C. albicans ++ ++ ++
C. glabrata ++ ++ ++
C. tropicalis ++ ++ ++
C. parapsilosis + + +
C. krusei ++ ++ ++
Cryptococcus neoformans
Trichosporon spp.
Aspergillus spp.
A. fumigatus + + +
A. flavus + + +
A. terreus + + +
A. niger + + +
A. nidulans + + +
Zygomycetes
Fusarium spp
S. prolificans
S. apiospermum
(continued)
7
Class Echinocandins

Eraxis (US)
Indication Cancidas: (EMEA) Mycamine: (EMEA) Ecalta: (EMEA)

(Please refer to specific licensed 1. Invasive candidiasis in adult or Adults [16 years: Treatment of invasive
indications and full prescribing paediatric patients 1. Invasive candidiasis candidiasis in adult non-
information according to jurisdiction; 2. Invasive aspergillosis in adult or 2. Oesophageal candidiasis neutropenic patients
adults only unless specified) paediatric patients who are refractory to 3. Prophylaxis of Candida infection in Eraxis:
or intolerant of first line therapy with patients undergoing allogeneic HSCT or 1. Oesophageal candidiasis
other antifungal agents patients expected to have neutropenia for 2. Candidaemia and other
3. Empirical therapy for presumed fungal [10 days forms of Candida infections,
infections in febrile, neutropenic adult or In children \16 years: including abdominal
paediatric patients 1. Invasive candidiasis abscesses and peritonitis
2. Prophylaxis of Candida infection in
patients undergoing allogeneic HSCT or
expected to have neutropenia [10 days
FDA: not licensed in the paediatric
population for prophylaxis of Candida
infection in patients expected to have
neutropenia [10 days
Routes of administration IV IV IV
(labelled and unlabelled routes)
PO dose NA NA NA
(Adult only)
(continued)
115
116
Class Echinocandins

Eraxis (US)
IV dosea 1., 2., and 3 [40kg: Ecalta:

(Adult only) 70 mg/day, then 50 mg OD 1. 100200 mg OD 200 mg/day, then 100 mg OD
If [80kg, 70 mg OD 2. 150 mg OD Eraxis:
3. 50 mg OD 1. 100 mg/day, then 50 mg OD
\40kg: 2. 200 mg/day, then 100 mg OD
1. 2 mg/kg/day
2. 3 mg/kg/day
3. 1 mg/kg/day
Dose adjustment if renal impairment No No No
Dose adjustment if liver impairment Yes No No
TDM recommended No No No
7
Class Echinocandins

Eraxis (US)
Main side effects Hypotension, peripheral oedema, tachycardia, fever, Fever, headache, hypokalaemia, Hypokalaemia,
chills, headache, rash, hypokalaemia, GI disturbance, hypomagnesaemia, GI GI disturbance,
abnormal LFTs, phlebitis, infusion reactions, anaemia disturbance, neutropenia, abnormal LFTs
thrombocytopenia
Drug interactions Cyclosporine, tacrolimus, rifampicin Few Few
Monitor for toxicity due to
sirolimus, nifedipine and
itraconazole if co-prescribed
S. Scedosporium, spp species, no activity, slight activity, + modest activity, ++ good activity, N no data, Eur Europe, US United States, EMEA European
Medicines Agency, FDA US Food and drug administration, BMT bone marrow transplantation, IV intravenous, neb nebulised, IT intrathecal, PO oral, NA not
applicable, OD once daily, BD twice daily, TDS three times daily, QDS four times daily, TDM therapeutic drug monitoring, GI gastrointestinal, LFTs liver
function tests, CYP3A4, Cytochrome P450 3A4 enzyme, CYP2C19, Cytochrome P450 2C19 enzyme.
a
Where revelant, the doses categorised numerically correspond to the numerically categorised indications in the Indication section
b
In immunocompromised patients with invasive mycoses
117
in-hospital mortality. Morrell et al. [42] found that in cases of candidaemia,

a [ 12-h delay in starting antifungal therapy after the time the first positive blood
culture sample was drawn was an independent predictor of mortality rate by
multiple logistic regression analysis. Collectively, these and related studies have
drawn into focus the way in which antifungal agents are being used in ICUs and
how prescribing practice might be improved.
7.6 Approaches to Antifungal Therapy in Intensive Care
How best to utilise systemic antifungal agents in ICU patients is the subject of
ongoing debate [43]. The challenges faced by intensivists and microbiologists are
to identify which patients are most at risk for invasive candidiasis, to determine at
what point it is appropriate to start antifungal therapy in a critically ill patient and
which antifungal agent to use.
It is now accepted that antifungals can be prescribed in one of several ways.
Firstly, there is prophylaxis, a practice that is more established in neutropenic
haematological malignancy patients. There are less convincing supportive data in
the ICU population. In considering what a preventive strategy should include,
Pfaller and Diekema [35] remind practitioners of the need to educate staff on the
importance of hand washing, optimal central vascular catheter care and prudent
antibiotic prescribing to prevent nosocomial candidiasis. They also reviewed meta-
analyses of published antifungal prophylaxis trials. The incidence of invasive
candidiasis was reduced in all meta-analyses by between 50 and 80%; three of the
five studies showed that the mortality rate was reduced while there was no
apparent overall increase in antifungal drug resistance. Despite these encouraging
results, the authors [35] suggest that further controlled trials are needed to identify
subsets of patients who would benefit most from prophylaxis and also to
emphasise the need to assess the efficacy of echinocandins, as earlier studies
predominantly evaluated fluconazole.
In the study of Bassetti et al. [38], stopping the use of prophylactic fluconazole
in their ICU resulted in a significant reduction in candidaemia caused by non-
albicans Candida spp., and there was no change in the incidence of invasive
candidiasis. Although studies showing that antifungal prophylaxis in ICU patients
reduced the number of IFIs, the effect on mortality rate and the overall benefits of
this strategy are not established. For reasons such as cost, concern about antifungal
resistance and drug-related adverse events, antifungal prophylaxis should only be
considered for selected high-risk patients.
Antifungal prophylaxis has received a grade BI recommendation from the
IDSA for patients at highest risk ([10% risk of invasive candidiasis) in ICUs with
high rates of invasive candidiasis (compared with normal rates of 12%) [13].
Fluconazole 400 mg (6 mg/kg) daily is recommended for such patients in this
setting. However, accurately identifying patients at [10% risk needs further
refining, as yet there is no universally agreed-upon risk prediction algorithm that
takes account of risk factors such as previous abdominal surgery, broad-spectrum

antibiotic therapy, presence of an intravascular catheter and number of sites col-
onised with Candida spp. A 2007 Cochrane review of the use of prophylactic
systemic antifungal agents in very-low-birth-weight infants cautiously found that
prophylactic fluconazole reduces the rates of invasive fungal infection but with no
statistically significant effect on overall mortality rates [44]. The role of other
prophylactic antifungals, such as orally administered nystatin and nebulised
amphotericin, is less clear, especially for ICU patients.
Alternative strategies to prophylaxis are, firstly, empirical therapy, where the
basis for giving antifungals is clinical evidence of fungal infection but without
mycological documentation; and, secondly, pre-emptive therapy, a newer concept
in which patients are identified for treatment according to predetermined risk
factors combined with several surrogate markers of fungal disease before it is
clinically evident [45]. Intensivists need to be cautious when interpreting micro-
biological culture results, for example, being careful not to equate the isolation of
Candida spp. from respiratory samples with the presence of an infection. Many
ventilated patients will have Candida spp. isolated from respiratory samples,
reflecting the altered flora due to antibiotic selection. Some studies used coloni-
sation indices as a basis for pre-emptive therapy, but many microbiology labora-
tories do not provide this service at the level equivalent to a clinical trial.
Furthermore, Candida spp. identified in colonisation samples do not necessarily
correlate with the species causing candidaemia; this appears to be especially the
case for C. glabrata candidaemias [46].
An alternative is to use the clinical approach initially to decide which patients
will receive antifungal treatment. Spellberg et al. [47] proposed an empirical
therapy algorithm for known or suspected cases of disseminated candidiasis based
on the patients general clinical status. Haemodynamically stable patients are
distinguished from those who unstable, the latter likely being those with either
severe sepsis and organ dysfunction or septic shock [45]. It is not clear whether or
not this is advocated for all such patients or limited to those in whom there may be
some additional evidence, such as fungal colonisation and prior broad-spectrum
antibiotic exposure. In the former, less seriously ill patient, the authors recommend
fluconazole but prefer an echinocandin, an amphotericin B formulation, or
voriconazole in the latter group. If the patient has colonisation with C. glabrata or
C. krusei, they recommend not using fluconazole.
Fluconazole is an attractive option for initial empirical therapy because it has
few side effects and is cheaper than the newer systemic antifungals. However, the
response rates of treating candidaemia reported in the comparative clinical trials
suggest that either amphotericin B or an echinocandin may be more efficacious,
even though superior response rates in individual studies were not significantly
different on statistical analysis [27, 48]. Zilberberg et al. [49] designed a decision
model aimed at assessing the cost-effectiveness of using an echinocandinin this
case, micafunginrather than fluconazole for empirical therapy of suspected
ICU-acquired candidaemia. It was based on a hypothetical 1,000-patient cohort.
A number of parameters were compared, including differences in mortality rates,
drug costs and cost per quality adjusted life year (QALY). The authors concluded
that preferred use of the echinocandin was cost effective. Their study builds on
earlier analysis of the optimal choice of empirical therapy, which also favoured an
echinocandin [50]. In an accompanying commentary to the Zilberberg study [49],
Golan [51] suggests the study may have overestimated the benefits of micafungin
compared with fluconazole and suggest further studies should be undertaken in
this area.
For antifungal therapy of a documented case of candidaemia in the ICU, the
IDSA guidelines [13] recommend either fluconazole or an echinocandin (AI grade)
with some caveats: Liposomal amphotericin B (AmBisome) was found to be as
effective as micafungin in a randomised double-blind trial [18], suggesting that it
is an alternative option, although as with other amphotericin B formulations, a
higher rate of treatment-related adverse events might be expected.
References
1. Cooke FJ, Rogers T (2005) Systemic antifungals. In: van Saene HKF, Silvestri L, de la Cal
MA (eds) Infection control in the intensive care unit, 2nd edn. Springer, Berlin, pp 155170
2. Schiller DS, Fung HB (2007) Posaconazole: an extended-spectrum triazole antifungal agent.
Clin Ther 29:18621886
3. Ullmann AJ, Lipton JH, Vesole DH et al (2007) Posaconazole or fluconazole for prophylaxis
in severe graft-versus-host disease. N Engl J Med 356:335347
4. Cornely OA, Maertens J, Winston DJ et al (2007) Posaconazole vs. fluconazole or
itraconazole prophylaxis in patients with neutropenia. N Engl J Med 356:348359
5. Walsh TJ, Anaissie EJ, Denning DW et al (2008) Treatment of aspergillosis: clinical practice
guidelines of the Infectious Diseases Society of America. Clin Infect Dis 46:327360
6. Walsh TJ, Raad I, Patterson TF et al (2007) Treatment of invasive aspergillosis with
posaconazole in patients who are refractory to or intolerant of conventional therapy: an
externally controlled trial. Clin Infect Dis 44:212
7. Greenberg RN, Mullane K, van Burik JA et al (2006) Posaconazole as salvage therapy for
zygomycosis. Antimicrob Agents Chemother 50:126133
8. van Burik JA, Hare RS, Solomon HF et al (2006) Posaconazole is effective as salvage therapy
in zygomycosis: a retrospective summary of 91 cases. Clin Infect Dis 42:e61e65
9. Page RL 2nd, Schwiesow J, Hilts A (2007) Posaconazole as salvage therapy in a patient with
disseminated zygomycosis: case report and review of the literature. Pharmacotherapy
27:290298
10. Zoller E, Valente C, Klepser ME (2010) Development, clinical utility, and place in therapy of
posaconazole for prevention and treatment of invasive fungal infections. Drug Des Devel
Ther 4:299311
11. Strorzinger D, Lichtenstern C, Weigand MA et al (2011) Posaconazole as part of the
antifungal armamentarium in the intensive care unit-case reports from a surgical ICU.
Mycoses 54(suppl 1):4548
12. Shields RK, Clancy CJ, Vadnerkar A et al (2010) Posaconazole serum concentrations among
cardiothoracic transplant recipients: factors impacting levels and correlation with clinical
response. Antimicrob Agents Chemother 55(3):13081311
13. Pappas PG, Kauffman CA, Andes D et al (2009) Clinical practice guidelines for the
management of candidiasis: 2009 update by the Infectious Diseases Society of America. Clin
14. Pappas PG, Rotstein CM, Betts RF et al (2007) Micafungin versus caspofungin for treatment
of candidemia and other forms of invasive candidiasis. Clin Infect Dis 45:883893
15. Vazquez JA (2010) Invasive fungal infections in the intensive care unit. Semin Respir Crit
Care Med 31:7986
16. Wagner C, Graninger W, Presterl E et al (2006) The echinocandins: comparison of their
pharmacokinetics, pharmacodynamics and clinical applications. Pharmacology 78:161177
17. Betts RF, Nucci M, Talwar D et al (2009) A multicenter, double-blind trial of a high-dose
caspofungin treatment regimen versus a standard caspofungin treatment regimen for adult
patients with invasive candidiasis. Clin Infect Dis 48:16761684
18. Kuse ER, Chetchotisakd P, da Cunha CA et al (2007) Micafungin versus liposomal
amphotericin B for candidaemia and invasive candidosis: a phase III randomised double-
blind trial. Lancet 369:15191527
19. Pappas PG, Rotstein CM, Betts RF et al (2007) Micafungin versus caspofungin for treatment
of candidemia and other forms of invasive candidiasis. Clin Infect Dis 45:883893
20. Dupont BF, Lortholary O, Ostrosky-Zeichner L et al (2009) Treatment of candidemia and
invasive candidiasis in the intensive care unit: post hoc analysis of a randomized, controlled
trial comparing micafungin and liposomal amphotericin B. Crit Care 13:R159
21. Queiroz-Telles F, Berezin E, Leverger G et al (2008) Micafungin versus liposomal
amphotericin B for pediatric patients with invasive candidiasis. Pediatr Infect Dis 27:820826
22. de Wet N, Llanos-Cuentas A, Suleiman J et al (2004) A randomized, double-blind, parallel-
group, dose-response study of micafungin compared with fluconazole for the treatment of
esophageal candidiasis in HIV-positive patients. Clin Infect Dis 39:842849
23. de Wet NT, Bester AJ, Viljoen JJ et al (2005) A randomized, double blind, comparative trial
of micafungin (FK463) vs. fluconazole for the treatment of oesophageal candidiasis. Aliment
Pharmacol Ther 21:899907
24. van Burik JA, Ratanatharathorn V, Stepan DE et al (2004) Micafungin versus fluconazole for
prophylaxis against invasive fungal infections during neutropenia in patients undergoing
hematopoietic stem cell transplantation. Clin Infect Dis 39:14071416
25. Denning DW, Marr KA, Lau WM et al (2006) Micafungin (FK463), alone or in combination
with other systemic antifungal agents, for the treatment of acute invasive aspergillosis.
J Infect 53:337349
26. Perkhofer S, Lass-Florl C (2009) Anidulafungin and voriconazole in invasive fungal disease:
pharmacological data and their use in combination. Expert Opin Investig Drugs 18:13931404
27. Reboli AC, Rotstein C, Pappas PG et al (2007) Anidulafungin versus fluconazole for invasive
candidiasis. N Engl J Med 356:24722482
28. Krause DS, Simjee AE, van Rensburg C et al (2004) A randomized, double-blind trial of
anidulafungin versus fluconazole for the treatment of esophageal candidiasis. Clin Infect Dis
39:770775
29. Kauffman CA, Bustamante B, Chapman SW et al (2007) Clinical practice guidelines for the
management of sporotrichosis: 2007 update by the Infectious Diseases Society of America.
30. Perfect JR, Dismukes WE, Dromer F et al (2010) Clinical practice guidelines for the
management of cryptococcal disease: 2010 update by the Infectious Diseases Society of
America. Clin Infect Dis 50:291322
31. Chapman SW, Dismukes WE, Proia LA et al (2008) Clinical practice guidelines for the
management of blastomycosis: 2008 update by the Infectious Diseases Society of America.
32. Galgiani JN, Ampel NM, Blair JE et al (2005) Coccidioidomycosis. Clin Infect Dis
41:12171223
33. Wheat LJ, Freifeld AG, Kleiman MB et al (2007) Clinical practice guidelines for the
management of patients with histoplasmosis: 2007 update by the Infectious Diseases Society
of America. Clin Infect Dis 45:807825
34. Cuenca-Estrella M, Rodriguez-Tudela JL (2010) The current role of the reference procedures
by CLSI and EUCAST in the detection of resistance to antifungal agents in vitro. Expert Rev
Anti Infect Ther 8:267276
35. Pfaller MA, Diekema DJ (2007) Epidemiology of invasive candidiasis: a persistent public
health problem. Clin Microbiol Rev 20:133163
36. Kett DH, Azoulay E, Echeverria PM et al. for the Extended Prevalence of Infection in the
ICU Study (EPIC. II) Group of investigators (2010) Candida bloodstream infections in
intensive care units: analysis of the extended prevalence of infection in an intensive care unit
study. Crit Care Med 39(4):665-670
37. Lass-Florl C (2009) The changing face of epidemiology of invasive fungal disease in Europe.
Mycoses 52:197205
38. Bassetti M, Ansaldi F, Nicolini L et al (2009) Incidence of candidaemia and relationship with
fluconazole use in an intensive care unit. J Antimicrob Chemother 64:625629
39. Pfaller MA, Moet GJ, Messer SA et al (2010) Candida bloodstream infections: comparison of
species distribution and antifungal resistance in community onset and nosocomial isolates in
the SENTRY antimicrobial surveillance program (20082009). Antimicrob Agents
Chemother 55(2):561566
40. Parkins MD, Sabuda DM, Elsayed S et al (2007) Adequacy of empirical antifungal therapy
and effect on outcome among patients with invasive Candida species infections. J Antimicrob
Chemother 60:613618
41. Garnacho-Montero J, Garcia-Garmendia JL, Barrero-Almodovar A et al (2003) Impact of
adequate empirical antibiotic therapy on the outcome of patients admitted to the intensive
care unit with sepsis. Crit Care Med 31:27422751
42. Morrell M, Fraser VJ, Kollef MH (2005) Delaying the empiric treatment of Candida
bloodstream infection until positive blood culture results are obtained: a potential risk factor
for hospital mortality. Antimicrob Agents Chemother 49:36403645
43. Cruciani M, Serpelloni G (2008) Management of Candida infections in the adult intensive
care unit. Expert Opin Pharmacother 9:175191
44. Clerihew L, Austin N, McGuire W (2007) Prophylactic systemic antifungal agents to prevent
mortality and morbidity in very low birth weight infants. Cochrane Database Syst Rev
(4):CD003850
45. Guery BP, Arendrup MC, Auzinger G et al (2009) Management of invasive candidiasis and
candidemia in adult non-neutropenic intensive care unit patients. Part II: Treatment. Intensive
Care Med 35:206214
46. Troughton JA, Browne G, McAuley DF et al (2010) Prior colonisation with Candida species
fails to guide empirical therapy for candidaemia in critically ill adults. J Infect 61:403409
47. Spellberg BJ, Filler SG, Edwards JE Jr (2006) Current treatment strategies for disseminated
candidiasis. Clin Infect Dis 42:244251
48. Rex JH, Bennett JE, Sugar AM et al (1994) A randomized trial comparing fluconazole with
amphotericin B for the treatment of candidemia in patients without neutropenia. N Engl J
Med 331:13251330
49. Zilberberg MD, Kothan S, Shorr AF (2009) Cost-effectiveness of micafungin as an
alternative to fluconazole empiric treatment of suspected ICU-acquired candidemia among
patients with sepsis: a model simulation. Crit Care 13:R94
50. Golan Y, Wolf MP, Pauker SG et al (2005) Empirical anti-Candida therapy among selected
patients in the intensive care unit: a cost effectiveness analysis. Ann Intern Med 143:857869
51. Golan Y (2009) Empiric anti-Candida therapy for patients with sepsis in the ICU: how little
is too little? Crit Care 13:180
Enteral Antimicrobials
8
M. Sanchez Garca, M. Nieto Cabrera, M. A. Gonzalez Gallego
and F. Martnez Sagasti
8.1 Introduction
Healthy individuals may carry one or more of the six potentially pathogenic
microorganisms (PPM), considered either normal or community-acquired, in their
upper [1, 2] and lower [3] gastrointestinal tract flora. These PPMs are Streptococcus
pneumoniae, Haemophilus influenzae, Moraxella catarrhalis, Staphylococcus
aureus, Escherichia coli, and Candida albicans. The presence of the opportunistic or
abnormal aerobic gram-negative bacilli (AGNB), Klebsiella, Enterobacter, Proteus,
Morganella, Citrobacter, Serratia, Acinetobacter, and Pseudomonas species, and of
methicillin-resistant S. aureus (MRSA) in the oropharynx and gastrointestinal tract
of healthy individuals is uncommon (see Chap. 2), although extended-spectrum
beta-lactamases (ESBL) harboring Enterobacteriaceae, as well as MRSA, are
increasingly being detected at hospital admission [4]. The nine abnormal bacteria are
predominantly carried by patients with an underlying condition, either chronic or
acute [5]. Illness severity and antibiotic use [3, 5, 6] are the most important factors in
conversion of the normal to the abnormal carrier state. Carriage of abnormal flora
invariably leads to high concentrations, i.e., abnormal bacteria overgrowth, in the
throat and gut of the critically ill [7, 8]. Overgrowth is defined as C105 of abnormal
flora per milliliter of saliva and/or gram of feces [9]. In turn, intestinal overgrowth
with AGNB is associated with systemic immune-system suppression [10, 11] and has
also been shown to be an independent risk factor for endogenous infection, endo-
toxemia, emergence of resistance, transmission via hands of carers, and outbreaks.
M. Snchez Garca (&)

Servicio de Medicina Intensiva,
Hospital Clnico San Carlos Universidad Complutense,
Madrid, Spain
124 M. Sanchez Garca et al.
8.2 Eradication and Control of Abnormal Flora Carriage

and Overgrowth
8.2.1 Parenterally Administered Antimicrobials
The most commonly used parenterally antimicrobials have been shown to

clear the three oropharyngeal microorganisms S. pneumoniae, H. influenzae, and
M. catarrhalis. For example, cefotaxime is excreted into saliva in concentrations
high enough to eradicate oropharyngeal carriage of these three community-
acquired respiratory PPMs [12]. The failure rate in clearing E. coli from the gut is
substantially due to the nonlethal fecal concentrations following excretion via bile
and mucus [12, 13], although relatively high concentrations of 153 kg have been
documented for ceftriaxone in human volunteers [14]. Systemic antifungals are not
associated with significant eradication or prevention of yeast carriage in critically
ill patients, as shown in a randomized controlled trial (RCT) of fluconazole [15].
Although penicillinase-resistant penicillins are widely regarded as the antistaph-
ylococcal agents of choice, there are no accurate data on the clearance of methi-
cillin-sensitive S. aureus (MSSA) carriage [16]. Effective eradication of MSSA
from throat and gut has, however, been reported for intravenously administered
(IV) cephradine [17].
Two grams of vancomycin IV produce concentrations of between 6 and 11 mg/kg
of feces. Enteral administration, however, results in fecal levels between 3,000 and
24,000 lg/g of feces [18, 19]. These pharmacokinetic data may explain why
systemic vancomycin fails to eradicate MRSA gut carriage. The same concept that
salivary, fecal, and mucus concentrations are in general nonlethal for AGNB
carried in the throat and gut applies to the commonly used systemic antimicrobials,
including beta-lactams, aminoglycosides, and fluoroquinolones [2022]. Even
the newer and more potent carbapenems and the combination of beta-lactams
and beta-lactamase inhibitors fail to clear AGNB, including Pseudomonas and
Acinetobacter spp. [23, 24]. This emphasizes the importance of measuring drug
levels at all relevant sites in addition to blood levels.
8.2.2 Enterally Administered Antimicrobials
Enterally administered antimicrobials are used to prevent and, if already present,

eradicate carriage and overgrowth with abnormal bacteria, including AGNB and
MRSA. Critically ill patients are unable to clear these opportunistic PPMs [8] due
to their underlying disease. Only recovery from their acute condition promotes
return to the normal carrier state [25]. Therefore, during the period of critical
illness, enterally administered antimicrobials are required to combat gut over-
growth of abnormal flora.
Administration of nonabsorbable enterally administered antimicrobials is based
on the experience that critically ill patients given parenterally administered anti-
biotics only have harmful gut overgrowth with abnormal flora [26]. Selective
8 Enteral Antimicrobials 125
decontamination of the digestive tract (SDD), a maneuver based mainly on enteral

administration of antimicrobials, is designed to convert the abnormal to the normal
carrier state. The purpose of SDD is therefore to prevent, or eradicate if initially
present, carriage and overgrowth of abnormal AGNB and MRSA. In addition, SDD
is used to eradicate MSSA and yeasts, which are carried by varying percentages of
healthy people. The overall aim of SDD using enterally administered antimicro-
bials is to reduce mortality and morbidity rates following recovery of systemic
immunity [11, 27] by preventing endogenous infections, resistance, and outbreaks
and reducing endotoxemia. A combination of nonabsorbable antimicrobials is
given enterally, which leaves indigenous flora relatively undisturbed, believed to
play a defensive role against abnormal bacterial carriage [28]. To qualify for
selective elimination of abnormal flora, antimicrobials should fulfil four criteria:
1. adequate spectrum, covering AGNB, MRSA, MSSA and yeasts, while sparing
indigenous, mostly anaerobic, flora as much as possible [28];
2. nonabsorbable, in order to achieve constant high intraluminal levels, which
would be lowered by absorption [29];
3. minimal inactivation by salivary, fecal, and food compounds, and no degra-
dation by fecal enzymes produced by indigenous anaerobic flora [30]; inter-
actions between feces, bacteria, and decontaminating antimicrobials
determining the microbiologically active fecal concentrations are of great
importance in the ultimate outcome of SDD;
4. bactericidal, with low minimal bactericidal concentrations (MBC) for AGNB,
MRSA, MSSA and yeasts, because there are no leucocytes in the human
alimentary canal to assist antimicrobial decontamination [31].
The most commonly used protocol is a combination of polymyxin and tobra-
mycin to clear AGNB and MSSA, amphotericin or nystatin to eradicate yeasts, and
vancomycin in case of MRSA (Table 8.1) [32].
8.2.2.1 Polymyxins
Polymyxins given enterally are nonabsorbable [33] and cover AGNB, including
ESBLs in general, as well as ESBLs conferring carbapenem resistance to
enterobacteria [34, 35]. These compounds have excellent activity against
P. aeruginosa and Acinetobacter spp., although not against Proteus, Morganella,
or Serratia spp. Polymyxins are selective in that they are not active against the
indigenousmainly anaerobicflora [36]. The mode of action is disruption of the
bacterial cell wall, making the bacterial cell permeable and leading to cell death.
This mechanism is independent of enzymatic systems [37], and acquired resistance
against polymyxins is uncommon [38]. Polymyxins are inactivated to a moderate
extent by proteins, fiber, food, cell debris, and salivary and fecal compounds, and
should therefore be given in a relatively high daily dose of 400 mg of polymyxin
E (300 mg of polymyxin B) [36]. Polymyxins should also be combined with
an aminoglycoside due to the gap in activity against Proteus, Morganella and
Serratia spp. The aminoglycoside should be active against P. aeruginosa because
polymyxins lose activity against this common intensive care unit (ICU) bacterium
Table 8.1 Enterally administered antimicrobials for eradication of carriage of potential

pathogens
Selective digestive decontamination Total daily dose in four divided doses

Target microorganisms \5 years 512 years [12 years
Oropharynx
AGNB Polymyxin E with tobramycin 2 g of 2% paste or gel
Yeasts Amphotericin B, or nystatin 2 g of 2% paste or gel
MRSA Vancomycin 2 g of 4% paste or gel
Gut
AGNB Polymyxin E (mg) 100 200 400
With tobramycin (mg) 80 160 320
Yeasts Amphotericin B (mg) 500 1,000 2,000
Nystatin (Units) 2 9 106 4 9 106 8 9 106
MRSA Vancomycin (mg) 2040/kg 2040/kg 5002,000
AGNB aerobic gram-negative bacilli, MRSA methicillin-resistant Staphylococci aureus
in the presence of feces. Polymyxins also neutralize endotoxin in the intestinal

lumen by adsorption [39].
8.2.2.2 Aminoglycosides
Aminoglycosides have several ideal features cited above for enterally adminis-
tered antimicrobials. They are active against a wide range of AGNB, including
P. aeruginosa, have a potent bactericidal activity similar to polymyxins, and
synergistic activity with polymyxins. Antipseudomonal aminoglycosides include
gentamicin, tobramycin, and amikacin. They are nonabsorbable, and the bacteri-
cidal activity is by inhibiting protein synthesis. Tobramycin is the least inactivated
by feces, followed by amikacin and gentamicin [4042]. According to an older
study [43] based on culture results, tobramycin is considered to be selective in
terms of leaving the indigenous flora undisturbed in doses \500 mg a day. A recent
investigation [8], ancillary to a randomized clinical trial [44], used more precise
molecular quantification methods and found reductions in the Faecalibacterium
prausnitzii group, which were attributed to enterally administered tobramycin.
Although an important component of intestinal microbiotaF. prausnitzii seems
to have several beneficial functions in the normal hostit is important to note that
the study is potentially biased because it was performed in a very small subgroup,
approximately 20%, of patients enrolled in one center and that, as recently shown
by the same group [45], other factors, including nutrition type and systemic
administration of antibiotics, may have been responsible for reductions in the
F. prausnitzii group count. The clinical relevance of this finding in critically ill
patients therefore remains to be elucidated. Blood levels of tobramycin and gen-
tamicin have been monitored during SDD [4648]. Aminoglycoside levels were
undetectable in most patients. Low concentrations, \1 mg/l, were measured in

ICU patients, particularly those with renal impairment. In a series of 19 patients
under standard SDD doses and with severe, acute renal failure requiring contin-
uous renal replacement therapy, tobramycin serum concentrations were B1 mg/l
[49]. In three patients with concomitant bowel ischemia, serum levels ranged
between 1.1 and 3.0 mg/l [49]. Although the three antipseudomonal aminogly-
cosides and polymyxins have a similar bactericidal activity, the total daily dose
recommended for tobramycin is 320 mg, which is lower compared with the
400 mg for the polymyxins, which are inactivated to a moderate extent by fecal
material. Aminoglycosides require the addition of polymyxins, as the emergence
of aminoglycoside-neutralizing enzymes is not uncommon [50]. Polymyxins are
thought to protect tobramycin from being inactivated by fecal enzymes.
Practically all ESBL-producing AGNB are sensitive to the enterally adminis-
tered combination of polymyxin/tobramycin [51]. Although rare, some ESBL-
producing AGNB, such as Klebsiella spp., may be resistant to tobramycin [52]. In
this case, tobramycin may be replaced by an aminoglycoside active against such
species, e.g., neomycin [53] or paromomycin [54]. Parenterally administered
antimicrobials that disregard the patients gut ecology may promote acquisition,
carriage, and subsequent overgrowth of ESBL-producing AGNB [55, 56].
Tobramycin has also been found to reduce endotoxin release [57].
8.2.2.3 Polyenes
The two polyenes used as decontaminating agents are either amphotericin B or
nystatin. They are fungicidal and highly selective, as fungi are the only PPM
covered by polyenes. They bind to a sterol of the plasma membrane and alter the
membrane permeability of the fungal cell, which leads to leakage of essential
metabolites and finally to fungal cell lysis. Absorption of polyenes is minimal
[47, 48], and emergence of resistance to them amongst yeasts and fungi is highly
uncommon [58]. Fecal inactivation of polyenes is high, explaining the high daily
dose of 2 g of amphotericin B and of 8 9 106 U of nystatin required for decon-
tamination purposes [59, 60] (Table 8.1).
8.2.2.4 Glycopeptides
Of the two glycopeptides, vancomycin and teicoplanin, the most experience as
decontaminating agents has been gathered for enterally administered vancomycin
[19]. Vancomycin is active against MRSA but cannot be considered as a selective
decontaminating agent as it covers the vast majority of the anaerobic Clostridium
spp. Thus, SDD protocols do not routinely include enterally administered van-
comycin because of its negative impact on the gut ecology. Enterally administered
vancomycin is only recommended to eradicate MRSA carriage and overgrowth
and should always be given in combination with polymyxin/tobramycin/ampho-
tericin B (PTA) to offset the potential for AGNB and yeast overgrowth as a
consequence of using a nonselective decontaminating agent [61]. The mode of
action is bactericidal, as vancomycin is bound rapidly and irreversibly to cell
walls of sensitive bacteria, thereby inhibiting cell-wall synthesis. Vancomycin
absorption is rare [48]. Inactivation by proteins, fiber, food, and feces is substantial,
hence the high daily dose of 2 g (Table 8.1).
8.3 Efficacy of Enterally Administered Polymyxin/

Tobramycin, Polyenes, and Vancomycin in Eradicating
AGNB, Yeast, and MRSA Carriage and Overgrowth
Sixty randomized controlled trials (RCTs) evaluating SDD were conducted [62]
between 1987 and 2010 (Chap. 13). The antibiotic combinations most frequently
used were polymyxin/tobramycin and polymyxin/gentamicin, in approximately
two-thirds and one-third of trials, respectively. Ten meta-analyses of RCTs on SDD
all invariably show a significant reduction in infection, and five meta-analyses report
a mortality rate reduction [63].
Surveillance cultures of the throat and rectum are an integral part of enterally
administered antimicrobial protocols [64]. Monitoring the carrier state in the
critically ill receiving antimicrobials enterally is essential, as only surveillance
cultures allow monitoring of compliance and efficacy. SDD is considered to be
effective only if surveillance samples show AGNB, MRSA, MSSA, and yeast
eradication. In addition, comparison of microorganisms cultured in surveillance
and diagnostic samples allows classification of ICU-acquired infections in
endogenous, i.e., SDD failure, and exogenous, i.e., hygiene failure [64]. Surveil-
lance samples were taken in most trials, predominantly twice-weekly throat and/or
rectal swab, although only a few provided classification of the type of infections
detected [65].
8.3.1 Aerobic Gram-Negative Bacilli (AGNB)
Most RCTs report effective clearance of AGNB carriage following polymyxin/

tobramycin administration. Reductions of AGNB in surveillance cultures can be
detected within 23 days of starting SDD [65]. Abnormal carriage in the throat is
eradicated within 3 days, whereas it takes 7 days for abnormal rectal carriage to be
eradicated, depending on the return of peristalsis. A meta-analysis of the impact
of SDD on AGNB carriage [66] shows impressive 87 and 85% reductions in
oropharyngeal and rectal carriage, respectively. A recent study employing more
sensitive molecular methods shows significantmore than tenfoldmean
reduction in Enterobacteriaceae [8].
8.3.2 Yeasts
Data on yeast carriage show that enterally administered polyenes significantly

reduce the odds ratio (OR) for carriage to 0.31 (0.180.54) [67]. More recent trials,
such as one multicenter RCT enrolling 6,000 patients [44], confirm this effect.
8.3.3 MRSA
Six RCTs included vancomycin enterally but none analyzed the impact on MRSA
carriage and infections [48, 6872]. A 4% vancomycin gel applied in the lower
cheeks was associated with a significantly reduced OR for oropharyngeal MRSA
carriage to 0.25 (0.090.69) in one RCT [73]. A beforeafter study performed in a
burn unit observed significant relative risk (RR) reductions [80% for acquisition
of MRSA carriage and infection [74] after the introduction of routine enterally
administered vancomycin, and a prospective nonrandomized study performed in a
medical/surgical unit revealed similar findings [75]. The recent detection of an
outbreak of linezolid-resistant but vancomycin-susceptible MRSA [76] implies a
potential application for prevention and control of this type of problem with
topical vancomycin.
Seven studies report fecal levels of one or more decontaminating agents poly-
myxin, tobramycin, gentamicin, amphotericin B, nystatin, and vancomycin [19, 43,
59, 60, 7779]. Compared with polymyxin, tobramycin was less inactivated by
fecal material. In one study, fecal specimens contained tobramycin levels of at least
100 mg/L feces following the daily intake of 300 mg of tobramycin [43]. In
another study, individuals taking 600 mg of tobramycin daily showed [500 mg/L
of fecal sample [77]. Polymyxin is moderately inactivated by mucosal cells, fiber
and feces, and hence the variation in fecal drug levels. Polymyxin was not detected
in one-third of individuals who took 600 mg of polymyxin daily [78]. One-third
had fecal levels [1,000 mg/L of feces, whereas the remaining individuals showed
polymyxin levels between 16 and 1,000 mg/L of feces. Tobramycin at a daily dose
of 320 added to 400 mg of polymyxin is the most commonly used combination for
eradicating AGNB carriage and overgrowth due to its synergism and relatively less
fecal inactivation [80]. Vancomycin inactivation by fecal material is high. In one
study, vancomycin po was given in doses of 2 g daily for 7 days, and the mean
concentration in 25 stool samples obtained during treatment was 3,100 400 lg/g
(range of 9058,760 lg/g) [26]. Fecal concentrations of polymyxin E and genta-
micin were measured in 38 stool samples obtained from 15 patients [79]. The levels
of both were \20 lg/ml of feces in ten stools. The remaining 28 samples showed
fecal polymyxin E levels of 94 174 mg/L (median 42 lg/ml, range 01,055 mg/L)
and gentamicin levels of 466 545 mg/L (median 196 mg/L, range 02,098 lg/ml).
Inactivation of polyenes, including amphotericin B and nystatin, by fecal material
is high. Daily doses of 2,000 mg of amphotericin B or 8 9 106 U of nystatin were
associated with fecal levels of 60 and 20 mg/L of feces, respectively [59, 60].
Most parenterally administered antimicrobials do not act upon gut flora.
However, fluoroquinolones including ciprofloxacin have been shown to possess
the pharmacokinetic characteristic of transintestinal secretion [81]. A substantial
amount of intravenously administrated ciprofloxacin is excreted via mucus rather
than via bile (15 vs. 1%), leading to high fecal ciprofloxacin concentrations. The
mean fecal level of ciprofloxacin was 108.7 mg/L of feces following parenteral
administration of a daily dose of 400 mg of ciprofloxacin [81].
The antifungal 5-flucytosine is a small molecule that also possesses the phar-
macokinetic property of transintestinal secretion, i.e., 10% of systematically
administered 5-flucytosine is excreted via mucus into the gut [82]. The good
penetration of flucytosine into most body tissues and fluids has been ascribed to its
high water solubility, low molecular weight, and low protein binding properties.
These features of ciprofloxacin and 5-flucytosine can be useful in critically ill
patients in whom rectal swabs remain positive for AGNB and yeasts following
1 week of polymyxin/tobramycin/amphotericin B treatment. Failure of the classic
PTA protocol may be due to AGNB and yeasts already translocated into the gut-
associated lymphoid tissue on admission and hence escape the intraluminal lethal
activity of nonabsorbable PTA. Three days of high intravenous doses of cipro-
floxacin or 5 flucytosine has been shown to assist effective SDD as measured by
surveillance samples negative for AGNB and yeasts in patients with inflamed gut.
Apart from monitoring efficacy and compliance of enterally administered
antimicrobials, surveillance samples, in particular rectal swabs, provide the unique
method of detecting antimicrobial resistance at an early stage, allowing prompt
treatment adjustment. In three RCTs [22, 70, 83], throat and rectal swabs were also
cultured on agar plates containing 2 mg/l of polymyxin, 4 mg/l of tobramycin, and
6 mg/l of vancomycin. For example, Proteus and Serratia spp. intrinsically
resistant to polymyxins may become resistant to tobramycin. There are reports that
describe the enteral use of amikacin [84] and paromomycin [29] replacing
tobramycin to eradicate tobramycin-resistant Proteus and Serratia, respectively.
8.4 Impact on Clinical and Epidemiological Endpoints

of Abnormal Flora Eradication Using Enterally
Administered Antimicrobials
8.4.1 Infectious Morbidity and Attributable Mortality
Infection-control policies that include SDD have four fundamental features [64]:
1. enterally administered antimicrobials to decontaminate the gastrointestinal
tract; they are combined with an oropharyngeal decontamination procedure
using a paste or gel containing 2% PTA; the aim of administering oropha-
ryngeal and intestinal nonabsorbable antimicrobials is preventing secondary
endogenous infections;
2. antibiotics given parenterally immediately upon admission to control primary
endogenous infections;
3. hygiene to control exogenous infections;
4. surveillance samples to monitor the SDD protocol.
The two most complete meta-analyses demonstrate that enterally administered
polymyxins with tobramycin or gentamicin significantly reduce AGNB infections
[85, 86]. Enterally administered polyenes significantly reduced both the number of
patients with yeast infections and episodes of yeast infections in 42 of 54 RCTs
evaluating the efficacy of antifungal polyenes as part of SDD [67]. Lower-airway
infections due to MRSA were significantly reduced in the RCT using 4%

vancomycin gel [73], as well as in two open trials employing both a 4% vanco-
mycin paste and vancomycin enterally [74, 75].
The most compelling evidence that infection is responsible for increased
mortality derives from RCTs of SDD. Several trials of large enough sample size
[22, 44, 70] documented a significant absolute mortality-rate reduction in patients
receiving the prophylactic regimen and a significantly reduced RR for developing
pneumonia of 0.20 (0.070.58) and bloodstream infections of 0.38 (0.170.83)
[70]. Whereas it is possible that these clinical benefits of SDD arise from reasons
other than preventing both early primary endogenous and late secondary endog-
enous infectionse.g., reduced absorption of gut endotoxinthe most plausible
interpretation of these recent data is that infection is responsible for a relative
increase in risk of ICU mortality rates of 2040%. A recent trial found a lower
mortality risk in both medical and surgical patients and in all severity groups [22].
8.4.2 Endotoxemia
AGNBs present as overgrowth in the gut are the major source of endotoxin in the
human body. Up to 10 mg of fecal endotoxin per gram of feces has been measured
in critically ill patients with AGNB gut overgrowth [87]. For example, gut
ischemia at the time of cardiac surgery and liver transplantation promotes trans-
mural migration or translocation of AGNB present in concentrations of C105/g of
feces [88]. Most translocating AGNB are killed by macrophages of gut-associated
lymphoid tissue, including the liver, which maintains bloodstream sterility.
However, the subsequent release of endotoxin may spill over into the bloodstream
and often lead to fluctuating levels of endotoxemia [89]. The enterally adminis-
tered combination of polymyxin/tobramycin has been shown to significantly
reduce fecal endotoxin load by a factor of 104 [90]. Five RCTs evaluated the
impact of SDD on endotoxemia: three during cardiopulmonary bypass surgery
[9193] and two in liver transplant patients [94, 95]. Two trials reported a sig-
nificant reduction following SDD [91, 92]; three trials failed to show a difference
[9395]. The cardiac patients received enterally administered polymyxin/tobra-
mycin 3 days preoperatively in the two positive studies. In the negative cardiac
study, tobramycin was replaced by neomycin, a poor antiendotoxin agent [96, 97].
In the two liver transplant studies, polymyxin/tobramycin was started 12 h
preoperatively and postoperatively only [98].
8.4.3 Antimicrobial Resistance
The most frequently used argument against the SDD strategy, and also a sincere
concern of some intensivists, is that the routine use of SDD may promote the
development of bacterial resistance or select resistant PPMs. This type of rea-
soning directly derives from the well-known fact that extensive systemic antibiotic
use is associated with induction of resistance and selection of resistant microor-

ganisms, sometimes with dramatic reductions in therapeutic options [35, 76].
Mechanisms contributing to an increase in the number of patients who carry
resistant microorganisms are: (1) admission to the ICU of patients already carriers
of resistant microorganisms; (2) selection and induction or mutation of resistant
microorganisms following antibiotic pressure; (3) transmission of resistant
microorganisms while in the ICU. Effective control of carriage/overgrowth using
antimicrobials enterally guarantees control of the import of resistance, control of
induction or mutation at the gut level, and significantly reduced transmission due
to the reduced number of carriers combined with a lower level of carriage of
resistant microorganisms (Chaps. 12 and 28).
The most extensive meta-analyses demonstrate the virtual absence of any
reported resistance, with all assessed RCTs being free of this problem and of
subsequent superinfections and/or epidemics of multiresistant strains [85, 86].
In one multicenter RCT from Madrid [65], MRSA carriage but not infection
was significantly greater in the test group. An immediately posttrial observational
study in 100 consecutive patients receiving SDD (enterally administered antimi-
crobial regimen: polymyxin E-gentamicin-amphotericin B, without vancomycin)
showed a progressive significant reduction from 27.6 to 8% and ruled out a causal
relationship [65]. The coordinating ICU of the trial has been using SDD in long-
term ([48 h) intubated patients for [20 years now, without significant resistance
problems. This accords with data from four SDD studies in which the endpoint of
resistance was evaluated over periods of 2, 2.5, 6 and 7 years [84, 99102],
whereas two studies evaluating the emergence of resistant microorganisms fol-
lowing SDD discontinuation failed to show any negative effect [102, 103]. Two
trials designed to evaluate the effect of SDD on mortality and resistance rate
showed both significant mortality-rate reductions [22, 44] and demonstrated sig-
nificant reductions in carriage of resistant bacteria. The trial comparing patients on
an SDD unit versus patients receiving parenterally administered antibiotic-only
approach in a different ICU showed significantly fewer carriers of multiresistant
AGNB and P. aeruginosa in the SDD unit than in the control unit [22]. A mul-
ticenter trial enrolling 6,000 patients also showed significant reductions in ceft-
azidime-resistant AGNB and P. aeruginosa [44].
A follow-up study in patients in hospital wards after discharge from the ICU
performed in two of the 13 participating centers detected an increased incidence of
infections in patients who had been in the SDD test group [104]. Per 100 surgical
procedures, the incidences of respiratory tract and blood-stream infections were
similar between the control and the decontaminated groups. The global increase in
infection was mainly due to, presumably exogenous, surgical site infections, 18 of
26 of which were superficial and 15 had no or negative cultures.
Another ancillary study of this trial [105] evaluated the influence of SDD on
bacterial resistance in point prevalence cultures obtained during the trial from all
concomitant ICU patients not enrolled in the trial. The study confirms that SDD
not only reduces resistant AGNB carriage in individual patient [8, 22, 44] but
protects other patients not receiving SDD probably by reducing transmission
[105]. This beneficial effect has been known for many years [106] and explains
why the efficacy of SDD increases with higher prevalence of the maneuver, i.e.,
during routine use in all eligible patients or without a simultaneous placebo control
group during trials. After SDD discontinuation, significant increases in some
resistance markers were detected (ceftazidime), although not in others (cipro-
floxacin, tobramycin). Resistance levels remained stable over the next 12 months
without SDD [105]. The various sources of bias for the results of this are appro-
priately addressed by the authors in the discussion section of their manuscript.
Six RCTs conducted in ICUs in which MRSA was endemic at the time of the
study showed a trend toward higher MRSA colonization [65] or infection rates in
patients receiving SDD [107112]. MRSA, by design, is not covered by SDD
antimicrobials. Inevitably, SDD exerts selective pressure on this PPM. Hence,
proponents of SDD have always accepted this possibility and proposed an SDD
strategy consisting of surveillance cultures to detect MRSA overgrowth in carriers,
combined with vancomycin administered either oropharyngeally [73, 113] or
oropharyngeally and enterally to control MRSA overgrowth [74, 75]. When there
is a serious clinical MRSA problem, this approach can be used as a prophylactic
policy in all high-risk groups. These patients should receive PTA combined with
enterally administered vancomycin on admission, and throat and rectal swabs
should be taken during the entire ICU stay. With an incidence of less than one
event per week, SDD that includes vancomycin enterally can be commenced as
treatment if swabs prove positive [114].
8.4.4 Transmission of Abnormal Flora Via Hands of Carers
The high density of S. aureus, both methicillin sensitive and resistant, in the
oropharynx and gut promotes skin carriage and hand and environmental con-
tamination. Quantitative studies from the early 1950s demonstrated that S. aureus
overgrowth in the nasal cavity leads to skin carriage of S. aureus in 44% of
individuals, but only in 16% if the level of contamination of nasal secretions was
\105 S. aureus [115]. Airborne dissemination was also a function of the number of
microorganisms present in the nose. However, similar research from the late 1950s
demonstrated that the weight of microorganisms released to the environment by
the fecal carrier greatly exceeded that of organisms released by the nasal carrier
[116, 117]. Twenty years later, the importance of fecal carriage of MRSA in
children was shown from air contamination studies during nappy changing, when
patients who were fecal carriers yielded the same type from the air [118]. More
recently, gut overgrowth of MRSA was shown to be associated with a significant
amount of MRSA dispersed from the perianal site into clothing and bedding and
hence into the environment [119]. Long-stay patients invariably have overgrowth
in their throat and gut of [109 potential pathogens per milliliter of saliva or gram
of feces [7]. Washing a patient or changing a nappy may lead to hand contami-
nation of health-care workers to levels of [106 PPM/cm2 of finger surface [120].
For hand hygiene to be effective, a disinfecting agent such as 0.5% chlorhexidine
in 70% alcohol is required, and the procedure must take at least 2 min. Under these
circumstances, contamination levels are lowered at most by 104 microorganisms,
still leaving up to 102 per cm2 of finger surface [121]. These quantitative data show
that the intervention of hand disinfection in a busy ICU with a few long-stay
patients can only ever hope to reduce transmission but never abolish it [122]. From
an SDD perspective, even under the hypothetical circumstances of completely
clearing hand contamination, hand hygiene could never exert an influence on the
major infection problem of primary endogenous infection with a magnitude of
between 60 and 85% of all infections (Chap. 5). The intervention of hand disin-
fection also fails to clear oropharyngeal and gastrointestinal carriage and/or
overgrowth of PPM present on arrival. However, high standards of hygiene,
including hand disinfection, are part of the SDD infection-control protocol. This
protocol aims to reduce the level of hand contamination below which transmission
occurs. It is possible to achieve these low levels, as the enterally administered
antimicrobials eradicate throat and gut carriage of PPM and substantially reduce
overall levels of PPM density on the patients skin. In this way, handwashing
becomes more effective in controlling transmission of PPM and subsequent
endogenous and exogenous infections.
8.4.5 Outbreaks
Observations on the characteristics of infection outbreaks show that once

approximately half the ICU population carries the outbreak strain in overgrowth
concentrations of minimally 3+ on the semiquantitative culturing scale [123],
spread via the hands of carers is impossible to prevent, and an outbreak is difficult
to avoid. Dissemination of AGNB invariably occurs whether there is oropharyn-
geal or intestinal overgrowth [53, 106]. In ICUs with ongoing endemicity of
multiresistant Klebsiella [53, 124], C. parapsilosis [125], and MRSA [61, 75],
reinforcement of conventional hygiene measures, including hand disinfection,
invariably failed to stop the outbreaks. The introduction of enterally administered
antimicrobials was reported to be successful in controlling AGNB, yeast, and
MRSA outbreaks in five reports [53, 61, 75, 124, 125]. In a French study, patients
were randomized and given antimicrobials either enterally or not enterally [53].
Fecal carriage of the ESBL-producing Klebsiella strain was cleared in 50% of
patients, and the outbreak was under control within 8 weeks. In a Manchester, UK,
ICU all patients received SDD, and a Klebsiella outbreak was stopped within
3 weeks [124]. Similarly, gut carriage/overgrowth is the main source of Candida
and MRSA, which are then transmitted via the hands of health-care workers [61,
75, 125]. On a Liverpool, UK, neonatal ICU, enterally administered nystatin was
given as treatment to neonates who were identified to carry the outbreak strain of
C. parapsilosis, and the outbreak was controlled within 14 weeks [125]. Enterally
administered vancomycin added to the classic SDD was an effective outbreak
control measure in Italian [61] and Spanish [75] ICUs with MRSA endemicity.
In the Spanish study, enterally administered vancomycin given as prophylaxis to
all patients at high risk was more effective in controlling endemicity compared
with when administered to confirmed carriers only. The main concerns about its
prophylactic use is vancomycin-resistant enterococci (VRE) [126]. Vancomycin
resistance among low-level pathogens such as enterococci is an endemic problem
in ICUs in the USA.
SDD, including enterally administered vancomycin, has been evaluated in five
European studies in mechanically ventilated patients [61, 73, 75, 113, 127]. None of the
studies report an increased infection rate due to VRE. They evaluated SDD including
vancomycin in ICUs without VRE history, and in one study VRE was imported into the
unit but no change in policy was required, as rapid and extensive spread did not occur
[75]. Recent literature shows that parenterally administered antibiotics that do not
respect the patients gut ecology rather than high doses of enterally administered
vancomycin promote the emergence of VRE in the gut [128, 129].
8.5 Conclusions
SDD using enterally administered antimicrobials is now an evidence-based proto-

col. It is the best-ever evaluated intervention in intensive care medicine for reducing
infectious morbidity and mortality. It is inexpensive and without side effects in
terms resistance emergence. In intensive care units (ICUs) using enterally admin-
istered antimicrobials, gut carriage of potential pathogens, both sensitive and
resistant, is significantly reduced with SDD. However, hand disinfection as a gen-
eral hygiene procedure is still valuable and can be expected to be more effective in
ICUs where all long-stay patients are successfully decontaminated and are thus free
from aerobic gram-negative bacilli (AGNB), yeast, and Staphylococcus aureus
overgrowth, This subsequently reduces hand contamination and, consequently, the
chance of transmission. SDD is the gold standard to which new maneuvers of
infection control should be compared.
References
1. Aas JA, Paster BJ, Stokes LN et al (2005) Defining the normal bacterial flora of the oral
cavity. J Clin Microbiol 43(11):57215732
2. Preza D, Olsen I, Willumsen T et al (2009) Diversity and site-specificity of the oral
microflora in the elderly. Eur J Clin Microbiol Infect Dis 28(9):10331040
3. Raum E, Lietzau S, von Baum H et al (2008) Changes in Escherichia coli resistance
patterns during and after antibiotic therapy: a longitudinal study among outpatients in
Germany. Clin Microbiol Infect 14(1):4148
4. Friedmann R, Raveh D, Zartzer E et al (2009) Prospective evaluation of colonization with
extended-spectrum beta-lactamase (ESBL)-producing Enterobacteriaceae among patients at
hospital admission and of subsequent colonization with ESBL-producing Enterobacteriaceae
among patients during hospitalization. Infect Control Hosp Epidemiol 30(6):534542
5. Tacconelli E, De Angelis G, Cataldo MA et al (2009) Antibiotic usage and risk of
colonization and infection with antibiotic-resistant bacteria: a hospital population-based
study. Antimicrob Agents Chemother 53(10):42644269
6. Rodriguez-Bano J, Alcala J, Cisneros JM et al (2009) Escherichia coli producing SHV-type

extended-spectrum beta-lactamase is a significant cause of community-acquired infection.
J Antimicrob Chemother 63(4):781784
7. van Saene HK, Taylor N, Donnell SC et al (2003) Gut overgrowth with abnormal flora: the
missing link in parenteral nutrition-related sepsis in surgical neonates. Eur J Clin Nutr
57(4):548553
8. Benus RF, Harmsen HJ, Welling GW et al (2010) Impact of digestive and oropharyngeal
decontamination on the intestinal microbiota in ICU patients. Intensive Care Med
36(8):13941402
237(4):419427
10. Marshall JC, Christou NV, Meakins JL (1988) Small-bowel bacterial overgrowth and
systemic immunosuppression in experimental peritonitis. Surgery 104(2):404411
11. Horton JW, Tan J, White DJ et al (2004) Selective decontamination of the digestive tract
attenuated the myocardial inflammation and dysfunction that occur with burn injury. Am J
Physiol Heart Circ Physiol 287(5):H2241H2251
12. Alcock SR (1990) Short-term parenteral antibiotics used as a supplement to SDD regimens.
Infection 18(suppl 1):S14S18
13. Sompolinsky D, Yaron V, Alkan WJ (1967) Microbiological changes in the human fecal
flora following the administration of tetracyclines and chloramphenicol. Am J Proctol
18(6):471478
14. Pletz MW, Rau M, Bulitta J et al (2004) Ertapenem pharmacokinetics and impact on
intestinal microflora, in comparison to those of ceftriaxone, after multiple dosing in male
and female volunteers. Antimicrob Agents Chemother 48(10):37653772
15. Garbino J, Lew DP, Romand JA et al (2002) Prevention of severe Candida infections in
nonneutropenic, high-risk, critically ill patients: a randomized, double-blind, placebo-
controlled trial in patients treated by selective digestive decontamination. Intensive Care
Med 28(12):17081717
16. Chang FY, Peacock JE Jr, Musher DM et al (2003) Staphylococcus aureus bacteremia:
recurrence and the impact of antibiotic treatment in a prospective multicenter study.
Medicine (Baltimore) 82(5):333339
17. van Saene R, Fairclough S, Petros A (1998) Broad- and narrow-spectrum antibiotics: a
different approach. Clin Microbiol Infect 4(1):5657
18. Geraci JE, Heilman FR, Nicolson DR et al (1956) Some laboratory and clinical experiences
with a new antibiotic, vancomycin. Proc Staff Meet Mayo Clin 31(21):564582
19. Tedesco F, Markham R, Gurwith M et al (1978) Oral vancomycin for antibiotic-associated
pseudomembranous colitis. Lancet 2(8083):226228
20. DAgata EM, Venkataraman L, DeGirolami P et al (1999) Colonization with broad-
spectrum cephalosporin-resistant gram-negative bacilli in intensive care units during a
nonoutbreak period: prevalence, risk factors, and rate of infection. Crit Care Med
27(6):10901095
21. Petros AJ, OConnell M, Roberts C et al (2001) Systemic antibiotics fail to clear multidrug-
resistant Klebsiella from a pediatric ICU. Chest 119(3):862866
22. De Jonge E, Schultz MJ, Spanjaard L et al (2003) Effects of selective decontamination of
randomised controlled trial. Lancet 362(9389):10111016
23. Corbella X, Montero A, Pujol M et al (2000) Emergence and rapid spread of carbapenem
resistance during a large and sustained hospital outbreak of multiresistant Acinetobacter
baumannii. J Clin Microbiol 38(11):40864095
24. Toltzis P, Yamashita T, Vilt L et al (1998) Antibiotic restriction does not alter endemic
colonization with resistant gram-negative rods in a pediatric intensive care unit. Crit Care
Med 26(11):18931899
25. Ketai LH, Rypka G (1993) The course of nosocomial oropharyngeal colonization in patients
recovering from acute respiratory failure. Chest 103(6):18371841
26. van Saene HK, Petros AJ, Ramsay G, Baxby D (2003) All great truths are iconoclastic:
Care Med 29(5):677690
27. Yao YM, Lu LR, Yu Y et al (1997) Influence of selective decontamination of the digestive
tract on cell-mediated immune function and bacteria/endotoxin translocation in thermally
injured rats. J Trauma 42(6):10731079
28. van der Waaij D (1992) History of recognition and measurement of colonization resistance
of the digestive tract as an introduction to selective gastrointestinal decontamination.
Epidemiol Infect 109(3):315326
29. Bodey GP (1981) Antibiotic prophylaxis in cancer patients: regimens of oral, nonabsorbable
antibiotics for prevention of infection during induction of remission. Rev Infect Dis
3(suppl):S259S268
30. van Saene HK, Stoutenbeek CP (1987) Selective decontamination. J Antimicrob Chemother
20(4):462465
31. Harris JC, Dupont HL, Hornick RB (1972) Fecal leukocytes in diarrheal illness. Ann Intern
Med 76(5):697703
32. Stoutenbeek CP (1989) Topical antibiotic regimen. In: van Saene HK, Stoutenbeek CP,
Lawin P, Ledingham IM (eds) Infection control by selective decontamination. Springer,
Heidelberg, pp 95101
33. Guyonnet J, Manco B, Baduel L et al (2010) Determination of a dosage regimen of colistin
by pharmacokinetic/pharmacodynamic integration and modeling for treatment of G.I.T.
disease in pigs. Res Vet Sci 88(2):307314
34. Nordmann P, Cuzon G, Naas T (2009) The real threat of Klebsiella pneumoniae
carbapenemase-producing bacteria. Lancet Infect Dis 9(4):228236
35. Kumarasamy KK, Toleman MA, Walsh TR et al (2010) Emergence of a new antibiotic
resistance mechanism in India, Pakistan, and the UK: a molecular, biological, and
epidemiological study. Lancet Infect Dis 10(9):597602
36. van Saene JJ, van Saene HK, Tarko-Smit NJ, Beukeveld GJ (1988) Enterobacteriaceae
suppression by three different oral doses of polymyxin E in human volunteers. Epidemiol
Infect 100(3):407417
37. Sogaard H (1982) The pharmacodynamics of polymyxin antibiotics with special reference
to drug resistance liability. J Vet Pharmacol Ther 5(4):219231
38. Hawser SP (2010) Susceptibility of Klebsiella pneumoniae clinical isolates from 2007 to
2009 to colistin and comparator antibiotics. Int J Antimicrob Agents 2010 36(4):383384
39. Danner RL, Joiner KA, Rubin M et al (1989) Purification, toxicity, and antiendotoxin
activity of polymyxin B nonapeptide. Antimicrob Agents Chemother 33(9):14281434
40. Veringa EM, van der Waaij D (1984) Biological inactivation by faeces of antimicrobial
drugs applicable in selective decontamination of the digestive tract. J Antimicrob
41. van Saene JJ, van Saene HK, Stoutenbeek CP, Lerk CF (1985) Influence of faeces on the
activity of antimicrobial agents used for decontamination of the alimentary canal. Scand J
Infect Dis 17(3):295300
42. Hazenberg MP, Pennock-Schroder AM, van de Merwe JP (1985) Binding to and
antibacterial effect of aztreonam, temocillin, gentamicin and tobramycin on human
faeces. J Hyg (Lond) 95(2):255263
43. Mulder JG, Wiersma WE, Welling GW, van der WD (1984) Low dose oral tobramycin
treatment for selective decontamination of the digestive tract: a study in human volunteers.
and oropharynx in ICU patients. N Engl J Med 360(1):2031
45. Benus RF, van der Werf TS, Welling GW et al (2010) Association between
Faecalibacterium prausnitzii and dietary fibre in colonic fermentation in healthy human
subjects. Br J Nutr 29:18
46. Cavaliere F, Sciarra M, Crociani E et al (1988) Serum levels of tobramycin during selective
decontamination of the gastrointestinal tract. Minerva Anestesiol 54(5):223226
47. Zobel G, Kuttnig M, Grubbauer HM et al (1991) Reduction of colonization and infection
rate during pediatric intensive care by selective decontamination of the digestive tract. Crit
Care Med 19(10):12421246
48. Gaussorgues P, Salord M, Sirodot M et al (1991) Efficacit de la dcontamination digestive
sur la survenue des bactrimies nosocomiales chez les patients sous ventilation
mchanique et recevant des betamimtiques. Ranimation Soins Intensifs Mdecin
dUrgence 7:169174
49. Mol M, van Kan HJ, Schultz MJ, De Jonge E (2008) Systemic tobramycin concentrations
during selective decontamination of the digestive tract in intensive care unit patients on
continuous venovenous hemofiltration. Intensive Care Med 34(5):903906
50. Smith CA, Baker EN (2002) Aminoglycoside antibiotic resistance by enzymatic
deactivation. Curr Drug Targets Infect Disord 2(2):143160
51. Rodriguez-Bano J, Navarro MD, Romero L et al (2006) Clinical and molecular
epidemiology of extended-spectrum beta-lactamase-producing Escherichia coli as a cause
of nosocomial infection or colonization: implications for control. Clin Infect Dis 42(1):
3745
52. Al Naiemi N, Heddema ER, Bart A et al (2006) Emergence of multidrug-resistant gram-
negative bacteria during selective decontamination of the digestive tract on an intensive care
unit. J Antimicrob Chemother 58(4):853856
53. Brun-Buisson C, Legrand P, Rauss A et al (1989) Intestinal decontamination for control of
nosocomial multiresistant gram-negative bacilli. Study of an outbreak in an intensive care
unit. Ann Intern Med 110(11):873881
54. Abecasis F, Kerr S, Sarginson RE et al (2007) Comment on: emergence of multidrug-
resistant gram-negative bacteria during selective decontamination of the digestive tract on
an intensive care unit. J Antimicrob Chemother 60(2):445
55. Hoyen CK, Pultz NJ, Paterson DL et al (2003) Effect of parenteral antibiotic administration
on establishment of intestinal colonization in mice by Klebsiella pneumoniae strains
producing extended-spectrum beta-lactamases. Antimicrob Agents Chemother
47(11):36103612
56. Martins IS, Pessoa-Silva CL, Nouer SA et al (2006) Endemic extended-spectrum beta-
lactamase-producing Klebsiella pneumoniae at an intensive care unit: risk factors for
colonization and infection. Microb Drug Resist 12(1):5058
57. Sjolin J, Goscinski G, Lundholm M et al (2000) Endotoxin release from Escherichia coli
after exposure to tobramycin: dose-dependency and reduction in cefuroxime-induced
endotoxin release. Clin Microbiol Infect 6(2):7481
58. Vanden Bossche H, Dromer F, Improvisi I et al (1998) Antifungal drug resistance in
pathogenic fungi. Med Mycol 36(Suppl 1):119128
59. Hofstra W, Vries-Hospers HG, van der WD (1979) Concentrations of nystatin in faeces after
oral administration of various doses of nystatin. Infection 7(4):166170
60. Hofstra W, de Vries-Hospers HG, van der Waaij D (1982) Concentrations of amphotericin
B in faeces and blood of healthy volunteers after the oral administration of various doses.
Infection 10(4):223227
61. Silvestri L, Milanese M, Oblach L et al (2002) Enteral vancomycin to control methicillin-
resistant Staphylococcus aureus outbreak in mechanically ventilated patients. Am J Infect
Control 30(7):391399
62. Silvestri L, van Saene HK, Zandstra DF et al (2010) Impact of selective decontamination of
the digestive tract on multiple organ dysfunction syndrome: systematic review of
randomized controlled trials. Crit Care Med 38(5):13701376
63. Silvestri L, van Saene HK, Folla L, Milanese M (2010) Selective digestive decontamination
is superior to oropharyngeal chlorhexidine in preventing pneumonia and reducing mortality
in critically ill patients. J Bras Pneumol 36(2):270272
64. Baxby D, van Saene HK, Stoutenbeek CP, Zandstra DF (1996) Selective decontamination
of the digestive tract: 13 years on, what it is and what it is not. Intensive Care Med
22(7):699706
65. Snchez Garca M, Cambronero Galache JA, Lopez DJ et al (1998) Effectiveness and cost
of selective decontamination of the digestive tract in critically ill intubated patients. A
randomized, double-blind, placebo-controlled, multicenter trial. Am J Respir Crit Care Med
158(3):908916
66. Silvestri L, van Saene HK, Casarin A et al (2008) Impact of selective decontamination of
the digestive tract on carriage and infection due to gram-negative and gram-positive
bacteria: a systematic review of randomised controlled trials. Anaesth Intensive Care
36(3):324338
67. Silvestri L, van Saene HK, Milanese M, Gregori D (2005) Impact of selective
decontamination of the digestive tract on fungal carriage and infection: systematic review
of randomized controlled trials. Intensive Care Med 31(7):898910
68. Bergmans DC, Bonten MJ, Gaillard CA et al (2001) Prevention of ventilator-associated
pneumonia by oral decontamination: a prospective, randomized, double-blind, placebo-
controlled study. Am J Respir Crit Care Med 164(3):382388
69. Korinek AM, Laisne MJ, Nicolas MH et al (1993) Selective decontamination of the
digestive tract in neurosurgical intensive care unit patients: a double-blind, randomized,
placebo-controlled study. Crit Care Med 21(10):14661473
70. Krueger WA, Lenhart FP, Neeser G et al (2002) Influence of combined intravenous and
topical antibiotic prophylaxis on the incidence of infections, organ dysfunctions, and
mortality in critically ill surgical patients: a prospective, stratified, randomized, double-
blind, placebo-controlled clinical trial. Am J Respir Crit Care Med 166(8):10291037
decreases incidence of ventilator-associated pneumonia. A randomized, placebo-controlled,
double-blind clinical trial. JAMA 265(20):27042710
72. Schardey HM, Joosten U, Finke U et al (1997) The prevention of anastomotic leakage after
total gastrectomy with local decontamination. A prospective, randomized, double-blind,
placebo-controlled multicenter trial. Ann Surg 225(2):172180
73. Silvestri L, van Saene HK, Milanese M et al (2004) Prevention of MRSA pneumonia by oral
vancomycin decontamination: a randomised trial. Eur Respir J 23(6):921926
74. Cerda E, Abella A, de La Cal MA et al (2007) Enteral vancomycin controls methicillin-
resistant Staphylococcus aureus endemicity in an intensive care burn unit: a 9-year
prospective study. Ann Surg 245(3):397407
75. de la Cal MA, Cerda E, van Saene HK et al (2004) Effectiveness and safety of enteral
medical/surgical intensive care unit. J Hosp Infect 56(3):175183
76. Sanchez Garcia M, De la Torre MA, Morales G et al (2010) Clinical outbreak of linezolid-
resistant Staphylococcus aureus in an intensive care unit. JAMA 303(22):22602264
77. Bodey GP (1980) Absorption of tobramycin after chronic oral administration. Curr Ther Res
28:394401
78. Gotoff SP, Lepper MH, Fiedler MA (1965) Treatment of salmonella carriers with colistin
sulfate. Am J Med Sci 249:399403
79. Misset B, Kitzis MD, Conscience G et al (1994) Mechanisms of failure to decontaminate the
gut with polymixin E, gentamicin and amphotericin B in patients in intensive care. Eur J
Clin Microbiol Infect Dis 13(2):165170
80. Stoutenbeek CP, van Saene HK (1990) Infection prevention in intensive care by selective
decontamination of the digestive tract. J Crit Care 5:137156
81. Krueger WA, Ruckdeschel G, Unertl K (1999) Elimination of fecal Enterobacteriaceae by

intravenous ciprofloxacin is not inhibited by concomitant sucralfatea microbiological and
pharmacokinetic study in patients. Infection 27(6):335340
82. Daneshmend TK, Warnock DW (1983) Clinical pharmacokinetics of systemic antifungal
drugs. Clin Pharmacokinet 8(1):1742
83. Winter R, Humphreys H, Pick A et al (1992) A controlled trial of selective decontamination
of the digestive tract in intensive care and its effect on nosocomial infection. J Antimicrob
84. Stoutenbeek CP, van Saene HK, Zandstra DF (1987) The effect of oral non-absorbable
antibiotics on the emergence of resistant bacteria in patients in an intensive care unit.
85. DAmico R, Pifferi S, Leonetti C et al (1998) Effectiveness of antibiotic prophylaxis in
critically ill adult patients: systematic review of randomised controlled trials. BMJ
316(7140):12751285
86. Liberati A, DAmico R, Pifferi S (2009) Antibiotic prophylaxis to reduce respiratory tract
infections and mortality in adults receiving intensive care. Cochrane Database Syst Rev
(4):CD000022
87. van Saene HK, Stoutenbeek CP, Faber-Nijholt R, van Saene JJ (1992) Selective
decontamination of the digestive tract contributes to the control of disseminated
intravascular coagulation in severe liver impairment. J Pediatr Gastroenterol Nutr
14(4):436442
88. Fink MP, Mythen MG (1999) The role of gut-derived endotoxin in the pathogenesis of
multiple organ dysfunction. In: Brade H, Opal SM, Vogel SN, Morrison DC (eds)
Endotoxin in health and disease. Dekker, New York, pp 855864
89. Cohen J (2000) The detection and interpretation of endotoxaemia. Intensive Care Med
26(Suppl 1):S51S56
90. van Saene JJ, Stoutenbeek CP, van Saene HK et al (1996) Reduction of the intestinal
endotoxin pool by three different SDD regimens in human volunteers. J Endotoxin Res
3:337343
91. Martinez-Pellus AE, Merino P, Bru M et al (1993) Can selective digestive decontamination
avoid the endotoxemia and cytokine activation promoted by cardiopulmonary bypass? Crit
Care Med 21(11):16841691
92. Martinez-Pellus AE, Merino P, Bru M et al (1997) Endogenous endotoxemia of intestinal
origin during cardiopulmonary bypass. Role of type of flow and protective effect of selective
digestive decontamination. Intensive Care Med 23(12):12511257
93. Bouter H, Schippers EF, Luelmo SA et al (2002) No effect of preoperative selective gut
decontamination on endotoxemia and cytokine activation during cardiopulmonary bypass: a
randomized, placebo-controlled study. Crit Care Med 30(1):3843
94. Bion JF, Badger I, Crosby HA et al (1994) Selective decontamination of the digestive tract
reduces gram-negative pulmonary colonization but not systemic endotoxemia in patients
undergoing elective liver transplantation. Crit Care Med 22(1):4049
95. Maring JK, Zwaveling JH, Klompmaker IJ et al (2002) Selective bowel decontamination in
elective liver transplantation: no improvement in endotoxaemia, initial graft function and
post-operative morbidity. Transpl Int 15(7):329334
96. Oudemans-van Straaten HM, van Saene HK, Zandstra DF (2003) Selective decontamination
of the digestive tract: use of the correct antibiotics is crucial. Crit Care Med 31(1):334335
97. Schippers EF, van Dissel JT (2003) Selective gut decontamination. Crit Care Med
31(11):27152716
98. van Saene HK, Silvestri L, Bams JL et al (2003) Selective decontamination of the digestive
tract: use in liver transplantation is evidence based. Crit Care Med 31(5):16001601
99. Hammond JM, Potgieter PD (1995) Long-term effects of selective decontamination on
antimicrobial resistance. Crit Care Med 23(4):637645
100. Leone M, Albanese J, Antonini F et al (2003) Long-term (6-year) effect of selective

digestive decontamination on antimicrobial resistance in intensive care, multiple-trauma
patients. Crit Care Med 31(8):20902095
101. Tetteroo GW, Wagenvoort JH, Bruining HA (1994) Bacteriology of selective
decontamination: efficacy and rebound colonization. J Antimicrob Chemother 34(1):
139148
102. OchoaArdila ME et al (2011) Long-term use of selective decontamination of the digestive
tract does not increase antibiotic resistance: a 5-year prospective cohort study. Intensive
Care Med 37:1458
103. Saunders GL, Hammond JM, Potgieter PD et al (1994) Microbiological surveillance during
selective decontamination of the digestive tract (SDD). J Antimicrob Chemother 34(4):
529544
104. de Smet AM, Hopmans TE, Minderhoud AL et al (2009) Decontamination of the digestive
tract and oropharynx: hospital acquired infections after discharge from the intensive care
unit. Intensive Care Med 35(9):16091613
105. Oostdijk EA, de Smet AM, Blok HE et al (2010) Ecological effects of selective
decontamination on resistant gram-negative bacterial colonization. Am J Respir Crit Care
Med 181(5):452457
106. Bonten MJ, Gaillard CA, Johanson WG Jr et al (1994) Colonization in patients receiving
and not receiving topical antimicrobial prophylaxis. Am J Respir Crit Care Med 150(5 Pt
1):13321340
107. Gastinne H, Wolff M, Delatour F et al (1992) A controlled trial in intensive care units of
selective decontamination of the digestive tract with nonabsorbable antibiotics. The french
study group on selective decontamination of the digestive tract. N Engl J Med 326(9):594
599
108. Hammond JM, Potgieter PD, Saunders GL, Forder AA (1992) Double-blind study of
selective decontamination of the digestive tract in intensive care. Lancet 340(8810):
59
109. Ferrer M, Torres A, Gonzalez J et al (1994) Utility of selective digestive decontamination in
mechanically ventilated patients. Ann Intern Med 120(5):389395
110. Wiener J, Itokazu G, Nathan C et al (1995) A randomized, double-blind, placebo-controlled
trial of selective digestive decontamination in a medical-surgical intensive care unit. Clin
Infect Dis 20(4):861867
111. Lingnau W, Berger J, Javorsky F et al (1997) Selective intestinal decontamination in
multiple trauma patients: prospective, controlled trial. J Trauma 42(4):687694
112. Verwaest C, Verhaegen J, Ferdinande P et al (1997) Randomized, controlled trial of
selective digestive decontamination in 600 mechanically ventilated patients in a
multidisciplinary intensive care unit. Crit Care Med 25(1):6371
113. Snchez M, Mir N, Canton R et al (1997) The effect of topical vancomycin on acquisition,
carriage and infection with methicillin-resistant Staphylococcus aureus in critically ill
patients (abstract). 37th Interscience Conference on Antimicrobial Agents and
Chemotherapy (ICAAC) (28 Sept1 Oct) Toronto.
114. van Saene HK, Weir WI, de La Cal MA et al (2004) MRSAtime for a more pragmatic
approach? J Hosp Infect 56(3):170174
115. White A (1961) Relation between quantitative nasal cultures and dissemination of
staphylococci. J Lab Clin Med 58:273277
116. Brodie J, Kerr MR, Sommerville T (1956) The hospital Staphylococcus; a comparison of
nasal and faecal carrier states. Lancet 270(6906):1920
117. Greendyke RM, Constantine HP, Agruder GB, Dean DC, Gardner JH, Morgan HR et al
(1958) Staphylococci on a medical ward, with special reference to fecal carriers. Am J Clin
Pathol 30(4):318322
118. Hone R, Keane CT, Fitzpatrick S (1974) Faecal carriage of Staphylococcus aureus in
infantile enteritis due to enteropathic Escherichia coli. Scand J Infect Dis 6(4):329332
119. Brady LM, Thomson M, Palmer MA, Harkness JL (1990) Successful control of endemic
MRSA in a cardiothoracic surgical unit. Med J Aust 152(5):240245
120. Salzman TC, Clark JJ, Klemm L (1967) Hand contamination of personnel as a mechanism
of cross-infection in nosocomial infections with antibiotic-resistant Escherichia coli and
Klebsiella-Aerobacter. Antimicrobial.Agents Chemother 7:97100
121. Nystrom B (1983) Optimal design/personnel for control of intensive care unit infection.
Infect Control 4(5):388390
122. Crossley K, Landesman B, Zaske D (1979) An outbreak of infections caused by strains of
Staphylococcus aureus resistant to methicillin and aminoglycosides. II. Epidemiologic
studies. J Infect Dis 139(3):280287
123. Viviani M, van Saene HK, Dezzoni R et al (2005) Control of imported and acquired
methicillin-resistant Staphylococcus aureus (MRSA) in mechanically ventilated patients: a
dose-response study of enteral vancomycin to reduce absolute carriage and infection.
Anaesth Intensive Care 33(3):361372
124. Taylor ME, Oppenheim BA (1991) Selective decontamination of the gastrointestinal tract as
an infection control measure. J Hosp Infect 17(4):271278
125. Damjanovic V, Connolly CM, van Saene HK et al (1993) Selective decontamination with
nystatin for control of a Candida outbreak in a neonatal intensive care unit. J Hosp Infect
24(4):245259
126. Rice LB (2001) Emergence of vancomycin-resistant enterococci. Emerg Infect Dis
7(2):183187
127. Snchez M, Mir N, Cantn R et al (1997) Incidence of carriage and colonization pattern of
vancomycin-resistant enterococcus (VRE) in intubated patients (Pts) receiving topical
vancomycin (V) (abstract). 37th Interscience Conference on Antimicrobial Agents and
Chemotherapy (ICAAC)
128 Stiefel U, Paterson DL, Pultz NJ et al (2004) Effect of the increasing use of piperacillin/
tazobactam on the incidence of vancomycin-resistant enterococci in four academic medical
centers. Infect Control Hosp Epidemiol 25(5):380383
129. Salgado CD, Giannetta ET, Farr BM (2004) Failure to develop vancomycin-resistant
Enterococcus with oral vancomycin treatment of Clostridium difficile. Infect Control Hosp
Epidemiol 25(5):413417
Part III
Infection Control
Evidence-Based Infection Control
in the Intensive Care Unit 9
J. Hughes and R. P. Cooke
9.1 Introduction
The rate of healthcare-associated infections (HCAIs) is higher in the intensive care

unit (ICU) patients than in the general hospital population, mainly due to its unique
environment where the most severely ill patients are brought together in one
unit [1]. This increases patients susceptibility to HCAIs, particularly with multi-
drug-resistant pathogenic microorganisms (MDRPMs). Risk factors for HCAIs
include antimicrobials, and invasive devices such as peripheral and central
intravenous catheters, urinary catheters, endotracheal tubes, tracheostomy, and
surgical and chest drains [2]. Due to the amount of patient-care contact in the ICU
environment, the risk of HCAIs is further increased. With the often urgent nature
of ICU interventions, suboptimal compliance with infection prevention and control
(IPC) practices can occur [3].
As a result of the high priority given to quality management, clinical gover-
nance and patient safety, HCAIs rates are increasingly in the political and public
spotlight. In the UK, trusts are held accountable by the 2008 Health Act. Code of
Practice for the Prevention and Control of Health Care Associated Infections [4].
In addition, from 2009, trusts had to register with the Care Quality Commission, a
new regulatory body that has enforcement powers against noncompliant organi-
sations [5].
To achieve sustainable reductions in HCAIs in the ICU, the support, commit-
ment and engagement of all healthcare workers (HCWs), both clinical and
nonclinical, are essential [6]. Though no single action on its own will guarantee
J. Hughes (&)
Infection Prevention and Control, 5 Boroughs Partnership NHS
FoundationTrust/University of Chester Warrington,
Winwick, Warrington, UK
146 J. Hughes and R. P. Cooke
Table 9.1 Classification of the US Centre for Disease Control guidelines
Category Application
IA Strongly recommended for implementation and then supported by
well-designed experimental, clinical or epidemiological studies
IB Strongly recommended for implementation and then supported by
certain experimental, clinical or epidemiological studies and a strong
theoretical rationale
IC Required for implementation, as mandated by federal and/or state
regulation or standard
II Suggested for implementation and then supported by suggestive
clinical or epidemiological studies or a theoretical rationale
No recommendation/ Practices for which insufficient evidence or no consensus regarding
unresolved issue efficacy exist
effective results, adherence to IPC practices is one of the best means of minimising
HCAIs [7].
Universally accepted standards for IPC include hand hygiene, use of personal
protective equipment (PPE), patient isolation, care of equipment and environ-
mental cleanliness [7, 8]. The role of invasive devices, antibiotics and surveil-
lance-related policies are discussed in Chaps. 10 and 11. Although there are a
number of studies on the benefits of such IPC measures, the evidence base is often
questionable or lacking [9]. However, several studies and reviews have tried to
address this issue [1012]. In 2010, the European Centre for Disease Prevention
and Control (ECDC) convened a group of experts to discuss patient safety and
infection prevention and control to formulate risk-based strategies and evidence-
based measures for effective HCAI reduction [13].
9.2 Evidence-Based Practice in the ICU
Evidence-based practice (EBP) should enable HCWs to be confident that their

interventions are informed by a sound knowledge base. Clinical care guidelines
provide key principles for good practice and ensure that practices can be stand-
ardised and auditable [914]. Therefore, the following review and recommenda-
tions are based on guidance provided by the US Centre for Disease Control (CDC)
(Table 9.1), the UK Department of Health Epic 2: National Evidence-Based
Guidelines for Preventing Healthcare-Associated Infections in Hospitals and the
World Health Organisation (WHO) [11, 12, 15]. Although some of the evidence
for prevention and control is based on studies in the general hospital population
(Table 9.2), many interventions to prevent and control HCAIs, in particular,
methicillin-resistant Staphylococcus aureus (MRSA), were pioneered initially in
the ICU, being a reflection of the magnitude of the problem across the entire
patient population [16]. Therefore, adherence to infection prevention and control
EBP will affect not only the ICU but the rest of the healthcare facility (Table 9.3).
9 Evidence-Based Infection Control in the Intensive Care Unit 147
Table 9.2 Recommendations for hand hygiene (HCWs healthcare workers; classification of the
US Centre for Disease Control guidelines)
Recommendation Level of
evidence
When hands are visibly dirty or contaminated with proteinaceous material or are IA
visibly soiled with blood or other body fluids, wash hands with water and either
a nonantimicrobial or an antimicrobial soap
If hands are not visibly soiled, use an alcohol-based hand rub for routinely IC
decontaminating hands in all other clinical situations described
Decontaminate hands after contact with a patients intact skin (e.g. when taking IB
a pulse or blood pressure and lifting)
Decontaminate hands after contact with body fluids or excretions, mucous IA
membranes, nonintact skin and wound dressings if hands are not visibly soiled
Decontaminate hands after removing gloves IB
Before eating and after using a restroom, wash hands with water and either a IB
nonantimicrobial or antimicrobial soap
No recommendation can be made regarding the routine use of non-alcohol-based Unresolved
hand rubs for hand hygiene in healthcare settings issue
When decontaminating hands with an alcohol-based hand rub, apply product to IB
palm of one hand and rub hands together, covering all surfaces of hands and
fingers, until hands are dry
When washing hands with soap and water, wet hands first with water, apply an IB
amount of product recommended by the manufacturer to hands, and rub hands
together vigorously for at least 15 s, covering all surfaces of the hands and
fingers. Rinse hands with water and dry thoroughly with a disposable towel.
Use paper towel to turn off the faucet
Provide personnel with efficacious hand-hygiene products that have low irritancy IB
potential, particularly when these products are used many times per shift. This
recommendation applies to products used for hand antisepsis before and after
patient care in clinical areas and to products used for surgical hand antisepsis by
surgical personnel
Do not add soap to a partially empty soap dispenser. This practice of topping IA
up dispensers can lead to a bacterial contamination of the soap
Provide HCWs with hand lotions or creams to minimise the occurrence of IA
irritant contact dermatitis associated with hand antisepsis or handwashing
As part of a multidisciplinary programme to improve hand-hygiene adherence, IA
provide HCWs with a readily accessible alcohol-based hand-rub product
To improve hand-hygiene adherence among personnel who work in areas in IA
which high workloads and high intensity of patient care are anticipated, make an
alcohol-based hand rub available at the entrance to all patient rooms or at the
bedside, in other convenient locations and in individual pocket-sized containers
to be carried by HCWs
HCW healthcare workers
Table 9.3 Recommendations for protective clothing and care of equipment and environment
Recommendation Level of
evidence
Select protective equipment on the basis of a risk assessment of microorganism II
transmission
Gloves Wear gloves (clean, nonsterile gloves are adequate) when touching IB
blood, body fluids, secretions, excretions and contaminated items. Put on clean
sterile gloves just before touching mucous membranes and nonintact skin.
Change gloves between tasks and procedures on the same patient after contact
with material that may contain a high concentration of microorganisms. Remove
gloves promptly after use, before touching noncontaminated items and
environmental surfaces and before going to another patient, and decontaminate
hands immediately to avoid transfer of microorganisms to other patients or
environments
Face and eye protection Wear a mask or a face shield to protect mucous IB
membranes of the eyes, nose and mouth during procedures and patient-care
activities that are likely to generate splashes or sprays of blood, body fluids,
secretions or excretions
Gown Wear a gown (a clean, nonsterile gown is adequate) to protect skin and IB
prevent soiling of clothing during procedures and patient-care activities that are
likely to generate splashes or sprays of blood, body fluids, secretions or
excretions. Select a gown that is appropriate for the activity and amount of fluid
likely to be encountered. Remove a soiled gown as promptly as possible, and
wash hands to avoid transfer of microorganisms to other patients or
environments
Isolation Place in a private room a patient who contaminates the environment or IB
who does not (or cannot be expected to) assist in maintaining appropriate
hygiene or environmental control. If a private room is not available, consult with
infection-control professionals regarding patient placement or other alternatives
Patient-care equipment Handle used equipment soiled with blood, body fluids, IB
secretions or excretions in a manner that prevents skin and mucous-membrane
exposure, clothing contamination and microorganism transfer to other patients
and environments. Ensure that reusable equipment is not used for the care of
another patient until it has been cleaned and reprocessed appropriately. Ensure
that single-use items are discarded properly
Environmental control Ensure that the hospital has adequate procedures for the IB
routine care, cleaning and disinfecting of environmental surfaces, beds, bedrails,
bedside equipment and other frequently touched surfaces, and ensure that these
procedures are being followed. Handle, transport, and process used linen soiled
with blood, body fluids, secretions or excretions in a manner that prevents skin
and mucous-membrane exposure and clothing contamination and that avoids
microorganism transfer to other patients and environments
9.3 Five Main IPC Manoeuvres
9.3.1 Hand Hygiene
Hand washing is often referred to as the single most important means of preventing
HCAIs [11, 12, 15, 16]. The UK epic2: Guidelines for Preventing Healthcare
Associated Infections in NHS Hospitals in England suggest that effective hand
decontamination can significantly reduce infection rates in high-risk areas such as
the ICU [11]. However, there is no supporting level 1 evidence, (i.e. randomised
controlled trials), as ethical approval for such studies would be not possible. The
available evidence is based on expert consensus opinion and several observational
epidemiological studies [11, 12, 15]. WHO guidelines on hand hygiene identify
Five Moments for Hand Hygiene [12]. Hands must be decontaminated and dried
thoroughly immediately before each direct patient contact/care episode and after
any activity or contact that can result in hands becoming contaminated, such as
contact with patient-care equipment and their immediate environment.
Many guidelines and studies review a variety of differing products for hand
decontamination, with some studies suggesting that soap and water is as effective
as antimicrobial-based hand-washing products [11]. Alcohol-based hand rubs
(ABHRs) are highly popular in ICUs but alone are not effective in removing
physical dirt of soiling, with some studies suggesting that they are not deemed as
effective in removing spore-forming bacteria such as Clostridium difficile,
although other studies repute this [17]. Some studies have also shown that
once alcohol gel has evaporated, a residue is left that can be deposited in the
environment and which may facilitate growth of organisms such as Acinetobacter
species [18]. To combat these problems, newer gels with an aloe vera base and
containing copper-based biocidal formulations are being trialled that may also be
less irritating to HCWs hands [19].
Most studies fail to find compelling evidence for the general use of one hand
contaminant agent over another. However, all infer that HCWs acceptability and
effective hand hygiene techniques and adequate drying are the most essential
factors when selecting products to promote compliance with hand washing.
In Europe, there are a set of standards, referred to as European Norms (EN), which
are laboratory tests that any product needs to pass before it can be marketed.
Therefore, any product chosen should comply with EN standards [20].
Hand decontamination is often poorly performed, and several studies demon-
strate that compliance can be suboptimal [11, 12, 21], particularly by some phy-
sicians. The various reasons for this include availability of hand-decontamination
facilities, selection of harsh hand-care products, workload and HCW attitudes.
ICUs should always have enough easily accessible hand-wash basins, the rec-
ommended ratio being 1 to 1 per ICU bed in the UK. The increased availability of
ABHRs has also been found to improve compliance [12], as have regular hand-
decontamination training sessions/audits and feedback to all ICU staff [11, 12, 21].
Reinforcing the importance of physicians, in particular, as positive role models to
influence behaviour could also help, as they are often the main leads for ICU.
Hand-decontamination policies should also ensure that no jewellery, including
watches and rings (other than a plain wedding band) are worn, nails are kept short
and acrylic/false fingernails and nail polish are prohibited [11, 12]. Clinical staff
should wear short sleeves (or long sleeves rolled up to above the elbow) when
performing patient-care procedures. In the UK, this is referred to as the principle of
bare below the elbow [22]. However, although hand hygiene is the cornerstone
of IPC, it is nonetheless not a stand-alone procedure and should be part of a
multifaceted approach [23].
9.4 Personal Protective Equipment (PPE)
PPE includes the use of aprons, gowns, gloves, eye protection and face masks and
should be easily accessible to all HCWs. Employers have a duty to provide PPE
for staff who in turn have to wear PPE to comply with health and safety guidelines,
such as the Personal Protective Equipment Regulations (1992), which are com-
pulsory in some countries, e.g. the UK [24]. However, wearing PPE should always
be based on a patient- and task-risk assessment.
Gloves are one of the most effective barriers against microorganisms, although
studies have shown both lack of understanding and poor compliance with glove use
[11, 15, 25]. This includes gloves being worn when not required or worn for prolonged
periods with hands not being washed following glove removal. Gloves can also cause
adverse skin reactions and skin sensitivity [11, 25]. Sterile gloves should be worn for all
invasive procedures and contact with nonintact skin. Nonsterile gloves are suitable for
all other procedures. Gloves should also be single-use items and discarded immedi-
ately after each activity, followed by hand decontamination [25].
Disposable plastic aprons are just as effective as gowns in most situations.
However, where there is a possibility of contamination of the HCWs clothing with
blood or body fluids, it is recommended that disposable plastic aprons are worn.
Gowns are only required where there is a risk of gross contamination of splashing,
such as in major burn patients or severe trauma, or when dealing with biohazard
group 4 pathogens (e.g. Lassa fever) [11, 15, 26]. HCWs should also have access to
facial and eye protection. Personal respiratory protection is required when caring for
patients with serious infections, such as tuberculosis pandemic flu when aerosol-
generating procedures are performed [26, 27]. The choice of mask/respirator will
depend on the level and extent of protection required. Eye protection and visors
should be worn when there is a risk of body-fluid contamination to the face or eyes.
9.5 Isolation
In addition to standard IPC precautions, further procedures (i.e. side-room isola-

tion) should be implemented if a patient is identified as carrying a highly trans-
missible microbe. Several studies have demonstrated that isolation precautions are
effective in achieving a significant reduction in transmission, particularly in

relation to glycopeptide-resistant enterococci, C. difficile, respiratory syncytial
virus (RSV), MRSA, chickenpox and Acinetobacter [11, 15, 16, 2729]. However,
some studies have disputed the role of isolation in preventing MRSA transmission
[30]. Nonetheless, current guidance suggest isolating or cohorting most patients
with transmissible infections wherever possible. Furthermore, these precautions
should be commenced on suspicion without awaiting microbiological confirmation
[11, 15, 3133]. Isolating patients may also be difficult due to the lack of single
rooms. In this event, a risk assessment with the infection prevention and control
team (IPCT) should be undertaken, and procedures should be simple to implement
and follow [34].
9.6 Patient-Care Equipment
Patient-care equipment such as ventilators, humidifiers, analysers and transducers

have all been implicated in ICU and anaesthesia-associated outbreaks [1, 21, 35],
particularly where equipment is shared between patients. Common procedures and
device-related infections are discussed within Chap. 11.
9.6.1 High-Risk Items
High-risk items include invasive devices in contact with a break in the skin or with
mucous membranes, or those introduced into a sterile body area, such as surgical
instruments, catheters, prosthetic devices etc. These items require sterilisation, but
if that is not practically achievable, e.g. in the case of flexible endoscopes, then
high-level disinfection.
9.6.2 Intermediate Risk
These include items in contact with intact mucous membranes or body fluids,
are contaminated with particularly virulent or readily transmissible organisms, or
items to be used on highly susceptible patients or sites, e.g. endoscopes and
respiratory equipment require disinfection.
9.6.3 Low-Risk Items
These items are those in contact with normal or intact skin, e.g. thermometers,
stethoscopes, washbowls, toilets and bedding. Cleaning and drying is usually
sufficient, but disinfecting is required in the case of a patient with a known
infection risk, e.g. MRSA.
9.6.4 Minimal-Risk Items
Minimal-risk items are items that do not come into contact with the patient, such
as floors and surfaces. Cleaning and drying is usually adequate unless in an
outbreak situation. Equipment used for invasive procedures should be decontam-
inated after being used on every patient, not only on those known to be infected.
Equipment such as stethoscopes and thermometers should be designated to
individual patients wherever possible. When dedicated equipment is not practical,
e.g. portable X-ray and ultrasound machines, staff should ensure that such
equipment is decontaminated between patients. Policies should ensure that all staff
receive training and are aware of their roles and responsibilities in relation to
decontamination. The policy should also ensure close liaison with procurement
officers and the IPC team when purchasing any new equipment.
9.7 Patient-Care Environment
Although the ICU patients surrounding is classed as low risk, a clean environment
provides the background for good standards of hygiene as well as maintaining
patient, staff and visitor confidence [1]. In the UK, the ICU environment is cate-
gorised as a high-risk area under the Standards of Environmental Cleanliness in
Hospitals Guidelines [36]. When considering the role of the environment in the
ICU, the design of the unit should ensure that there is adequate space between beds
to allow easy access for staff and equipment [37, 38]. Furniture and fixtures should
be kept to a minimum and be of materials that are easy to clean. A suitable area
should also be designated for safe storage of equipment. Dirty and clean
areas should be separate so that no mixing of equipment occurs, and clinical waste
should be stored where there is no risk of transporting waste through clean or
patient areas. Despite the pressure for ICU beds, ICU mangers should maintain
close liaison with domestic services staff to maintain standards.
9.8 Audit and Surveillance
As stated the IPC, manoeuvres referred to should be used in conjunction and are
rarely stand alone. Therefore, it is often a challenge to gauge exactly which
manoeuvres reviewed reduce HCAIs, and the evidence should be viewed in light
of this. However, to ensure compliance with IPC and monitor HCAIs, it is essential
that there are effective audit and surveillance systems in place to monitor HCAIs
[68, 39, 40].
The Saving Lives High Impact Interventions (HIIs), launched in the UK in
2005, are practical EBP tools to audit practice and compliance and help reduce
HCAI [68]. These are based on a care-bundle approach and have been imple-
mented by many ICUs in the UK. There are several available, e.g. care of invasive
devices such as peripheral, central venous and urinary catheters, reducing
surgical-site infections, reducing the risk of C. difficile and cleaning and decon-
tamination of the environment.
In relation to surveillance the rapid rise in the rate of bacterial antibiotic
resistance in ICU patients is well recognised [41]. Four key organism groups
warrant regular surveillance in the ICU setting, as highlighted by the National
Nosocomial Infections Surveillance (NNIS) system report in MDRPMs in US
ICUs [42]. These include MRSA, vancomycin-resistant Enterococcus faecium
(VRE), Pseudomonas aeruginosa resistant to carbapenems or fluoroquinolones
and Enterobacteriaceae resistant to third-generation cephalosporins [including
production of extended-spectrum beta-lactamases (ESBLs)]. ICU screening pro-
tocols for MRSA and VRE are well established through infection control;
guidelines for ESBL surveillance are lacking.
Many factors contribute to the high incidence of nosocomial infections in the ICU
and associated poor patient outcomes. Though most studies have come from
industrialised countries, the rates of infection in developing countries may be higher
[43]. This relates particularly to the use of invasive devices, the use of which should
be linked to an ICU audit programme. Such a programme should look at not only
infection rates but mortality rates, length of ICU stay and hospital costs.
Indwelling devices, such as central venous catheters (VCs), Foley catheters and
endotracheal tubes, bypass natural host defence mechanisms and predispose to
infection. The Surveillance and Control of Pathogens of Epidemiological Impor-
tance (SCOPE) database found that 51% of all catheter-associated blood-stream
infections (CR-BSIs) in 49 US hospitals occurred in the ICU. CR-BSI rates per
1,000 catheter days for a variety of vascular devices were listed in the report [44].
A recent systematic review indicated that CR-BSIs per 1,000 catheter days range
from 0.5 for peripheral VCs to 1.6 for long-term central VCs, 1.7 for arterial
catheters, 2.1 for peripherally inserted central VCs and 2.7 for short-term central
VCs [45]. EPIC guidelines recommend using antimicrobial-impregnated central
VC if short-term CR-BSI rates are high despite introducing a comprehensive
strategy to reduce rates [11].
The highest risk for nosocomial pneumonia is in patients on mechanical venti-
lation [i.e. ventilator-associated pneumonia (VAP)], which has been the most studied
form of nosocomial pneumonia. ICUs should therefore identify cases of VAP and
collect data on the number of ventilator-associated days on a routine basis.
Catheter-associated urinary tract infection (UTI) is overall the most common
nosocomial infection. In the ICU setting, auditing infection per 1,000 catheter days
will allow a rational approach for using urinary catheters impregnated with either
antimicrobial or antiseptic agents.
9.9 Use of the Microbiology Laboratory by the ICU
The purpose of a microbiology laboratory is to provide an accurate and clinically

relevant service in a timely manner. In the context of ICU, this must be available
7 days per week. Whenever possible, every effort should be made to obtain
specimens prior to starting antibiotic therapy, especially by taking two indepen-

dent blood cultures from septic patients.
Prompt transport to the laboratory is equally essential. Once local criteria are
agreed upon, this can usually be by means of a pneumatic tube system directly to
the microbiology laboratory. The ICU must also be supported by good information
technology access to microbiology reports. This should be complemented by daily
ward rounds involving a consultant in microbiology or infectious diseases.
In certain circumstances, point-of-care microbiology testing in the ICU may
offer a more rapid response service. For MRSA screening, this may have a number
of advantages, but it must be offset against local MRSA prevalence, standard
laboratory turnaround times and equipment/staff consumable cost [46].
Selective decontamination of the digestive tract (SDD) is an antimicrobial
strategy that reduces the incidence of severe infections of lower airways and
blood by eradicating throat and gut carriage of normal and abnormal flora. Sixty
randomised controlled trials (RCTs) and 11 meta-analyses demonstrated that SDD
confers protection against severe infections of lower airways and blood [47].
Despite the optimism regarding SDD, it has not been used widely in the ICU
setting, principally due to anxiety in the emergence of antibiotic-resistant bacteria
[48]. Hence, in units that wish to perform SDD, a bacteriological protocol for
gut-surveillance cultures as opposed to diagnostic cultures is recommended. SDD
is discussed in Chap. 13.
9.10 Summary and Conclusions
In this chapter, recommendations for EBP IPC practice and strategies have been
made following review of the literature. However, to ensure best practice, ICUs
should also be adequately staffed, as many studies demonstrate the detrimental effect
of poor staffing-to-patient ratios and increased workload [49, 50]. Good IPC practices
in the ICU rely on strong medical and nursing leadership plus good multidisciplinary
working practices. It also needs to be recognised that IPC in the ICU patient must be
practised slightly differently from the non-ICU patient. Hence, specific IPC policies
and procedures for the ICU should be drawn up. These should be underpinned by
audit and surveillance to demonstrate that they are working effectively, and they
should be reviewed by the multidisciplinary group on a regular basis.
References
1. Cooper M (2009) Prevention of infection in special wards and departmentscritical care
units. In: Fraise AP, Bradley C (eds) Ayliffes control of healthcare-associated infection, 5th
edn. Hodder Arnold, Kent
2. Fraise AP, Bradley C (eds) (2009) Ayliffes control of healthcare-associated infection, 5th
edn. Hodder Arnold, Kent UK
3. Department of Health (2006) Infection prevention and control in adult critical care reducing
the risk of infection through best practice. Department of Health, London
4. Department of Health (2009) The health and social care act 2008: code of practice for the
NHS on the prevention and control of health care associated infections and related guidance.
Department of Health, London
5. Care Quality Commission (2008) Registering with the Care Quality Commission in relation
to healthcare associated infection. Guidance for trust 2009/10. Care Quality Commission,
London
6. Department of Health (2006) Saving lives: a delivery programme to reduce healthcare
associated infection, including MRSA screening for methicillin-resistant Staphylococcus
aureus (MRSA) colonisation: a strategy for NHS trusts: a summary of best practice.
7. Department of Health (2006) Going further faster: implementing the saving lives delivery
programme sustainable change for cleaner safer care. Department of Health, London
8. Department of Health (2008) Clean safe care reducing infections and saving lives.
9. Ayliffe GJ (2000) Evidence-based practises in infection control. Br J Infect Control 1(4):59
10. NICE (2004) Guideline development methods: information for national collaborating centres
and guideline developers. National Institute for Health and Clinical Excellence (updated 2005)
11. Pratt RP, Pellowe CM, Wilson JA et al (2007) epic2: national evidence-based guidelines for
preventing healthcare-associated infections in hospitals in England. J Hosp Infect 65S:S1S64
12. World Health Organization (2009) Patient safety a world alliance for safer health care. WHO
guidelines on hand hygiene in health care. First global patient safety challenge clean care is
safer care. WHO, Geneva
13. European centre for disease prevention and control (2010) meeting report: expert consultation
on healthcare-associated infection prevention and control. ECDC Stockholm
14. Harbour R, Miller JA (2001) A new system for grading recommendations in evidence-based
guidelines. BMJ 323:334336
15. Siegel JD, Rhineheart E, Jackson M, Chiarello L (2007) Guideline for isolation precautions:
preventing transmission of infectious agents in healthcare settings. The Healthcare Infection
Control Practices Advisory Committee. Available at http://www.cdc.gov/ncidod/dhqp/pdf/
guidelines/isolation2007.pdf. Accessed Jun 2011
16. Humphreys H (2008) Can we do better in controlling and preventing methicillin-resistant
Staphylococcus aureus (MRSA) in the intensive care unit (ICU)? Eur J Clin Micro Infect Dis
27(6):409413
17. Boyce JM, Ligi C, Kohan C et al (2006) Lack of association between the increased incidence
of Clostridium difficile-associated disease and the increasing use of alcohol-based hand rubs.
Infect Control Hosp Epidemiol 27(5):479483
18. Edwards J, Patel G, Wareham DW (2007) Low concentrations of commercial alcohol hand
rubs facilitate the growth of and secretion of extracellular proteins by multidrug-resistant
strains of Acinetobacter baumannii. J Med Micro 56:15951599
19. Hall TJ, Wren MWD, Wareham DW et al (2010) Effect of the dried residues of two hand gels
on the survival of methicillinresistant Staphylococcus aureus and Acinetobacter
calcoaceticus-baumannii. J Inf Prev 11(3):7072
20. Fraise A, Bradley C (2007) Decontamination of equipment, the environment and the skin. In:
Fraise AP, Bradley C (eds) Ayliffes control of healthcare-associated infection, 5th edn.
Hodder Arnold, Kent
21. Pittet D, Simon A, Hugonnet S (2004) Hand hygiene among physicians: performance, beliefs,
and perceptions. Ann Intern Med 141:1118
22. Department of Health (2010) Uniform and workwear: guidance on uniform and workwear
policies for NHS employers. Department of Health, London
23. Silvestri L, Petros AJ, Sarginson RE et al (2005) Review: handwashing in the intensive care
unit: a big measure with modest effects. J Hosp Inf 59:172179
24. Health and Safety Executive (1992) The personal protective equipment at work regulations
guidance on regulations. The Stationery Office, London
25. Infection Control Nurses Association (2002) Protective clothing: principles and guidance.
Fitwise, Bathgate
26. Ho PL, Tang XP, Seto WH (2003) SARS: hospital infection control and admission strategies.
Respirology 8:S41S45
27. Humphreys H (2007) Control and prevention of healthcare-associated tuberculosis: the role
of respiratory isolation and personal respiratory protection. J Hosp Inf 66(1):15
28. Trautmann M, Pollitt A, Loh U et al (2007) Implementation of an intensified infection control
program to reduce MRSA transmission in a German tertiary care hospital. Am J Infect
Control 35(10):643649
29. Raineri E, Crema L, De Sivestri A et al (2007) Methicillin-resistant Staphylococcus aureus in
an intensive care unit: a 10 year analysis. J Hosp Infect 67(4):308315
30. Cepeda JA, Whitehouse T, Cooper B et al (2005) Isolation of patients in single rooms or
cohorts to reduce spread of MRSA in intensive-care units: prospective two-centre study.
Lancet 36(945):295304
31. Farr BM, Bellingan G (2004) Pro/con clinical debate: isolation precautions for all intensive
care unit patients with methicillin-resistant Staphylococcus aureus colonisation are essential.
Crit Care 8(3):153156
32. Groothuis J, Bauman J, Malinoski F, Eggleston M (2008) Strategies for prevention of RSV
nosocomial infection. J Perinatol 28(5):319323
33. Kappstein I, van der Muhlen K, Meschzan D et al (2009) Prevention of transmission of
methicillin-resistant Staphylococcus aureus (MRSA) infection: standard precautions instead
of isolation: a 6-year surveillance in a university hospital. Chirurg 80(1):4961
34. Kilpatrick C, Prieto J, Wigglesworth N (2008) Single room isolation to prevent transmission
of infection: development of a patient journey tool to support safe practice. Br J Inf Control
9(6):1925
35. King TA, Cooke RPD (2005) Cleaning, disinfection and sterilisation. In: Darcy AJ, Diba A
(eds) Wards anaesthetic equipment, 5th edn. Elsevier, London
36. National Patient Safety Agency (2007) The national specifications for cleanliness in the
NHS, London. NHS/NPSA. Available at www.npsa.nhs.uk/health/currentprojects/nutrition/
cleaning. Accessed Jun 2011
37. Dettenkofer M, Seegers S, Antes G et al (2004) Does the architecture of hospital facilities
influence infection rates? A systematic review. Infect Control Hosp Epidemiol 25(1):2125
38. Estates NHS (2005) Health building note 57: facilities for critical care. The Stationery Office,
London
39. McGingle KL, Gourlay ML, Buchanan IB (2008) The use of active surveillance cultures in
adult intensive care units to reduce methicillin-resistant Staphylococcus aureus-related
morbidity, mortality, and costs: a systematic review. Clin Infect Dis 46(11):17171725
40. Aragon D, Sole ML (2006) Implementing best practice strategies to prevent infection in the
ICU. Crit Care Clin N Am 18(4):441452
41. Ahern JW, Alston WK (2009) Use of longitudinal surveillance data to assess the effectiveness
of infection control in critical care. Infect Control Hosp Epidemiol 30(11):11091112
42. National Nosocomial Infections Surveillance (2000) NNIS system report, data summary from
January 1992April. Am J Infect Control 28:429
43. Depuydt PO, Blot SI, Benoit DD et al (2006) Antimicrobial resistance in nosocomial
bloodstream infection associated with pneumonia and the value of systematic surveillance
cultures in an adult intensive care unit. Crit Care Med 34(3):653659
44. Wisplinghoff H, Bischoff T, Tallent SM (2004) Nosocomial bloodstream a prospective
nationwide surveillance study. Clin Infect Dis 39:309317
45. Maki DG, Kluger DM, Crivich CJ (2006) The risk of bloodstream infection in adults with
different intravascular devices: a systematic review of 200 published prospective studies.
Mayo Clin Pro 23006:11591171
46. Cunningham R, Jenks B, Northwood J (2007) Effect of MRSA transmission of rapid PCR
testing of patients admitted to critical care. J Hosp Infect 65:2438
47. Silvestri L, van Saene HK, Casarin A (2008) Impact of selective decontamination of the
digestive tract on carriage and infection due to Gram-negative and Gram-positive bacteria: a
systematic review of randomised controlled trials. Anaesth Intensive Care 36:324338
48. Oostdijk EA, de Smet AM, Blok HE et al (2010) Ecological effects of selective
decontamination on resistant Gram-negative bacterial colonization. Am J Respir Crit Care
Med 181(5):452457
49. Stone PW, Mooney-Kane C, Larson EL (2007) Nurse working conditions and patient safety
outcomes. Med Care 45(6):571578
50. Hugonnet S, Chevrolet JC, Pittet D (2007) The effect of workload on infections in critically
ill patients. Crit Care Med 35(11):7681
Device Policies
10
A. R. De Gaudio, A. Casini and A. Di Filippo
10.1 Introduction
This chapter presents a summary of the latest guidelines for preventing infections
in intensive care units (ICUs), particularly regarding ventilation-associated
pneumonia (VAP), catheter-related bloodstream infections (CRBSI), and catheter-
associated urinary tract infections (UTI). Recent clinical trials were evaluated to
investigate the latest products, procedures, and treatments aimed at preventing
infections in ICUs. A summary table is provided, describing the most recent
acquisitions and their effectiveness (Table 10.1).
10.2 Ventilation-Associated Pneumonia
The 2004 guidelines for preventing VAP indicate the existence of fundamental
principles that possess level 1A evidence [1, 2]. These principles include educating
all workers about health epidemiology and VAP control procedures, instructing
them on how to effectively use various techniques, and continuous technique
updating through regular audits. The guidelines advise thorough cleaning of all
equipment (sterilization when possible, or high-level disinfection), with attention
to complete rinsing with sterile water, drying, and packaging. The ventilation
circuit should not be changed routinely on the basis of duration of use if used on a
single patient; rather, the circuit should be changed only when it is visibly dirty or
mechanically malfunctioning. When removing the condensation tube, healthcare
workers hands must be cleaned with soap and water or alcohol-based
A. Di Filippo (&)
Department of Critical Care Section of Anaesthesia,
c/o Careggi Teaching Hospital,
University of Florence, Firenze, Italy
Table 10.1 Recent advances in preventing intensive care patient infections and their
effectiveness assessed by large clinical trials
Method Study contents Efficacy Reference

Ventilation-associated pneumonia
Oral cavity Chlorhexidine (0,2%) versus potassium Not effective [3]
decontamination permanganate
Saline versus chlorhexidine 2% versus Effective [4]
chlorhexidine 2% ? colistin 2%
Chlorhexidine (0,2%) gel Not Effective [5]
Povidone iodine versus saline Effective [6]
Electric toothbrushes versus chlorhexidine Not effective [7]
Endotracheal Washing with isotonic saline Effective [8]
secretion Closed tracheal suction system versus open system Not effective [9]
removal
Subglottic Effective [10]
continuous
suction
Materials Polyurethane versus polyvinyl chloride tube Effective [11]
Polyurethane cuff and subglottic suction tube Effective [12]
versus polyvinyl tube with no subglottic suction
Cuff inflation Automatic Not effective [13]
control Low-pressure and low-volume inflating cuff Effective [14]
Cuff lubrication Effective [15]
Silver-coated Preliminary study Effective [16]
endotracheal NASCENT study Effective [17]
tube
Humidification Humidification method Not effective [18]
device Humidification method Not effective [19]
Antibacterial filters Not effective [20]
Tracheotomy Early tracheotomy Not effective [21]
Early tracheotomy Effective [22]
Mechanical Prophylactic PEEP Effective [23]
ventilation
mode
Patient Prone position Not effective [24, 25]
positioning
Selective Effective [26]
decontamination
of the
digestive tract
(continued)
10 Device Policies 161

Method Study contents Efficacy Reference
Catheter-related bloodstream infections
Medicated Catheters with silverplatinumcarbon versus Effective but [29]
catheters catheters with rifampicinminocycline not different
Catheters with chlorhexidine and sulfadiazine Effective [30]
versus standard catheters
Silver-impregnated versus standard multilumen Not effective [31]
catheters
Catheter Long-stay medicated catheter versus tunneled Effective [32]
tunneling catheter
Site chosen Jugular versus femoral Not effective [33]
Medication Chlorhexidine versus povidone iodine alcohol Effective [34]
Chlorhexidine for daily hygiene Effective [35]
Chlorhexidine-impregnated wipes Effective [36]
Urinary tract infections
Medicated Nitrofurazone-impregnated versus silicone Effective [38]
catheters catheters
Hydrophilic catheters Not effective [39]
Antibiotic Trimethoprim/sulfamethoxazole in three doses Effective [40]
prophylaxis
NASCENT North American Silver-Coated Endotracheal Tube investigation group
disinfectants. Nebulizers should only be used with sterile liquid and should be
placed with the most aseptic maneuver possible.
To prevent infection transmission from person to person, contaminated hands
should be frequently and thoroughly washed with water and soap (antimicrobial or
nonantimicrobial) or with an alcoholic antiseptic solution. This procedure must be
performed before and after any contact with intubated or tracheostomized
patients, as well as before and after contact with any respiratory device. It is also
recommended to wear gloves for handling respiratory secretions or contaminated
objects and to change gloves and decontaminate hands between contact with
different patients and between contact with a contaminated body site and the
respiratory tract.
To prevent inhalation of gastric content, it is good practice to remove all
devices, such as endotracheal, tracheostomy, or enteral feeding tubes when clinical
indications are absent. When the patients condition permits, orotracheal intuba-
tion should be preferred over nasotracheal intubation. Enteral tube placement
should be routinely controlled to prevent enteral nutrition-combined inhalation.
Postoperative pneumonia can be prevented by directing high-risk patients to
breathe deeply and ambulate as soon as possible. Indeed, this education would
benefit all postsurgery patients.
Recent clinical trials investigated the main methods for preventing VAP and
found that poor oral hygiene in patients undergoing mechanical ventilation is often
associated with secondary colonization of the respiratory tract, leading to sub-
sequent development of pneumonia. This observation implies that proper oral
hygiene should reduce VAP incidence; however, the existence of highly resistant
microorganisms makes the benefit of this strategy unclear.
Multiple strategies are used to decontaminate the oral cavity. For example,
Panchabhai et al. compared chlorhexidine (CHX) 0.2% with a control solution of
0.01% potassium permanganate in a prospective randomized trial [3]. Oral hygiene
was performed twice daily on 512 patients admitted to ICUs. Statistical analysis
indicated that the VAP incidence was not correlated with oral hygiene using CHX
or potassium permanganate. Mortality, seen as a secondary outcome, occurred at
rates of 34.8% for CHX-treated patients and 28.3% for the control group. How-
ever, the VAP incidence was lower (7.4%) 3 months after treatment compared
with 3 months before (21.7%), regardless of treatment type. Therefore, oral
hygiene seems to act as a protective factor against VAP development indepen-
dently of the pharmacological agent employed [3].
Koeman et al. conducted a randomized double-blind trial of 385 patients
divided into three groups: 130 placebo-treated patients, 127 treated with CHX 2%,
and 128 treated with CHX 2% ? colistin 2% [4]. The daily risk of VAP was
reduced in both treatment groups compared with placebo: 65% [hazard ratio
(HR) = 0.352; 95% confidence interval (CI), 0.1600.791; p = 0.012] for CHX
and 55% (HR = 0.454; 95% CI, 0.2240.925; p = 0.030) for CHX/COL. The
combination of colistin (COL) with CHX resulted in a significant reduction in
colonization by Gram-positive and Gram-negative endotracheal and oral micro-
organisms [4]. However, in a multicenter prospective double-blind trial con-
ducted in 228 patients undergoing mechanical ventilation for at least 5 days, oral
CHX gel decontamination was not associated with significant decreases in
VAP incidence, hospitalization duration, or mortality, nor was it effective against
multi-drug-resistant microorganisms (Pseudomonas aeruginosa, Acinetobacter,
Enterobacteriaceae) [5].
The use of povidone iodine for oral decontamination seems to be related to
reduced development of secondary infections. Regular use was associated with a
significant reduction in the incidence of VAP in a prospective randomized study
conducted on 98 patients with severe head trauma who underwent mechanical
ventilation for at least 48 h [6]. In this study, 36 patients were treated with
povidone iodine, 31 with saline, and the rest with simple secretion aspiration; VAP
incidence was 8, 39, and 42%, respectively. There were no statistically significant
differences in duration of hospitalization or intra-ICU mortality among the
groups [6].
Effective VAP reduction due to the use of electric toothbrushes is still con-
troversial. In a recent prospective trial conducted on 147 patients intubated for
more than 48 h, Pobo et al. demonstrated that the use of electric toothbrushes did
not significantly reduce VAP development compared with 0.12% CHX oral
hygiene [7].
Removing endotracheal secretions appears to be a primary step in preventing

VAP and can be carried out continuously or at scheduled intervals. Periodic
secretion aspiration after instilling isotonic saline seems to reduce VAP incidence
in patients with tracheotomies. A randomized trial of 262 patients admitted to
ICUs revealed that VAP incidence was significantly lower in patients in whom
aspiration was performed following administered isotonic saline (23.5 vs. 10.8%,
p \ 0.008), with a relative risk of 54% (95% CI = 1874%). In contrast, the
incidence of atelectasis and endotracheal tube obstruction was similar in both
groups [8]. Secretion aspiration can be performed with closed or open systems. No
significant difference in VAP prevention was found in a randomized study of 443
patients, 210 treated with a closed tracheal suction system and 233 with an open
system [9]. However, continuous subglottic secretion aspiration can reduce the use
of antimicrobial agents and the incidence of VAP in patients undergoing cardiac
surgery [10].
For patients undergoing mechanical ventilation via endotracheal tube place-
ment, special attention must be paid to the material used to build the device and the
integrity of the structures of the device itself. These patients experience an
increased risk of nosocomial pneumonia, and microsubglottic contaminated
secretions are among the leading causes of VAP.
A polyurethane-cuffed tube may prevent VAP onset. For example, Poelaert
et al. in a study of 134 postcardiac surgery patients, showed that the incidence of
pneumonia during early postoperative mechanical ventilation was significantly
reduced when a polyurethane tube was used compared with a polyvinyl chloride
tube [11]. These results were corroborated by Lorente et al. in a randomized study
of 280 patients. VAP incidence was 22.1% in the patient group (n = 140) with
conventional polyvinyl chloride endotracheal tubes with no subglottic aspiration
compared with 7.9% (p \ 0.001) in patients with polyurethane cuff tubes and
subglottic continuous suction. These devices appear to reduce the incidence of
VAP whether used early or late [12].
Automated devices more accurately assess adherence of the endotracheal tube
cuff; however, available data suggest that this is not sufficient to significantly
reduce VAP occurrence [13]. In addition, studies in animal models have shown
that cuff inflation performed at low pressures and low volumes may reduce the risk
of inhalation, especially during tracheotomy. This situation cannot be true in cases
of selective intubation or when the positioning of the endotracheal tube excludes
areas of lung ventilation [14]. Cuff lubrication reduces fluid outflow during
thoracic surgery [15].
Endotracheal tubes coated with silver appear to contribute to VAP prevention.
A preliminary study demonstrated reduced airway bacterial colonization [16].
These data were partly confirmed by the 2003 North American Silver-Coated
Endotracheal Tube (NASCENT) investigation group study of patients undergoing
mechanical ventilation [24 h. Data analysis from bronchoalveolar lavage showed
that VAP onset was delayed and its incidence reduced compared with the control
group [17].
Even circuit components used for ventilation and humidification of the gas
mixtures administered to patients can serve as a source of bacterial contamination,
prompting the onset of pneumonia. However, the likelihood of this occurrence is in
dispute. A study of an unselected population of 369 ICU patients who underwent
mechanical ventilation [48 h demonstrated that VAP onset was not related to the
device types used to humidify the gas mixture [18]. Similar results emerged from
another study conducted on 181 patients [19]. Moreover, bacterial filters do not
appear to be involved in reducing VAP incidence. Bacterial filters used in a group
of 230 patients undergoing mechanical ventilation [24 h did not reduce the
prevalence of respiratory infections associated with mechanical ventilation [20].
Tracheotomy is commonly performed in patients undergoing mechanical ven-
tilation over shorter and longer periods. An early tracheotomy may reduce the
duration of mechanical ventilation and the incidence of respiratory tract infections,
as well as improve patient comfort and reduce respiratory dead space. Evidence
also exists to the contrary. Early tracheotomy (within 8 days) in a small group
(60) of trauma patients was unable to reduce ventilation duration, pneumonia
frequency, or ICU hospitalization compared with a control group in which the
tracheotomy was performed C28 days after the acute event [21]. However, a
prospective study of 62 patients with isolated severe head trauma showed
that early tracheostomy shortened mechanical ventilation duration after VAP
onset [22].
The mode of mechanical ventilation also appears to play a role in preventing
VAP, but the use of positive end-expiratory pressure (PEEP) seems to be an issue.
PEEP effectiveness was recently studied in 131 mechanically ventilated patients
with chest radiographs and Horowitz indexes [250 [23]. The primary outcome
was intra-ICU mortality and secondary outcomes were VAP, acute respiratory
distress syndrome (ARDS), barotrauma, occurrence of atelectasis, and develop-
ment of hypoxemia. The application of prophylactic PEEP in ventilated hypox-
emic patients did not reduce the number of hypoxemia episodes or the VAP
incidence [23].
Placing the patient undergoing mechanical ventilation in the prone position
appears to be associated with reduced gastric content aspiration and VAP inci-
dence. However, there is a lack of data on large-scale clinical studies that validate
the effectiveness and feasibility of this strategy. For example, a multicenter pro-
spective study conducted on ICU patients on mechanical ventilation showed that
the prone position is not easily achieved and its effectiveness is therefore probably
negative [24]; these data were also confirmed in children [25].
The role of selective decontamination (SDD) of the digestive tract is contro-
versial with regards to VAP. De la Cal et al. employed a randomized double-blind
study of 107 patients with severe burns at high risk of inhalation to assess whether
SDD could reduce the incidence of infections, morbidity, and mortality in criti-
cally ill patients. Statistical analysis demonstrated that the treatment was able to
reduce mortality and incidence compared with a control group [26].
10.3 Catheter-Related Bloodstream Infections
Given the incidence and clinical impact of catheter-related bloodstream infections

(CRBSI) (5.3 per 1,000 catheter days), practical guidelines for their prevention
were proposed based on type-A evidence [27, 28]. For example, educating medical
staff is encouraged regarding indications for using vascular catheters, proper
device insertion and maintenance, and careful surveillance of control measures to
prevent infection. Testing knowledge and guideline compliance should be con-
ducted, as well as management of staff who insert catheters, ensuring an adequate
number of nurses on duty. Monitoring catheter insertion sites visually or by pal-
pation through the intact dressing is important, depending on the clinical situation
of the individual patient. If patients exhibit tenderness at the insertion site without
an obvious source of infection, with or without fever, the dressing should be
removed to examine the site. It is important to observe correct procedures with
regard to hand hygiene by washing hands with conventional antiseptic soap and
water or an alcohol-based gel. Hand hygiene must be performed before and after
palpating the insertion site, as well as before and after insertion, replacement, or
medication. The use of gloves does not eliminate the need for hand hygiene.
It is important to maintain an aseptic technique during the insertion of intra-
venous catheters and dressings. Indeed, gloves should be cleaned or sterilized
before inserting an intravascular catheter. Wearing clean gloves when inserting
catheters is acceptable, as is not touching the peripheral access site after appli-
cation of skin antiseptics; sterile gloves should be worn for inserting arterial
catheters or central lines. Clean or sterile gloves should also be worn when
changing dressings.
Skin disinfection must be done with an appropriate antiseptic before inserting
the catheter and during dressing changes. Although it is preferable to use CHX 2%,
povidone iodine can be used with 70% alcohol. The antiseptic must be allowed to
air dry; povidone must remain on the skin for at least 2 min, or longer if not
completely dry. The dressing to be applied over the catheter insertion site should
be of sterile gauze or semipermeable dressings and be transparent. The dressing
should be replaced when wet, nonadhesive, or visibly soiled. Topical antibiotics
should not be used on the insertion site to avoid fungal infections and microbial
resistance.
Catheter type and insertion technique and site should be chosen for the lowest
risk of complications for the therapy type and expected duration. Any catheter that
is no longer essential should be promptly removed. In adults, it is necessary to
replace peripheral venous catheters at least every 7296 h to prevent phlebitis.
Short-term central venous catheters (CVCs) should be replaced if there are signs of
infection at the insertion site; they should not be used in guide-wire techniques to
replace catheters in patients with suspected CRBSI.
Administration sets must be frequently replaced at a minimum time interval of
72 h unless a CRBSI is suspected or documented. The infusion sets used to
administer blood (or its derivatives) or lipid emulsions should be replaced within
24 h of starting the infusion. If injections are performed, taps must first be dis-
infected, and all taps not in use should be covered. Contaminated fluids must not
be administered; it is therefore recommended that parenteral nutrition be prepared
aseptically in the pharmacy. It is also not recommended to use in-line filters for
preventing infections, as administering systemic antibiotic prophylaxis routinely
before insertion or during use of an intravascular catheter adequately prevents
colonization.
Devices with as few lines as possible should be selected for patient manage-
ment. Catheters impregnated with antimicrobial or antiseptic are best choices if
they remain in place [5 days and if the infection rate remains high in the ward
after using a global strategy. The risks and benefits of placing a CVC in the
recommended site must be weighed, comparing the reduction of infectious com-
plications and mechanical complications. Subclavian access is recommended for
reducing infection in nontunneled catheters, and using the jugular or femoral vein
is recommended regarding CVC dialysis (to avoid venous stasis). An episode of
fever should not lead to changing these devices, as a clinical evaluation will
determine whether to remove the catheter if infection is evidenced elsewhere or if
the cause is not infectious. Routine exchange by guidewire should not be used to
prevent infections; this technique should be reserved for cases of catheter mal-
function without signs of infection. Disposable peripheral arterial catheters should
be used with monitoring systems and should not substitute for routine peripheral
arterial catheters to prevent catheter infections.
Clinical trials have recently been used to assess various methods to prevent
CRBSI. For example, using particular types of medicated catheters appeared to
correlate with a reduced incidence of catheter-related infections. A randomized
study of 646 catheterizations compared the effectiveness of silverplatinum
carbon antimicrobial catheters with rifampicinminocycline medicated catheters.
The former served as effective antimicrobials, but the proportions of catheter-
related infections were extremely low in both groups (1.4% for silverplatinum
carbon catheters vs. 1.7% for medical catheters with rifampicinminocycline)
[29]. Rupp et al. initiated a multicenter randomized double-blind controlled study
of 780 patients to compare the incidence of infections due catheters medicated
with CHX and sulfadiazine versus standard catheters. The results suggested that
medicated catheters are highly tolerated by the patient, with less colonization by
pathogenic organisms at the time of removal. The main colonizing microorgan-
isms were coagulase-negative staphylococci and other Gram-positive microor-
ganisms. Noninfectious adverse events occurred in both groups with comparable
frequencies [30]. However, a recent prospective multicenter randomized con-
trolled study demonstrated that there was no effective reduction in the infection if
multilumen CVCs or catheters impregnated with silver were used. In both groups
of patients, the incidence of infections was high: 2.5% in the standard catheter
group versus 2.7% in the multilumen silver-impregnated catheter group [31].
Tunneling catheters are often used to prevent colonization and subsequent
infections related to long-standing catheters. Darouiche et al. showed that a group
of patients with long-stay antimicrobial-medicated catheters experienced a lower
incidence of catheter-related infections than patients with tunneled catheters

(3.6 vs. 1.43 per 1,000 catheter days) [32].
The impact of the site of catheter insertion has not yet been determined in
preventing catheter-related infections. In a recent multicenter randomized inves-
tigation, Parienti et al. studied the infection incidence in patients with CVCs,
comparing femoral versus jugular access. The entire patient cohort was subjected
to cycles of short-term dialysis. The risk of colonization (assessed per 1,000
catheter days) was similar in both groups: 40.8 for femoral access versus 35.7 for
jugular access. However, a higher infection incidence occurred in patients with
body mass index (BMI) [28.4 with femoral central venous access (50.9 femoral
vs. 24.5 jugular). The risk of hematoma was highest when using jugular
access [33].
Careful disinfection of the CVC insertion site may reduce the incidence of
infection. Mimoz et al. conducted a randomized study of 538 CVCs with jugular
and subclavian access, demonstrating that the use of CHX may represent a valid
alternative to the use of povidone iodine alcohol. CHX was able to reduce the
incidence of colonization by 50% (p \ 0.002) for the same incidence of overt
bacterial infection (p \ 0.09) [34]. Bleasdale et al. confirmed these data in a
randomized study, emphasizing that using CHX for daily patient hygiene can
reduce the incidence of primary systemic infections [35]. Timsit et al. evaluated
the effectiveness of using CHX-gluconate-impregnated sponges to reduce the
incidence of infection originating from the catheter insertion site. The randomized
controlled trial conducted in 3,778 catheters kept in place for an average of 6 days
demonstrated that the use of medicated wipes with CHX reduced the incidence of
infection [6].
10.4 Urinary Tract Infections
Several studies provide useful information regarding UTI prevention [2, 37].
These studies argue that UTI prevention occurs in surgical patients by limiting the
use of bladder catheters to only those cases with real need. The necessity of
the bladder catheter should be assessed on the basis of clinical circumstances, and
the routine use of these devices is therefore not recommended. Indwelling urinary
catheters are recommended in long-term-care patients for managing incontinence.
Closed drainage systems are recommended after aseptically inserting the uri-
nary catheter; catheters should only be inserted under the appropriate indications
and should be kept in place only as long as necessary. Use and duration should be
minimized in all patients, especially in those at increased risk for UTI, such as
women, the elderly, and patients with compromised immune systems. Aseptic
insertion should be ensured; and hospital staff, patients, and family should be
educated in the correct insertion technique. It is necessary to attain a free
urine flow.
The risks and benefits of various catheterization approaches have also been
assessed. If an intermittent catheterization is preferable, this technique should be
performed at regular intervals to avoid bladder distension. When urinary catheter
placement is indicated in surgical patients, it must be removed as soon as possible
after surgery, preferably within 24 h, unless there are appropriate indications for
continuous use. If UTI rate is not reduced after applying a comprehensive strategy
(appropriate positioning and proper aseptic maintenance), using catheters impreg-
nated with an antimicrobial or antiseptic should be considered. After inserting the
catheter aseptically, maintaining a closed drainage system is recommended.
In the absence of specific clinical indications, the use of systemic antibiotics as
routine UTI prophylaxis is not recommended. For UTI prevention, it is not necessary
to clean the periurethral area when routine cleaning during daily hygiene is required.
Replacing catheters or drainage bags at regular intervals is also not recommended.
Catheter replacement should be based on clinical indications, such as infections,
obstructions, or when the closed system is compromised. Disposable sterile gel is
also recommended for catheter insertion. Medical staff and professionals who
manage catheters should continue to take refresher courses and further education for
insertion and removal of the maintenance device. Hand hygiene before catheter
insertion or any manipulation of the catheter site or devices is mandatory.
Several recent clinical trials have been performed regarding the prevention of
UTI. A randomized double-blind controlled study of 212 patients enrolled consec-
utively after traumatic events compared the UTI incidence in patients with silicone
urinary catheters with patients with catheters impregnated with nitrofurazone. There
were significantly fewer bacteriurias and fungurias associated with nitrofurazone-
impregnated bladders than with the silicone urinary catheters (9.1 vs. 24.7%). The
clinical significance of asymptomatic bacteria and fungi in urine was unclear, lim-
iting the study [38]. However, the use of hydrophilic catheters in a randomized
controlled trial for self-intermittent catheterization after spinal injury did not sig-
nificantly reduce the UTI incidence compared with standard catheters [39].
Finally, the efficacy of prophylactic antibiotic therapy before bladder catheter
removal was assessed. The prospective randomized study was conducted on 239
patients undergoing major abdominal surgery who were catheterized periopera-
tively; the usefulness and effectiveness of trimethoprimsulfamethoxazole in three
doses was evaluated. Urine cultures were taken before and 3 days after bladder
catheter removal. Trimethoprimsulfamethoxazole administration significantly
reduced the incidence of symptomatic UTI if administered before catheter removal
(4.9% of the analysis group vs. 21.6% in the control group, p \ 0.001) [40].
10.5 Conclusions
Recent findings seem to confirm the effectiveness of certain procedures for

infection prevention, particularly with regard to VAP. Oral-cavity decontamina-
tion, continuous subglottic secretion suction, early tracheotomy, SDD and, for
high-risk patients, endotracheal tubes with silver were all shown to be useful in
preventing VAP. Medicated catheters and CHX-based dressings were efficacious
against CRBSI. UTIs were shown to be prevented through the use of medical
catheters. All these procedures can be incorporated into departmental protocols for
preventing nosocomial infections in ICUs.
References
1. Centers for Disease Control and Prevention (2004) Guidelines for preventing health-care-
associated pneumonia, 2003: recommendations of CDC and the Healthcare Infection Control
Practices Advisory Committee. MMWR 53(RR-3)
2. Di Filippo A, De Gaudio AR (2003) Device-related infections in critically ill patients. Part II:
prevention of ventilator-associated pneumonia and urinary tract infections. J Chemother
15(6):536542
3. Panchabhai TS, Dangayach NS, Krishnan A et al (2009) Oropharyngeal cleansing with 0.2%
chlorhexidine for prevention of nosocomial pneumonia in critically ill patients: an open-label
randomized trial with 0.01% potassium permanganate as control. Chest 135(5):11501156
4. Koeman M, van der Ven AJ, Hak E et al (2006) Oral decontamination with chlorhexidine
reduces the incidence of ventilator-associated pneumonia. Am J Respir Crit Care Med
173(12):13481355
5. Fourrier F, Dubois D, Pronnier P et al (2005) Effect of gingival and dental plaque antiseptic
decontamination on nosocomial infections acquired in the intensive care unit: a double-blind
placebo-controlled multicenter study. Crit Care Med 33(8):17281735
6. Seguin P, Tanguy M, Laviolle B et al (2006) Effect of oropharyngeal decontamination by
povidone-iodine on ventilator-associated pneumonia in patients with head trauma. Crit Care
Med 34(5):15141519
7. Pobo A, Lisboa T, Rodriguez A, RASPALL Study Investigators et al (2009) A randomized
trial of dental brushing for preventing ventilator-associated pneumonia. Chest 136(2):
433439
8. Caruso P, Denari S, Ruiz SA et al (2009) Saline instillation before tracheal suctioning
decreases the incidence of ventilator-associated pneumonia. Crit Care Med 37(1):3238
9. Lorente L, Lecuona M, Martn MM et al (2005) Ventilator-associated pneumonia using a
closed versus an open tracheal suction system. Crit Care Med 33(1):115119
10. Bouza E, Prez MJ, Muoz P et al (2008) Continuous aspiration of subglottic secretions in
the prevention of ventilator-associated pneumonia in the postoperative period of major heart
surgery. Chest 134(5):938946
11. Poelaert J, Depuydt P, De Wolf A et al (2008) Polyurethane cuffed endotracheal tubes to
prevent early postoperative pneumonia after cardiac surgery: a pilot study. J Thorac
Cardiovasc Surg 135(4):771776
12. Lorente L, Lecuona M, Jimnez A et al (2007) Influence of an endotracheal tube with
polyurethane cuff and subglottic secretion drainage on pneumonia. Am J Respir Crit Care
Med 176(11):10791083
13. Valencia M, Ferrer M, Farre R et al (2007) Automatic control of tracheal tube cuff pressure in
ventilated patients in semirecumbent position: a randomized trial. Crit Care Med 35(6):
15431549
14. Young PJ, Pakeerathan S, Blunt MC et al (2006) A low-volume, low-pressure tracheal tube
cuff reduces pulmonary aspiration. Crit Care Med 34(3):632639
15. Sanjay PS, Miller SA, Corry PR et al (2006) The effect of gel lubrication on cuff leakage of
double lumen tubes during thoracic surgery. Anaesthesia 61(2):133137
16. Rello J, Kollef M, Diaz E et al (2006) Reduced burden of bacterial airway colonization with a
novel silver-coated endotracheal tube in a randomized multiple-center feasibility study. Crit
Care Med 34(11):27662772
17. Kollef MH, Afessa B, Anzueto A et al (2008) Silver-coated endotracheal tubes and incidence
of ventilator-associated pneumonia: the NASCENT randomized trial. JAMA 300(7):805813
18. Lacherade JC, Auburtin M, Cerf C et al (2005) Impact of humidification systems on
ventilator-associated pneumonia: a randomized multicenter trial. Am J Respir Crit Care Med
172(10):12761282
19. Boots RJ, George N, Faoagali JL et al (2006) Double-heater-wire circuits and heat-and-
moisture exchangers and the risk of ventilator-associated pneumonia. Crit Care Med
34(3):687693
20. Lorente L, Lecuona M, Mlaga J et al (2003) Bacterial filters in respiratory circuits: an
unnecessary cost? Crit Care Med 31(8):21262130
21. Barquist ES, Amortegui J, Hallal A et al (2006) Tracheostomy in ventilator dependent trauma
patients: a prospective, randomized intention-to-treat study. J Trauma 60(1):9197
22. Bouderka MA, Fakhir B, Bouaggad A et al (2004) Early tracheostomy versus prolonged
endotracheal intubation in severe head injury. J Trauma 57(2):251254
23. Manzano F, Fernndez-Mondjar E, Colmenero M et al (2008) Positive-end expiratory
pressure reduces incidence of ventilator-associated pneumonia in nonhypoxemic patients.
Crit Care Med 36(8):22252231
24. van Nieuwenhoven CA, Vandenbroucke-Grauls C, van Tiel FH et al (2006) Feasibility and
effects of the semirecumbent position to prevent ventilator-associated pneumonia:
a randomized study. Crit Care Med 34(2):396402
25. Aly H, Badawy M, El-Kholy A et al (2008) Randomized, controlled trial on tracheal
colonization of ventilated infants: can gravity prevent ventilator-associated pneumonia?
Pediatrics 122(4):770774
26. de La Cal MA, Cerd E, Garca-Hierro P et al (2005) Survival benefit in critically ill burned
patients receiving selective decontamination of the digestive tract: a randomized, placebo-
controlled, double-blind trial. Ann Surg 241(3):424430
27. Centers for Disease Control and Prevention (2002) Guidelines for the prevention of
intravascular catheter-related infections. MMWR 2002 51(RR-10)
28. De Gaudio AR, Di Filippo A (2003) Device-related infections in critically ill patients. Part I:
prevention of catheter-related bloodstream infections. I Chemother 15(5):419427
29. Fraenkel D, Rickard C, Thomas P et al (2006) A prospective, randomized trial of rifampicin-
minocycline-coated and silver-platinum-carbon-impregnated central venous catheters. Crit
Care Med 34(3):668675
30. Rupp ME, Lisco SJ, Lipsett PA et al (2005) Effect of a second-generation venous catheter
impregnated with chlorhexidine and silver sulfadiazine on central catheter-related infections:
a randomized, controlled trial. Ann Intern Med 143(8):570580
31. Kalfon P, de Vaumas C, Samba D et al (2007) Comparison of silver-impregnated with standard
multi-lumen central venous catheters in critically ill patients. Crit Care Med 35(4):10321039
32. Darouiche RO, Berger DH, Khardori N et al (2005) Comparison of antimicrobial impregnation
with tunneling of long-term central venous catheters: a randomized controlled trial. Ann Surg
242(2):193200
33. Parienti JJ, Thirion M, Mgarbane B et al (2008) Femoral vs jugular venous catheterization
and risk of nosocomial events in adults requiring acute renal replacement therapy:
a randomized controlled trial. JAMA 299(20):24132422
34. Mimoz O, Villeminey S, Ragot S et al (2007) Chlorhexidine-based antiseptic solution vs
alcohol-based povidone-iodine for central venous catheter care. Arch Intern Med 167(19):
20662072
35. Bleasdale SC, Trick WE, Gonzalez IM et al (2007) Effectiveness of chlorhexidine bathing to
reduce catheter-associated bloodstream infections in medical intensive care unit patients.
Arch Intern Med 167(19):20732079
36. Timsit JF, Schwebel C, Bouadma L et al (2009) Chlorhexidine-impregnated sponges and less
frequent dressing changes for prevention of catheter-related infections in critically ill adults:
a randomized controlled trial. JAMA 301(12):12311241
37. Healthcare Infection Control Practices Advisory Committee (2009) Guideline for prevention
of catheter-associated urinary tract infections. http://www.cdc.gov/ncidod/dhqp/pdf/
guidelines/CAUTI_Guideline2009final.pdf. Accessed June 2011
38. Stensballe J, Tvede M, Looms D et al (2007) Infection risk with nitrofurazone-impregnated
urinary catheters in trauma patients: a randomized trial. Ann Intern Med 147(5):285293
39. Cardenas DD, Hoffman JM (2009) Hydrophilic catheters versus noncoated catheters for
reducing the incidence of urinary tract infections: a randomized controlled trial. Arch Phys
Med Rehabil 90(10):16681671
40. Pfefferkorn U, Lea S, Moldenhauer J et al (2009) Antibiotic prophylaxis at urinary catheter
removal prevents urinary tract infections: a prospective randomized trial. Ann Surg
249(4):573575
Antibiotic Policies in the Intensive
Care Unit 11
H. K. F. van Saene, N. J. Reilly,
A. de Silvestre and F. Rios
11.1 Introduction
Every intensive care unit (ICU) should have well-structured guidelines on the use
of antimicrobial agents to guarantee that patients requiring intensive care receive
appropriate antimicrobials for a relevant period to prevent and treat infections.
These guidelines should meet the therapeutic needs of the consultants and allow
the intensivist, clinical microbiologist, and pharmacist to monitor efficacy, toxic-
ityincluding allergy and diarrheaand side effects, such as the emergence of
resistant strains and subsequent outbreaks of superinfections. Calculation of
infection rates is only feasible following implementation of an antibiotic policy.
Apart from audit and research, antimicrobial guidelines aid educational programs
and enable the clinical pharmacist to control drug expenditure.
11.2 Main Feature of Antibiotic Guidelines
The main feature of an antibiotic policy in the ICU is the use of a minimum of
well-established antimicrobial agents that are associated with a minimum of side
effects but also allow the control of the three patterns of ICU infections due to the
15 potentially pathogenic microorganisms (PPM) (Chaps. 3 and 5).
H. K. F. van Saene (&)

Institute of Ageing and Chronic Disease, University of Liverpool,
Duncan Building, Liverpool, UK
11.2.1 Antimicrobials Chosen on the Basis

of Three Characteristics
11.2.1.1 Preserving Indigenous Flora

Control of the abnormal flora, including (potentially resistant) aerobic Gram-
negative bacilli (AGNB) and methicillin-resistant Staphylococcus aureus
(MRSA), by normal indigenous, mainly anaerobic, flora is termed microbial
ecology [1]. Normal flora provides resistance to acquisition and subsequent
carriage of PPM and constitutes the microbial component of the carriage defense
(Chap. 2). Antimicrobial agents active against indigenous anaerobic flora and
excreted into throat and gut via saliva, bile, and mucus in lethal concentrations,
may suppress indigenous flora. Diarrhea and candidiasis are well known side
effects of commonly used antibiotics. Disregarding ecology by using antimi-
crobial agents leads to an impaired microbial factor of carriage defense and
promotes abnormal AGNB [2] and MRSA [3] overgrowth, as well as
Clostridium difficile [4] and vancomycin-resistant enterococci (VRE) [5].
Narrow-spectrum antimicrobials can be defined as those that only cover the 15
aerobic PPM, leaving indigenous normal flora more or less intact [6]. Broad-
spectrum antimicrobials are not only active against the 15 disease-causing PPM
but affect the normal indigenous anaerobic flora that contributes to physiology
rather than infection. From an ecological perspective, ampicillin, amoxicillin,
and flucloxacillin are antimicrobials with a broader spectrum than that of
cephradine and cefotaxime [711]. Metronidazole, although active against
anaerobes, does not affect indigenous flora, as it is readily neutralized following
biliary excretion [12]. The most recent, ever-more potent, antimicrobials, such as
ceftriaxone, generally belong to the group of ecology-disregarding antimicrobials
that predispose to, e.g., AGNB, MRSA, and yeast overgrowth and subsequent
superinfection [3, 1317] (Table 11.1).
11.2.1.2 Limiting the Emergence of Resistant Microbes by Using

Antimicrobials with the Lowest Resistance Potential
If resistance occurs during drug development or clinical trials or within 2 years of
general use, the antibiotic has a high resistance potential [18]. Ceftazidime-,
ciprofloxacin-, and imipenem-resistant Pseudomonas aeruginosa were reported
during clinical trials and early after introduction for general use [1921]. Anti-
microbials with little or no resistance potential include cephradine, piperacillin,
cefotaxime, amikacin, and meropenem (Table 11.2) [2124]. Three observations
have emerged from this historical experience: (1) each antibiotic class has one or
more antimicrobials capable of causing antibiotic resistance, but this is not a class
phenomenon; antibiotic resistance is agent specific; (2) resistance is not related to
duration of use; and (3) resistance is not related to volume of use.
The underlying mechanisms explaining the difference between antibiotics with a
high and a low resistance potential are not fully understood but are thought to be
based on specific antibiotic-resistance mechanisms. For example, among the ami-
noglycosides, only gentamicin has been associated with widespread P. aeruginosa
11 Antibiotic Policies in the Intensive Care Unit 175
Table 11.1 Classification of

Indigenous flora friendly Indigenous flora suppressing
parenterally administered
antimicrobials based on Challenge studies in volunteers
respect for ecology
Cephradine, cefotaxime Ampicillin, amoxicillin
Comparative studies in patients
Cephradine, penicillin Amoxicillin, Ampicillin
Observational studies in patients
Gentamicin, tobramycin, Flucloxacillin,
amikacin, polymyxins, polyenes, amoxicillin ? clavulanic
metronidazole acid,
piperacillin ? tazobactam,
ceftriaxone, ciprofloxacin,
carbapenems, macrolides

Low resistance High resistance
parenteral antimicrobials
based on the potential for Piperacillin Ampicillin
resistance
Cephradine Ceftazidime
Cefotaxime Gentamicin
Cefepime Ciprofloxacin
Amikacin Imipenem
Levofloxacin
Meropenem
resistance, but amikacin continues to be effective against most gentamicin-resistant

P. aeruginosa. Gentamicin is an aminoglycoside that is highly susceptible to
inactivation by a variety of enzymes at six different loci on the gentamicin mole-
cule, but amikacin has only one such vulnerable point [25]. Substituting amikacin
for gentamicin decreased gentamicin-resistant P. aeruginosa when amikacin was
used in the same volume as gentamicin, and no subsequent resistance problems
developed. Some institutions with renewed gentamicin susceptibility to
P. aeruginosa returned to using gentamicin as the major hospital aminoglycoside
(i.e., rotating formularies) [26]. Predictably, gentamicin-resistant P. aeruginosa
isolates returned to previous levels and were perpetuated as long as gentamicin was
the primary aminoglycoside used in the hospital and ICU. Some people believe that
these aminoglycoside lessons of the past should be applied to the problems of
P. aeruginosa resistant to ceftazidime, ciprofloxacin, and imipenem. Substitution of
cefepime, levofloxacin, and meropenem for ceftazidime, ciprofloxacin, and imi-
penem may decrease or even eliminate resistant P. aeruginosa from the ICU
environment [2729]. According to the concept of low and high resistance potential
of antimicrobial agents, the key to controlling antibiotic resistance is agent related

and is not class, duration, or volume associated.
11.2.1.3 Controlling Inflammation by Using Antimicrobials

with Anti-Endotoxin/Inflammation Properties
Lipopolysaccharide (LPS), or endotoxin, is the major component of the outer
membrane of AGNB and is thought to be a key molecule in the induction of
generalized inflammation. LPS causes the production and release from host
effector cells of various proinflammatory cytokines and other mediators of
inflammation, such as nitric oxide and prostaglandins. The degree to which these
mediators are released depends, in part, on the amount of LPS that is presented to
CD14-bearing effector cells, such as monocytes, macrophages, and polymorpho-
nuclear leukocytes. Therefore, it is possible that factors affecting the amount of
LPS released in vivo may modulate the inflammatory response associated with
AGNB infection. Several in vitro and animal studies show that antimicrobial
agents may differentially release LPS from AGNB [30, 31]. They also demon-
strated that antibiotics associated with substantial release of endotoxin generate
higher levels of proinflammatory cytokines and suggested that using antimicro-
bials associated with the release of lower amounts of LPS leads to lower levels of
cytokinemia and better outcome. Carbapenems, such as imipenem and merope-
nem, release a small amount of endotoxin, whereas third-generation cephalospo-
rins, such as ceftazidime and cefotaxime, are associated with far greater release of
endotoxin [32]. Three clinical trials failed to support this hypothesis that differ-
ential antibiotic-induced endotoxin release is of clinical significance, as there were
no significant differences in clinical parameters (temperature, blood pressure, or
heart rate) or plasma endotoxin and cytokine levels [3335].
Fluoroquinolones, such as ciprofloxacin, release substantial amounts of endo-
toxin compared with the aminoglycosides gentamicin and tobramycin and with the
polymyxins E (colistin) and B [36, 37]. Although not active against AGNB,
glycopeptides, including teicoplanin [38] and vancomycin [39]; and polyenes,
such as amphotericin B [40], have been shown to downregulate LPS-induced
cytokine release (Table 11.3).
Classifying antimicrobials using these three criteria of flora friendliness, low
resistance potential, and anti-inflammation propensity is not evidence based, i.e.,
not evaluated in randomized trials. However, despite the lack of level 1 evidence,
we found the available data compelling and difficult to ignore in selecting anti-
microbials for our antibiotic policy.
Using a limited number of antimicrobial agents allows the control of the 15
potential pathogens implicated in the three types of ICU infections. The main
antimicrobial groups are:
1. b-lactams, i.e., antibiotics with a b-lactam ring in their structure, such as
penicillins and cephalosporins;
2. aminoglycosides, e.g., tobramycin;
3. polymyxins, e.g., polymyxin E or colistin;

Inflammation controlling Inflammation promoting
antimicrobials based on their
anti-inflammation Aminoglycosides Beta-lactams
propensities
Glycopeptides Cefotaxime
Polymyxins Ceftazidime
Polyenes Fluoroquinolones
Carbapenems Ciprofloxacin
4. glycopeptides, e.g., vancomycin;

5. polyenes, such as amphotericin B.
All these agents are lethal to microorganisms. Polymyxins, glycopeptides, and
polyenes have a rapid action on the microbial cell membrane. b-lactams interfere
with the cell-wall synthesis, a slower mechanism of action. Aminoglycosides
inhibit protein synthesis but still kill microbes in the rest phase. These differences
in mechanism of action may explain why aminoglycosides and polyenes are more
toxic to the ICU patient than are b-lactams when parenterally administered.
Table 11.4 shows the spectrum of activity of the five main antimicrobial groups
according to the minimal bactericidal concentration (MBC) for the 15 PPMs. The
MBC of an antimicrobial agent is defined as the amount of the antimicrobial
(milligram per liter) required to establish irreversible inhibition in the test tube
without the support of the killing activity of leukocytes of the ICU patient. Leu-
kocytes of the critically ill are thought to be less effective in killing PPM than
those of a healthy population [41]. Antimicrobial agents with an MBC of 1 mg/l
for PPM are in general suitable for clinical use. Nontoxic antibiotics, such as the
b-lactams, are ideal for systemic administration, and high doses (50100 mg/kg
per day) can be given. The more toxic agents, such as polymyxins and the poly-
enes, can be safely applied enterally and topically in high doses. Polyenes given
parenterally are toxic, and the daily doses are in the order of milligrams (1.5 mg/
kg per day for amphotericin B). Aminoglycosides and glycopeptides, although
toxic, are administered systemically in lower doses (525 mg/kg per day).
11.3 Antimicrobial Use for Both Prophylaxis and Therapy
Antimicrobial administration in the ICU falls into three categories: parenteral,

enteral nonabsorbable, and topical.
Parenteral administration is aimed at preventing and, if already present, eradi-

cating colonization/infection of internal organs, including lungs, and bladder, as
well as in the bloodstream. Systemic prophylaxis is generally accepted in surgery
Table 11.4 Spectrum of activity of the commonly used antimicrobial agents for the 15 potential pathogens expressed by the minimal bactericidal
178
concentration (MBC) (milligrams per liter)
PPM Antimicrobial agents
Penicillin G Cephradine Cefotaxime Ceftazidime Aminoglycosides, Polymyxins, Glycopeptides, Polyenes,

e.g., tobramycin e.g., polymyxin E e.g., vancomycin e.g., amphotericin B
Normal
S. pneumoniae 0.1 0.2 (2)
H. influenzae 0.06 0.2
M. catarrhalis 1 1 1
E. coli 0.02 0.2 0.2 0.1
S. aureus 1 1 (8.0) 0.2 0.05
Candida spp.
0.05
Abnormal
Klebsiella spp. 0.1 0.1 0.1 0.1
Proteus spp. 0.1 0.1 0.2
Morganella spp. 0.1 0.1 0.2
Citrobacter spp. 0.3 0.3 0.1 0.1
Enterobacter spp. 0.2 0.2 0.1 0.1
Serratia spp. 1 1 0.2
Acinetobacter spp. (8) (8) 0.1 0.1
Pseudomonas spp. 0.1 0.1 0.5
MRSA 1.0
PPM potentially pathogenic microorganisms, MRSA methicillin-resistant Staphylococcus aureus
H. K. F. van Saene et al.
[42]. The aim is to achieve a tissue concentration at the time of the surgical trauma
in order to prevent wound infections. Patients who require intensive care due to an
acute trauma or worsening of the underlying disease invariably receive invasive
devices, including ventilation tube, urinary catheter, and intravascular lines. These
interventions are well-known risk factors for lower airway, bladder, and blood-
stream infections in the ICU patient whose immunoparalysis is at its nadir during
the first week of admission. This is the time during which primary endogenous
infections occur. Only the immediate parenteral administration of antimicrobials
can prevent this type of infection and early therapy of an already incubating
primary endogenous infection. If the primary endogenous infection is the indica-
tion for ICU admission, parenterally administered antimicrobials are required to
treat the established infection. The philosophy that preventing infection is always
better than cure dictates the immediate systemic administration of antibiotics to a
critically ill patient requiring mechanical ventilation. Hence, systemically
administered cefotaxime is an integral part of the concept of the prophylactic
protocol of selective decontamination of the digestive tract (SDD) [43].
The criteria for parenteral administration include flora friendliness and low
resistance potential (after excretion into throat and gut via saliva, bile, and mucus)
and anti-inflammation propensity. In addition, the pharmacokinetic properties
should include a high excretion in the target organs and, in particular, in bronchial
secretions. The ratio of the antimicrobial level in the target organ and the MBC
(Table 11.4) determines the efficacy against a particular microorganism (Chap. 7).
Protein binding should be minimal. Parenterally administered antimicrobials
require a good safety profile in terms of allergy, nephro- and ototoxicity, and
influence on hemostasis. Finally, the target PPM is a criterion of paramount
importance for choosing an antimicrobial to be administered parenterally. Primary
endogenous infections are caused by both normal, e.g., Streptococcus pneumoniae,
Haemophilus influenzae, and S. aureus, and abnormal potential pathogens,
including AGNB and MRSA. General health before ICU admission influences the
carrier state (Chap. 2). Previously, healthy individuals such as trauma, burn, acute
liver, and pancreatitis patients only carry normal potential pathogens, whereas
patients with chronic underlying diseases, including chronic obstructive pulmon-
ary disease, diabetes, and alcoholism, may carry AGNB and MRSA. It is obvious
that patients referred to the ICU from other hospitals or wards are highly likely to
be carriers of abnormal potential pathogens. Taking these criteria into consider-
ation, there are only a few antimicrobials suitable for prophylaxis. The first- and
second-generation b-lactams cover the normal potential pathogens but are less
effective against the abnormal AGNB. Cefotaxime, a third-generation cephalo-
sporin, has an adequate spectrum toward both normal and abnormal AGNB, with
the exception of P. aeruginosa and Acinetobacter spp. Ceftazidime adequately
covers AGNB but at the expense of S. aureus. Cefepime may be a suitable
alternative for ceftazidime. Three randomized SDD trials used the fluoroquino-
lones ciprofloxacin and ofloxacin as systemic prophylaxis despite inadequate cover
of S. pneumoniae [4446]. Most SDD studies employed cefotaxime as the
parenterally administered antimicrobial. Many patients admitted to a medical/
surgical ICU receive perioperative prophylaxis. This surgical prophylaxis may be

replaced by cefotaxime as soon as SDD is started. In patients in whom prosthetic
material has been implanted, an appropriate endocarditis prophylaxis should also
be given. It is not uncommon that patients in a generalized inflammatory state on
ICU admission receive an aminoglycoside to control inflammation, in addition to
cefotaxime.
Colonization of the lower airways, particularly with S. aureus, MRSA,
P. aeruginosa, and Aspergillus fumigatus, may persist despite adequate systemic
therapy. Cephradine, vancomycin, polymyxin E, cefotaxime, ceftazidime, genta-
micin, tobramycin, and amphotericin B can be nebulized to achieve higher antibiotic
concentrations in the lower airways. The ventilation tube is often contaminated and
has to be changed and replaced by a new tube following nebulization. Surveillance
swabs of the oropharynx are required to monitor therapy efficacy. Chapter 22 dis-
cusses in detail the fundamental features of infection therapy:
1. sterilizing the infected internal organ using parenterally administered
antibiotic(s);
2. eliminating the source with enterally administered antibiotics;
3. removing or changing the invasive device;
4. evaluating therapeutic efficacy using surveillance samples.
11.3.2 Enterally Administered Nonabsorbable Antimicrobials
The purpose of enterally administered nonabsorbable antimicrobials is to prevent,

or if already present, eradicate abnormal flora carriage and overgrowth from throat
and gut. The abnormal carrier state always precedes secondary endogenous
infections. Whereas parenterally administered antimicrobials are required to
control primary endogenous infections, enterally administered nonabsorbable
antibiotics are used to prevent secondary endogenous infections [47]. The com-
monly used decontaminating agents are polymyxin E, tobramycin, amphotericin
B, or nystatin and vancomycin. They all fulfil the three requirements of narrow
spectrum or ecology friendliness, low resistance potential, and anti-inflammation
properties. Additionally, they are nonabsorbable in order to achieve high salivary
and fecal concentrations, andas all antimicrobialsare inactivated by fiber,
cells, and fecal material to varying extents. Tobramycin and polymyxin are min-
imally and moderately inactivated, respectively. The polyenes and vancomycin
require high oral doses due to a high inactivation rate (Chap. 8). For decontami-
nating agents to be effective, a minimal contact time of 15 min is required between
antimicrobial agent and an abnormal potential pathogen. This contact time is no
problem in the stomach and gut due to the ileus and is guaranteed by the appli-
cation of a paste or a gel in the oropharynx. High failure rates of orally admin-
istered sprays and rinses of antimicrobials have been reported (Chap. 27). Apart
from nystatin, all antimicrobials used for SDD are administered parenterally
despite their toxicity. Toxicity is a lesser problem following enterally administered
administration, even in high doses. Chapter 8 discusses the significant reductions

in AGNB, yeast, and MRSA carriage following enteral use of polymyxin/tobra-
mycin, polyenes, and vancomycin. Interestingly, among the 15 target PPMs,
P. aeruginosa, Candida spp., and MRSA are less easy to completely clear from the
throat and/or gut. It is not uncommon that surveillance swabs yield their very low
concentrations even during long-term SDD. Apparently, reducing overgrowth to
very-low-growth densities is sufficient to control infection, resistance, and trans-
mission [48, 49].
11.3.3 Topically Administered Antimicrobials
Topically administered antimicrobials are intended to prevent and, if already

present, eradicate colonization/infection of abnormal flora from wounds in which
plastic devices may be present, such as a tracheostoma and gastrostoma. Wounds, in
particular burn wounds, are prone to potential acquisition and subsequent exogenous
colonization/infection with pathogens, such as P. aeruginosa and MRSA, without
previous carriage in the digestive tract. Gels including AquaForm and IntraSite are
highly suitable for topical application of antimicrobials to burn wounds [50]. The
concentration of, for example, polymyxin E or vancomycin is in general 2%. Each
application has to be preceded by wound debridement and cleaning with a disin-
fecting agent, such as 2% taurolin, to remove necrotic tissue that may inactivate
polymyxin E or vancomycin. The transparent gels allow careful inspection of wound
healing and granulating tissue. Tracheostoma and gastrostoma are artificially created
long-term wounds kept open by plastic devices. Potential pathogens, such as
P. aeruginosa and non-aeruginosa, A. baumannii, and S. aureus, both methicillin-
sensitive and -resistant, have an intrinsic affinity for plastic. Colonization/infection
of exogenous origin is not uncommon in patients with tracheostoma and/or gas-
trostoma. As it is virtually impossible to sterilize plastic using parenterally admin-
istered agents, the devices must be removed and the wounds cleaned with taurolin.
A thin layer of a paste containing 2% polymyxin E and/or vancomycin is applied to
the stoma before a new device is put in. Obviously, preventing colonization/infection
of tracheostoma and gastrostoma using the paste twice daily throughout ICU treat-
ment is preferred to having to treat colonization/infection [51].
11.4 Efficacy in Relation to Antimicrobial Therapy Duration
Preventing primary endogenous infection is the main reason for parenteral

administration of antimicrobials, being one of the crucial components of the SDD
protocol. Supplementary prophylaxis is the second reason [52], i.e., cover while
establishing SDD on mucosal surfaces, cover for procedurally released microor-
ganisms, and elimination from mucosal surfaces of PPM resistant to enterally
administered antimicrobials (e.g., S. pneumoniae is resistant to polymyxins, ami-

noglycosides, and polyenes).
The duration of parenterally administered prophylaxis as an integral part of
SDD depends on the type of carrier state, i.e., normal versus abnormal. The
maximum period of parenterally administered prophylaxis is 5 days for patients
with abnormal flora. The oropharynx and gut (not rectal cavity) are expected to be
free from abnormal AGNB and MRSA within 3 days of enterally administered
antimicrobials. In patients who were previously healthy, a shorter period of 3 or
even 2 days has been shown to be effective [53].
According to the guidelines of the American Thoracic Society, lower airway
infections due to normal potential pathogens, including H. influenzae and S. aureus,
should be treated for 710 days, whereas episodes caused by P. aeruginosa and
Acinetobacter spp. should be treated for at least 1421 days [54]. The guidelines
are based on expert opinion. There is only one randomized controlled trial evalu-
ating whether a short course of intravenously administered antimicrobial therapy of
1 week is as effective as a prolonged treatment C2 weeks [55]. These investigators
showed that there is no evidence to support an antibiotic course exceeding 1 week.
A third approach for determining antibiotic therapy duration is the use of inflam-
mation markers, including C-reactive protein (CRP). Five prospective studies show
that a course of appropriate antimicrobials of \1 week results in significant
improvement in clinical, radiographic, and microbiological parameters [5660].
There is no evidence to support the superiority of a 2-week course of intravenously
administered antibiotics for [1 week, which prompted us to implement a paren-
terally administered antibiotic policy of 5 days, followed by careful evaluation of
patients clinical, radiographic, and microbiological parameters [61, 62].
11.4.2 Enterally Administered Antimicrobials
Enterally administered antimicrobials using a paste or gel and a suspension into

the throat and gut, respectively, is indicated as long as the patient is immuno-
paralyzed, i.e., at high risk of acquiring abnormal flora and subsequently of
developing an infection. SDD is generally given throughout the ICU treatment or,
in practice, until extubation.
There is general consensus that long-term use of enterally administered poly-
myxin E/tobramycin/amphotericin B (PTA) in the correct concentrations in throat
and gut abolishes oropharyngeal and gastric carriage, respectively, of the target
PPM within a few days. Eradication of rectal carriage appears to depend on the
presence of peristalsis and may vary until the patient produces feces [63].
11.4.3 Topically Administered Antimicrobials
Three days of topical application of PTA and/or vancomycin into wounds or

tracheostoma or gastrostoma has been shown to eradicate the potential pathogens
using daily sampling [50, 51]. Preventing colonization/infection of wounds and

stomas obviously requires topical application throughout the ICU treatment [51].
11.5 Endpoints of Antimicrobial Policies
A new antimicrobial agent launched by the pharmaceutical industry or a new

antibiotic policy implemented in the ICU should be assessed using four main
endpoints, including morbidity, mortality, antimicrobial resistance, and costs
(Table 11.5). The pharmaceutical industry is not often able to provide that
information. If the superinfection rate associated with a particular antimicrobial is
known, that figure is often diluted by the number of all patients enrolled in all
studies, whether community-, hospital-, or ICU-based, as denominator. Superin-
fection rates are invariably higher in critically ill patients requiring intensive care,
including the need for mechanical ventilation, compared with patients staying in
wards. The more severely ill the patient population studied, the higher the
superinfection rate. Mortality rate should be assessed following immediate
administration of the antimicrobial(s), as the delay of adequate therapy is asso-
ciated with increased mortality rates [64, 65]. Immediate administration of an
antimicrobial agent always implies an empirical decision in the absence of any
knowledge of the causative microorganism, although, of course, reasonable
assumptions can be made depending upon disease presentation [62]. Presence of
normal potential pathogens in admission flora is highly likely if the patient was
previously healthy (trauma, burn, acute liver failure, and pancreatitis patients).
Patients with chronic underlying diseases, including alcoholism, chronic
obstructive pulmonary disease, and diabetes, often import abnormal potential
pathogens in their flora on admission. Patients referred from other hospitals or
from wards are generally ill and may bring abnormal bacteria associated with the
hospital ecology into the ICU. Surveillance samples of throat and rectum taken at
the time of admission may help identify normal versus abnormal carrier state.
These samples are also required to assess the impact of an antibiotic on the
patients ecology and to monitor secondary or supercarriage of resistant bacteria.
This information is invariably missing for most new antibiotics or antibiotic
policies. New antibiotics are also recommended for 10 or more days and are
always more expensive than the older antimicrobials that are often out of patent
and inexpensive.
Our antimicrobial policy using the above criteria is shown in Chap. 17.
A prospective, observational, cohort study was performed for 4 years (19992003)
to assess efficacy, side-effects, and costs of this antibiotic policy. Only critically ill
children requiring 4 or more days of intensive care were included in the epide-
miological descriptive study [66]. This study group represents the sicker patients
with longer pediatric ICU stays. Approximately two thirds of pediatric ICU
admissions did not meet the 4-day stay entry criterion for the study and are thus
not included in the denominator for infection rates. Short-stay patients have a low
Table 11.5 Endpoints of antimicrobial policies
Desired endpoints
Efficacy: clinical endpoints
reduction in infectious morbidity (superinfection)
reduction in mortality following the immediate administration of empirical antimicrobial
Safety: microbiological endpoints (using surveillance samples)
impact on ecology: yeasts, Clostridium difficile (diarrhea)
antimicrobial resistance: supercarriage of aerobic Gram-negative bacilli, methicillin-resistant
Staphylococcus aureus, vancomycin-resistant enterococci
Costs
duration: an antimicrobial course of \1 week is as good as that of C2 weeks
is the newer, more expensive, antimicrobial superior than the older, less expensive one, in terms
of efficacy and safety?
risk of developing secondary endogenous and exogenous infection, and we believe

they should not be reported in the denominator of pediatric ICU infection rates.
A total of 1,241 children were enrolled in the study: 520 had infections, with an
overall infection rate of 41.9%; viral infections accounted for 14.5% and bacterial/
yeast infections for 33.0%. The incidence of blood-stream infection and lower
airway infection was 21.0 and 9.1 episodes per 1,000 patient days, respectively;
13.3% of the children were infected with a microorganism acquired in the pediatric
ICU; 4.0% of admitted patients developed infections due to resistant microor-
ganisms. The mortality rate was 9.6%.
References
1. Vollaard EJ, Clasener HAL (1994) Colonisation resistance. Antimicrob Agents Chemother
335:409414
2. Vlaspolder F, de Zeeuw G, Rozenberg-Arska M et al (1987) The influence of flucloxacillin
and amoxicillin with clavulanic acid on the aerobic flora of the alimentary tract. Infection
15:241244
3. Harbarth S, Liassine N, Dharan S et al (2000) Risk factors for persistent carriage of
methicillin-resistant Staphylococcus aureus. Clin Infect Dis 31:13801385
4. Shek FW, Stacey BSF, Rendell J et al (2000) The rise of Clostridium difficile: the effect of
length of stay, patient age and antibiotic use. J Hosp Infect 45:235237
5. Donskey CJ, Chowdry TK, Hecker MT et al (2000) Effect of antibiotic therapy on the density
of vancomycin-resistant enterococci in the stool of colonised patients. N Engl J Med
343:19251932
6. van Saene R, Fairclough S, Petros A (1998) Broad- and narrow-spectrum antibiotics: a
different approach. Clin Microbiol Infect 4:5657
7. Buck AC, Cooke EM (1969) The fate of ingested Pseudomonas aeruginosa in normal
persons. J Med Microbiol 2:521525
8. van Saene HKF, Stoutenbeek CP, Geitz JN et al (1988) Effect of amoxycillin on colonisation
resistance in human volunteers. Microb Ecol Health Dis 1:169177
9. Vollaard EJ, Clasener HAL, Janssen AJHM et al (1990) Influence of amoxycillin on

microbial colonisation resistance in healthy volunteers. A methodological study.
J Antimicrob Chemother 25:861871
10. Vollaard EJ, Clasener HAL, Janssen AJHM et al (1990) Influence of cefotaxime on microbial
colonisation resistance in healthy volunteers. J Antimicrob Chemother 26:117123
11. Vollaard EJ, Clasener HAL, Janssen AJHM (1992) Influence of cephradine on microbial
colonisation resistance in healthy volunteers. Microb Ecol Health Dis 5:147153
12. Krook A, Danielsson D, Kjellander J, Jarnerot G (1979) Changes in the fecal flora of patients
with Crohns disease during treatment with metronidazole. Scand J Gastroenterol 14:705710
13. Meijer BC, Kootstra GJ, Geertsma DG, Wilkinson MHE (1991) Effects of ceftriaxone on
faecal flora: analysis by micromorphometry. Epidemiol Infect 106:513521
14. Toltzis P, Yamashita T, Vilt L et al (1998) Antibiotic restriction does not alter endemic
colonisation with resistant Gram-negative rods in a pediatric intensive care unit. Crit Care
Med 26:18931899
15. Samonis G, Gikas A, Toloudis P et al (1994) Prospective study of the impact of broad-
spectrum antibiotics on the yeast flora of the human gut. Eur J Clin Microbiol Infect Dis
13:665667
16. Krcmery V Jr, Oravcova E, Spanik S et al (1998) Nosocomial breakthrough fungaemia
during antifungal prophylaxis or empirical anti-fungal therapy in 41 cancer patients receiving
anti-neoplastic chemotherapy: analysis of aetiology risk factors and outcome. J Antimicrob
Chemother 41:373380
17. Lode H, von der Hh N, Ziege S et al (2001) Ecological effects of linezolid versus amoxicillin/
clavulanic acid on the normal intestinal microflora. Scand J Infect Dis 33:899903
18. Cunha AC (2001) Effective antibiotic-resistance control strategies. Lancet 357:13071308
19. Lee SC, Fung CP, Liu PYF et al (1999) Nosocomial infections with ceftazidime-resistant
Pseudomonas aeruginosa: risk factors and outcome. Infect Control Hosp Epidemiol 20:
205207
20. Fink MP, Snydman DR, Niederman MS et al (1994) Treatment of severe pneumonia in
hospitalised patients: results of a multi-center, randomised, double-blind trial comparing
intravenous ciprofloxacin and imipenem-cilastatin. Antimicrob Agents Chemother 38:
547557
21. Carmeli Y, Troillet N, Eliopoulos GM, Samore MH (1999) Emergence of antibiotic-resistant
Pseudomonas aeruginosa: comparison of risks associated with different anti-pseudomonal
agents. Antimicrob Agents Chemother 43:13791382
22. Lacey RW, Lord VL, Howson GL et al (1983) Double-blind study to compare the selection
of antibiotic resistance by amoxicillin or cephradine in the commensal flora. Lancet 2:
529532
23. de Man P, Verhoeven BAN, Verburgh HA et al (2000) An antibiotic policy to prevent
emergence of resistant bacilli. Lancet 355:973978
24. Cunha BA (2000) Antibiotic resistance. Med Clin North Am 84:14071429
25. Mingeot-Leclercq MP, Glupczynski Y, Tulkens PM (1999) Aminoglycosides: activity and
resistance. Antimicrob Agents Chemother 43:727737
26. Gerding DN, Larson TA, Hughes RA et al (1991) Aminoglycoside resistance and
aminoglycoside usage: ten years of experience in one hospital. Antimicrob Agents
Chemother 35:12841290
27. Toltzis P, Dul M, ORiordan MA et al (2003) Cefepime use in a pediatric intensive care unit
reduces colonization with resistant bacilli. Pediatr Infect Dis J 22:109114
28. Gillespie T, Masterton RG (2002) Investigation into the selection frequency of resistant
mutants and the bacterial kill rate by levofloxacin and ciprofloxacin in non-mucoid
Pseudomonas aeruginosa isolates from cystic fibrosis patients. Int J Antimicrob Agents
19:377382
29. Livermore DM (2002) Multiple mechanisms of antimicrobial resistance in Pseudomonas
aeruginosa: our worst nightmare? Clin Infect Dis 34:634640
30. Dofferhoff AS, Nijland JH, de Vries-Hospers HG et al (1991) Effects of different types and
combinations of antimicrobial agents on endotoxin release from Gram-negative bacteria: an
in vitro and in vivo study. Scand J Infect Dis 23:745754
31. Jackson JJ, Kropp H (1992) Beta-lactam antibiotic-induced release of free endotoxin: in vitro
comparison of penicillin-binding protein (PBP) 2-specific imipenem and PBP 3-specific
ceftazidime. J Infect Dis 165:10331041
32. Prins JM, van Deventer SJH, Kuijper EJ, Speelman P (1994) Clinical relevance of antibiotic-
induced endotoxin release. Antimicrob Agents Chemother 38:12111218
33. Luchi M, Morrison DC, Opal S et al (2000) A comparative trial of imipenem versus
ceftazidime in the release of endotoxin and cytokine generation in patients with
Gram-negative urosepsis. J Endotoxin Res 6:2531
34. Prins JM, van Agtmael MA, Kuijper EJ et al (1995) Antibiotic-induced endotoxin release in
patients with Gram-negative urosepsis: a double-blind study comparing imipenem and
ceftazidime. J Infect Dis 172:888891
35. Holzheimer RG, Hirte JF, Reith B et al (1996) Different endotoxin release and IL-6 plasma
levels after antibiotic administration in surgical intensive care patients. J Endotoxin Res
3:261267
36. Crosby HA, Bion JF, Penn CW, Elliott TSJ (1994) Antibiotic-induced release of endotoxin
from bacteria in vitro. J Med Microbiol 40:2330
37. Artenstein AW, Cross AS (1989) Inhibition of endotoxin reactivity by aminoglycosides.
J Antimicrob Chemother 24:826828
38. Foca A, Matera G, Berlinghieri MC (1993) Inhibition of endotoxin-induced interleukin 8
release by teicoplanin in human whole blood. Eur J Clin Microbiol Infect Dis 12:940944
39. Siedlar M, Szczepanik A, Wieckiewicz J et al (1997) Vancomycin down-regulates
lipopolysaccharide-induced tumour necrosis factor alpha (TNF alpha) and TNF alpha
mRNA accumulation in human blood monocytes. Immunopharmacology 35:265271
40. Prentice HG, Hann IM, Herbrecht R et al (1997) A randomized comparison of liposomal
versus conventional amphotericin B for the treatment of pyrexia of unknown origin in
neutropenic patients. Br J Haematol 98:711718
41. Peters M, Petros A, Dixon G et al (1999) Acquired immuno-paralysis in paediatric intensive
care: prospective observational study. BMJ 319:609610
42. Kaiser AB (1986) Antimicrobial prophylaxis in surgery. N Engl J Med 315:11291138
43. Stoutenbeek CP (1989) The role of systemic antibiotic prophylaxis in infection prevention in
intensive care by SDD. Infection 17:418421
44. Lingnau W, Berger J, Javorsky F et al (1997) Selective intestinal decontamination in multiple
trauma patients: prospective, controlled trial. J Trauma 42:687694
45. Verwaest C, Verhaegen J, Ferdinande P et al (1997) Randomised, controlled trial of selective
digestive decontamination in 600 mechanically ventilated patients in a multi-disciplinary
intensive care unit. Crit Care Med 25:6371
mortality in critically ill patients. Am J Respir Crit Care Med 166:10291037
47. Baxby D, van Saene HKF, Stoutenbeek CP, Zandstra DF (1996) Selective decontamination of
the digestive tract: 13 years on, what it is and what it is not. Intensive Care Med 22:699706
48. Damjanovic V, Connolly CM, van Saene HKF et al (1993) Selective decontamination with
24:245259
50. Desai MH, Rutan RL, Heggers JP, Herndon DN (1992) Candida infection with and without
nystatin prophylaxis. Arch Surg 127:159162
51. Morar P, Makura Z, Jones A et al (2000) Topical antibiotics on tracheostoma prevents

exogenous colonisation and infection of lower airways in children. Chest 117:513518
52. Alcock SR (1990) Short-term parenteral antibiotics used as a supplement to SDD regimens.
Infection 13(Suppl 1):S14S18
53. Sirvent JM, Torres A, El-Ebiary M et al (1997) Protective effect of intravenously
administered cefuroxime against nosocomial pneumonia in patients with structural coma.
Am J Respir Crit Care Med 155:17291734
54. (No authors listed) (1996) Hospital-acquired pneumonia in adults: diagnosis, assessment of
severity, initial antimicrobial therapy, and preventative strategies. A consensus statement
November 1995. Am J Respir Crit Care Med 153:17111725
55. Chastre J, Wolff M, Fagon JY et al (2003) Comparison of 8 vs 15 days of antibiotic therapy
for ventilator-associated pneumonia in adults. A randomised trial. JAMA 290:25882598
56. ACourt CHD, Garrard CS, Crook D et al (1993) Microbiological lung surveillance in
culture. Q J Med 86:635648
57. Montravers P, Fagon JY, Chastre J et al (1993) Follow-up protected specimen brushes to
assess treatment in nosocomial pneumonia. Am Rev Respir Dis 147:3844
58. Singh N, Rogers P, Atwood CW et al (2000) Short-course empiric antibiotic therapy for
patients with pulmonary infiltrates in the intensive care unit. Am J Respir Crit Care Med
162:505511
59. Dennesen PJW, van der Ven AJAM, Kessels AGH et al (2001) Resolution of infectious
parameters after antimicrobial therapy in patients with VAP. Am J Respir Crit Care Med
163:13711375
60. Ibrahim EH, Ward S, Sherman G et al (2001) Experience with a clinical guideline for the
treatment of ventilator-associated pneumonia. Crit Care Med 29:11091115
61. Condon RE (2002) Bacterial resistance to antibiotics. Arch Surg 137:14171418
62. Torres A, Ewig S (2004) Diagnosing ventilator-associated pneumonia. N Engl J Med
350:433435
63. Ledingham I McA, Alcock SR, Eastaway AT et al (1988) Triple regimen of selective
decontamination of the digestive tract, systemic cefotaxime, and microbiological surveillance
for prevention of acquired infection in intensive care. Lancet 1:785790
64. Luna CM, Vujacich P, Niederman MS et al (1997) Impact of BAL data on the therapy and
outcome of ventilator-associated pneumonia. Chest 111:676685
65. Alvarez-Lerma F, ICU-acquired Pneumonia Study Group (1996) Modification of empiric
antibiotic treatment in patients with pneumonia acquired in the intensive care unit. Intensive
Care Med 22:387394
a prospective four year study based on knowledge of the carrier state. Crit Care Med
32:839847
Outbreaks of Infection in the ICU:
Whats up at the Beginning 12
of the Twenty-First Century?
V. Damjanovic, N. Taylor, T. Williets

12.1 Introduction
Two recent sets of publications were taken into consideration when preparing our
analysis of infectious outbreaks in the intensive care unit (ICU). The first concerns
the emergence of severe acute respiratory syndrome (SARS) and avian flu in 2003,
and a spread across the world of a novel influenza caused by SwH1N1 in 2009. These
viral infections had a major impact on intensive care and are described in Chap. 20.
This chapter is dedicated to describing outbreaks caused by bacteria and fungi, with
references to secondary infections associated with flu and SARS [1, 2]. The second
publication concerns the International Study of the Prevalence and Outcomes of
Infection in Intensive Care Units published in December 2009 [3]. Although this is
a point-prevalence study, it provides information about the global epidemiology of
Infection in ICUs. Unfortunately, it could not give insight into outbreaks of infection
in ICUs, so we searched for specific publications describing such outbreaks.
In the second (2005) edition of this book, we analysed the usefulness of
molecular techniques in selected outbreaks [4]. The majority of outbreaks occurred
in the last decade of the twentieth century. However, reports were usually published
several years later. A similar pattern was observed when we analysed outbreaks
published in the first decade of the twenty-first century: the actual outbreaks
occurred a few years earlier. Indeed, the above-mentioned point-prevalence study
was conducted on 8 May 2007 but published in December 2009 [3]. Therefore, for
accuracy, this analysis indicates when outbreaks actually happened and when they
were subsequently published. Acinetobacter outbreaks were selected to illustrate
V. Damjanovic (&)
Liverpool, UK
190 V. Damjanovic et al.
Fig. 12.1 Acinetobacter outbreaks published in 20002003 actually occurred in 19961999
this point (Fig. 12.1). In addition to the reported outbreaks, a number of publica-
tions considered many relevant aspects of infection and outbreaks in ICU. Some of
these are included in this chapter. We analysed 97 publications, the majority of
which met the definition of an outbreak in neonatal (NICU), paediatric (PICU) and
adult (AICU) ICUs and reported since 2000. The main objective of this analysis
was to find out whether there were any new features in the outbreaks of infection in
ICU at the beginning of the new century, including those influenced by new viruses.
12.2 Methods of Analysis
12.2.1 Search Strategy
We searched MEDLINE for outbreaks published between January 2000 and

September 2009. The search terms used were intensive care unit, adult ICU,
paediatric ICU, neonatal ICU and outbreaks.
12.2.2 Framework of Analysis
We used the same framework as in the second edition of this book; however,
outbreaks were not presented separately per ICU type but according to caus-
ative organisms, in the following order: methicillin-resistant Staphylococcus
aureus (MRSA), vancomycin-resistant enterococci (VRE), aerobic Gram-
negative bacilli (AGNB), Pseudomonas spp., Acinetobacter spp. and fungi,
together with the selected features searched (Table 12.1). The number of ana-
lysed outbreaks is stated, but only selected outbreaks are shown and listed in
the references.
12 Outbreaks of Infection in the ICU 191
Table 12.1 Outbreak microorganisms and features searched
Causative Features
microorganisms
Emerging Methods used Pathogenesis Prevention Endpoint
threats and control
MRSA, VRE, New Surveillance Exogenous versus Hygiene, Morbidity
AGNB, antibiotic cultures, endogenous, anti-sepsis, and mortality
Pseudomonas resistance, molecular endemic infection SDD associated
spp., SARS, techniques, versus outbreak with outbreak
Acinetobacter H1N1 statistics
spp., fungi
SDD selective digestive decontamination
12.3 MRSA Outbreaks
We retrieved reports on six outbreaks [510] published since 2000; five occurred
in AICUs and one in an animal ICU. Reports of two outbreaks were published
in 2002 and three in 2004, all occurring between 1997 and 2000. One report
published in 2007 did not report the actual time of the outbreak. These outbreaks
are summarised briefly according to their countries of origin. A paper from Italy
published in 2002 reported a unique experience of controlling a MRSA outbreak of
8 months duration in a medical/surgical AICU in 1998 using enterally adminis-
tered vancomycin in mechanically ventilated patients [5]. Another report from
Italy, published in 2004, described the identification of a variant of the Rome
clone of MRSA responsible for an outbreak in a cardiac surgery ICU, which
occurred in 1999 in a hospital in Rome. This strain had decreased sensitivity to
vancomycin and was resistant to many antibiotics [6]. A study from Germany
published in 2002 described the occurrence of MRSA in ICU in terms of endemic
and epidemic infections followed from January 1997 to June 2000.This study
involved 139 ICUs, 51 of which (37%) had MRSA infections. Outbreaks (three or
more MRSA infections within 3 months) were registered in 13 ICUs, clusters (two
MRSA infections within 3 months) in further 12 units and single events in 26 [7].
A publication from Spain showed that enterally administered vancomycin can
control endemic MRSA in ICUs without promoting VRE. This study was carried
out over a 49-month period from July 1996 to 2000 and published in 2004 [8].
In 2007, a report from Canada presented a recent outbreak of MRSA carriage in an
animal ICU. This finding appears important, as the strain responsible for the
animal outbreak was indistinguishable from a strain in humans commonly isolated
in Canada and the USA. Infection control measures, including active surveillance
of all animals in the ICU, were used to control the outbreak. As transmission of
MRSA within the unit occurred without infections and did not persist for a pro-
longed period of time, staff screening was surprisingly not initiated [9]. A paper
from China published in 2004 described an MRSA outbreak due to an increased
acquisition rate in ICU associated with an outbreak of SARS, which occurred in

2003. From 12 March to 31 May, only patients with SARS were admitted to the
22-bed unit. During this period, infection control precautions were upgraded,
which included wearing gloves and gowns at all times. However, data suggested
that MRSA transmission might be unexpectedly increased if gloves and gowns
were worn all the time [10].
12.4 Enterococcal Outbreaks
There have been ten outbreaks in AICUs published since 2000: eight were caused
by VRE, one was sensitive to vancomycin and one was sensitive to vancomycin
but resistant to linezolid. We selected seven reports and summarised them
according to the countries of origin and time of events and publishing.
A paper from Pakistan published in 2002 was the countrys first experience with
a vancomycin resistant Enterococcus faecium outbreak in the ICU and NICU. The
outbreak occurred in 2002, lasted 1 month and all but one isolate was of a single
clone [11]. All isolates were resistant to gentamicin, ampicillin and tetracycline
but sensitive to chloramphenicol. Six patients were colonised and four infected,
with positive blood cultures; two of each died before specific therapy could be
started (50% mortality rate). In 2005, a report from Italy described an outbreak of
VRE colonisation and infection in an ICU that lasted 16 months (20012002) [12].
Fifty-six patients were colonised by E. faecium, and E. faecalis was detected in
only two cases. Because of the low pathogenicity of VRE, the authors questioned
whether it was worthwhile to have a specific VRE surveillance programme. For
the 2004 Lowbury lecture, Pearman reported the Australian experience with VRE,
which he described as from disaster to ongoing control. This was the first
outbreak of VRE, which was caused by E. faecium in an ICU and hospital
wards and lasted 5 months in 2001. A vigilant VRE control programme
prevented the epidemic strain from becoming endemic in the hospital [13].
An outbreak due to glycopeptide-resistant enterococci (GRE) in an ICU with
simultaneous circulation of two different clones was reported from France in 2008.
The outbreak lasted several months in 2003 without infections, but the significant
colonisation caused organisational problems in the ICU [14]. An outbreak of
VRE in an ICU was reported from China in 2009. The outbreak was caused by
E. faecium and lasted 11 months (20062007). A detailed molecular analysis
showed that genetically unrelated isolates had transferred vancomycin resistance
by conjugation [15]. A paper from Korea reported an outbreak of VRE in a
neurological ICU. VRE was mainly isolated from urine specimens associated with
the presence of a Foley catheter. Of 52 patients colonised with VRE, only two had
active infection [16]. In 2009, a report from Spain presented an outbreak of lin-
ezolid-resistant E. faecalis in an ICU and reanimation unit [17]. This was the first
report of a clonal outbreak of linezolid-resistant E. faecalis in Spain. The strain
was sensitive to imipenem, vancomycin, teicoplanin and rifampicin. Most patients
were exposed to linezolid within a year (20052006). The use of linezolid began in
2002. The increase in its use continued until 2005 when a mutant was identified by
molecular analysis.
12.5 AGNB Outbreaks
Fourteen reports on outbreaks were retrieved since 2000. Eight were caused
by Klebsiella pneumoniae, four by Serratia marcescens, one by Enterobacter
cloacae and one by simultaneous infection of E. cloacae and S. marcescens.
Three Klebsiella, three Serratia and the remaining two were selected for analysis.
We discuss Pseudomonas and Acinetobacter outbreaks separately.
12.5.1 Klebsiella Outbreaks
An outbreak of Klebsiella infection in NICU and PICU was published from

Spain in 2004; this outbreak occurred in 20022003 and lasted 1 year [18]. The
outbreak was polyclonal. Two predominant clones of Klebsiella harboured a
special gene (SHV5) for the beta-lactamase enzyme responsible for multi-drug-
resistant Klebsiella. According to the authors, this type of Klebsiella was not
reported previously in Spain. Another clone harbouring two different genes
responsible for multidrug resistance but dissimilar from the above was reported.
A report from The Netherlands published in 2001 described an outbreak of
infections with a multi-drug-resistant Klebsiella strain [19] associated with
contaminated roll boards in operating rooms. This outbreak in 2000 showed
how an unusual source of the outbreak can be revealed by systematic sur-
veillance. In 2008, a polyclonal outbreak of extended spectrum beta-lactamase
(ESBL)-producing K. pneumoniae in an ICU of a university hospital in
Belgium was reported [20]. This was a 2-month outbreak that occurred in 2005
with 18 isolates. There was one predominant clone, two clones with several
isolates and four with unique isolates. The cause of the outbreak was not clear
but was associated with a dramatic increase in the number of imported carriers
during the previous weeks.
12.5.2 Enterobacter cloacae Outbreaks
An outbreak caused by ESBL-producing E. cloacae in a cardiothoracic ICU was

reported from Spain in 2007 [21]. The outbreak occurred in 2005, lasted 3 months,
and involved seven patients. Molecular analysis revealed two clones responsible
for the outbreak: one carried a single ESBL; the other carried two ESBLs.
Both clones showed resistance to quinolones and aminoglycosides. The outbreak
was brought under control by the implementation of barrier measures and
cephalosporin restrictions.
12.5.3 Serratia marcescens Outbreaks
An outbreak was reported from Germany in 2002 [22] in both the NICU and PICU,
lasted from September to November 1998 and involved 15 patients. Two epidemic
strains were associated with cross-infection in groups of five and ten patients,
respectively. Two epidemic clones were detected from the surfaces of an ICU room,
but an original source was not identified. The outbreak was stopped by routine
infection-control measures. A report from Malaysia in 2004 described an outbreak of
Serratia infections that lasted 10 days in an AICU [23]. The single outbreak strain
was found in insulin and sedative solutions administered to patients. An outbreak of
S. marcescens colonisation and infection in a neurological ICU that occurred from
May 2002 to March 2003 was reported from a Dutch university medical centre in
2006 [24]. The outbreak strain was traced to a healthcare worker (HCW) with long-
term carriage on the hands. The skin of the HCWs hands was psoriatic. The epidemic
ended after the colonised HCW went on leave, with subsequent eradication treatment.
A heterogeneous outbreak of E. cloacae and S. marcescens infections in a surgical
ICU was published by a group of authors from San Francisco, USA [25]. The outbreak
lasted from December 1997 through January 1998. Molecular techniques ruled out a
point source or significant cross-contamination as modes of transmission. The authors
concluded that patient-related factors, such as respiratory tract colonisation and
duration of central line placement might have played a role in this outbreak.
12.5.4 Pseudomonas Outbreaks
Several reports have been published on infections caused by multi-drug-resistant

Pseudomonas spp. in ICUs since 2000. We retrieved 19 reports; not all were
outbreaks, as some were described as endemic infections. In addition, one out-
break was caused by Burkholderia cepacia. We selected a few outbreaks that we
believed would represent the main problems occurring in ICUs, such as multidrug
resistance, clonality, transmission source and mode and infection severity.
In 2000, a publication from Norway reported an outbreak of multi-drug-resistant
P. aeruginosa associated with increased risk of death [26]. The outbreak occurred
from December 1999 to September 2000, was monoclonal and the strain was
introduced into the ICU early in 1998 and was maintained thereafter. All patients
were ventilated. The strain was resistant to carbapenems, quinolones and azlocillin.
In 13 infected patients, ten of whom died, Pseudomonas was found in one or all
specimens, such as respiratory secretions, ventilator tubes, connection tubes and the
water catcher of the ventilator system. The bacterium was also isolated from water
taps. In addition to enhanced control of infection measures, complete elimination of
the outbreak was achieved after water taps were pasteurised and sterile water was
used when a solvent was needed. In 2003, French authors published a report on the
epidemiology of P. aeruginosa in an ICU [27]. Although between 1996 and 1997 the
prevalence of P. aeruginosa infections reached 30% of all hospital-acquired
infections, the authors did not call this an outbreak, despite the fact that this was
twice the national prevalence of 15% observed in ICUs. However, this high prev-
alence prompted the authors to conduct a prospective epidemiological study from
July 1997 to February 1998. We selected this study as a good example of activities
necessary to prevent a major outbreak. The authors described how systematic sur-
veillance was carried out (oropharyngeal and rectal swabs on admission and twice
weekly afterwards). This practice revealed that during the study period, the overall
incidence of P. aeruginosa carriage was 43%: 17% on admission and 26% acquired
in the ICU. In addition 16/191 (8%) patients developed the infection. The authors
also pointed out that intestinal carriage was a prerequisite for colonisation or
infection. Genotyping analysis of 81 isolates indicated that 70% belonged to geno-
type 1, 4% to genotype 2 and that remaining isolates were not genetically related. It
has also been shown that mechanical ventilation was associated with P. aeruginosa
carriage and ineffective antibiotics significantly increased the risk of colonisation
and infection in ICU. The authors concluded that not only do endogenous sources
account for the majority of colonisation or infection due to P. aeruginosa but that
exogenous sources may be involved in some instances. In an epidemic setting, the
authors stance was to reinforce standard barrier precautions. However, the main
message of this study is the necessity to adopt and pursue preventive measures.
In 2008, an outbreak of severe B. cepacia infections in an ICU was reported from
Spain [28]. The outbreak occurred over a period of 18 days in August 2006 when
B. cepacia were recovered from different clinical samples associated with bacter-
aemia in three cases, lower respiratory tract infection in one and urinary tract
infection in one. Samples of antiseptics, eau de cologne and moisturising milk
available on treatment carts were collected and cultured. B. cepacia was isolated not
only from three samples of the moisturising body milk that had been applied to the
patients but also from two new hermetically closed units. All strains recovered from
environmental and clinical samples belonged to the same clone. The cream was
withdrawn from all hospital units, and no new cases of B. cepacia developed. The
authors concluded that the presence of bacteria in cosmetic products, even within
accepted limits, may lead to severe life-threatening infections in severely ill patients.
12.5.5 Acinetobacter Outbreaks
We retrieved 34 publications on Acinetobacter outbreaks, 11 of which were not

strictly outbreaks, and actually not reported as such, but rather described general
epidemiology, antibiotic resistance, infection control or treatment options. Most of
these problems are dealt with in relevant chapters of this edition. Following our
approach, we summarise only a few outbreaks, which appeared to offer some new
findings or insights.
A 2000 report from Italy described an outbreak of infusion-related A. baumannii
bacteraemia in an eight-bed ICU [29]. From 6 June to 15 July 2000, six cases were
identified. All patients received parenterally administered solutions prepared by ICU
nurses, which was subsequently proven to be the source of infection. Three patients
died from sepsis despite treatment with a combination of meropenem and amikacin,
which were shown by laboratory tests to be synergistic. This high mortality rate (50%)
was explained by the authors as being due to persistent bacteraemia related to the
repeated infusions of contaminated solutions. Once aseptic preparation was carried out
in the hospital pharmacy, this outbreak was controlled, and further infusion-related
nosocomial bacteraemia was prevented. From the USA, a publication in 2001 reported
an outbreak of multiresistant Acinetobacter colonisation and infection in an ICU [30].
The strain was sensitive only to polymyxin. The outbreak lasted an entire year between
1996 and 1997 and involved 57 patients, 27 of whom were infected and 25 colonised.
The arrival of a colonised burn patient ([50% total body surface area) from an outside
hospital was responsible for the outbreak. Although on typing two strains were found,
the only identified primary source was the original burn patient. Ten deaths resulted
from infections (37% of infected patients). The authors claimed that this outbreak
served as a model of eradication of multi-drug-resistant organisms, as the
Acinetobacter was eliminated from all ICU patients by multidisciplinary measures that
included the following: cohort and contact isolation of all colonised and infected
patients; introduction of strict aseptic measures such as hand washing, barrier isolation,
equipment and room cleaning; sterilisation of ventilator equipment; and individual
dedication of medical equipment to each patient. A paper was published from Australia
in 2007 regarding carbapenem-resistant A. baumannii [31]. We selected this publi-
cation as an illustration of an extensive molecular analysis rather than for a critical
review of the outbreak, which occurred in an ICU between 1999 and 2000. Based on
their findings, the authors claim that antibiotic-resistant genes are readily exchanged
between co-circulating strains in epidemics of phenotypically indistinguishable
organisms. In conclusion, they recommend that epidemiological investigation of major
outbreaks should include whole-genome typing as well as analysis of potentially
transmissible genes and their vehicles. Finally, we found a paper in a journal from
Kuwait not found by our Internet research [32]. The authors reported three different
outbreaks of multi-drug-resistant A. baumannii infections involving 24 patients aged
1675 years that occurred in an ICU in the course of 1 year between 2006 and 2007.
The outbreak was polyclonal and successfully controlled with tigecycline, to which
two causative clones were sensitive. Three additional distinct clones were isolated
from the environment. Due to lack of appropriate surveillance cultures, no explanation
was offered for the origin of epidemic clones. Subsequently, in a letter to the editor, our
interpretation that microbial gut overgrowth increased spontaneous mutation,
which led to polyclonality and antibiotic resistance in the critically ill was accepted by
the authors [33, 34].
12.6 Fungal Outbreaks
Thirteen publications were retrieved from MEDLINE, five of which described

outbreaks of remarkable findings. The remaining papers reported some important
aspects of fungal species, colonisation, infection and treatments, predominantly as
surveys, and as such were not included in our analysis.
Outbreaks presented here were caused by uncommon opportunistic fungi. Two

reports described ICU outbreaks caused by Hansenula anomala, an opportunistic
yeast first reported from a Liverpool, UK, NICU in 1986 [35]. In 2001, a report
from Croatia described an outbreak in a surgical ICU [36]. H. anomala was
isolated from blood taken from eight patients between 23 August and 6 December
1993. All patients were treated with antifungal therapy; three died from compli-
cations of underlying disease. The introduction of strict hygienic measures stopped
the spread of infection, but the outbreak ceased with the introduction of a new
batch of cotton from another manufacturer, which was used for venipuncture-site
disinfection. However, the authors could not find evidence for infection source and
transmission route. The second report, from Brazil (2005), describes an outbreak in
a PICU [37]. The authors reported their finding as an outbreak of Pichia anomala,
a newly introduced name for H. anomala. From October 2002 to January 2004, 17
children developed P. anomala fungemia. The median age was 1.1 year, and
the main underlying conditions were congenital malformations and neoplastic
disease. The overall mortality rate was 41.2% despite treatment with amphotericin B.
During a 2-week period in April 2003, when new cases occurred, surveillance cul-
tures revealed that 67.9% of patients were colonised with yeasts, but no single patient
was found to be colonised with P. anomala. Thus, no source was found at that time.
The outbreak was not controlled until orally administered prophylaxis with nystatin
and topical application of an iodoform to venipuncture sites were started.
An extraordinary outbreak of invasive gastritis caused by Rhizopus microsporus
in an adult ICU was reported from Spain in 2004 [38]. Over a 14-week period
(between November 1995 and March 1996), gastric mucormycosis was diagnosed
in five patients, four of whom were admitted to ICU with severe community-
acquired pneumonia and one with multiple trauma. The main symptom was
upper gastrointestinal haemorrhage. Isolated filamentous fungi were identified as
R. microsporus var. rhizopodiformis and were detected in gastric aspiration sam-
ples and traced to wooden tongue depressors used to prepare medication for oral
administration (and given to patients through a nasogastric catheter) and in some
tongue depressors stored in unopened boxes unexposed to the ICU environment.
The outbreak was terminated when contaminated tongue depressors were with-
drawn from use. This outbreak was attributable to the 40% mortality rate; wooden
material should not be used in the hospital setting.
In 2004, an outbreak of three cases of Dipodascus capitatus infection in an
ICU was reported from Japan [39]. The index case was pulmonary infection
with a fulminant course of fungal infection, which resulted in death, in a
patient with acute myelocytic leukaemia who shared a room for at least 1 week
with the two other patients, suggesting the possibility of transmission. One of
the other two patients died from multiple organ dysfunction. The presence
of D. capitatus might have been due to contamination in the respiratory ICU.
In all cases, D. capitatus was identified in sputum, deep tracheal aspiration
samples, blood and urine samples. The authors concluded that D. capitatus
should be added to the lengthening list of opportunistic fungal pathogens that
can cause infection in immune-compromised patients, with the danger of

transmission and potential outbreak.
An outbreak of Saccharomyces cerevisiae fungemia in an ICU was reported
from Spain in 2005 [40]. During the period from 15 to 30 April, three patients
with S. cerevisiae fungemia were identified. The only identified risk factor was
treatment with a probiotic containing this yeast. The three patients received the
product via nasogastric tube for a mean of 8.5 days before the culture was positive.
Surveillance cultures for the control patients admitted at the same time did not
reveal any carriers. All three patients died from causes unrelated to S. cerevisiae.
Discontinuation of use of the product for treatment or prevention of Clostridium
difficile-associated diarrhoea in the unit stopped the outbreak of infection.
In conclusion, the authors warned that the use of S. cerevisiae should be carefully
reassessed in immune-compromised or critically ill patients.
12.7 Discussion
An outbreak is defined as an event where two or more patients in a defined location

are infected by identical, often multi-drug-resistant, microorganisms transmitted via
the hands of HCW, usually within an arbitrary time period of 2 weeks. There are two
different types of infection involved in outbreaks: secondary endogenous and
exogenous. Outbreaks of secondary endogenous infections are invariably preceded
by outbreaks of carriage of abnormal flora, whereas outbreaks of exogenous
infections are not preceded by outbreaks of abnormal carriage. These two types of
outbreaks each require a different type of management: enterally and topically
administered antimicrobials for secondary endogenous and exogenous outbreaks,
respectively. Ongoing surveillance efforts, i.e. throat and rectal swabs on admission
and twice weekly thereafter, to monitor the efficacy of systematic decontamination
of the digestive tract (SDD) and to identify the emergence of antimicrobial resistant
threats, is an intrinsic component of any decontamination programme. In this sense,
a well-designed programme contains an intrinsic degree of protection against
antibiotic-resistant organism emergence. Surveillance cultures of throat and
rectum are more sensitive in detecting resistance than are diagnostic samples [41].
Additionally, there is a close relationship between surveillance and diagnostic
samples. Once a resistant microorganism reaches overgrowth concentrations,
i.e. C105/ml saliva and/or gram of faeces, diagnostic samples become positive [8].
In our review, 28 outbreaks were selected to illustrate the situation at the
beginning of this century. As a matter of fact, the majority of the outbreaks was
related to the previous decade. However, biased or not, our analysis described 19
outbreaks that occurred after 2000 and nine from last century, although the out-
breaks were published in this century (Fig. 12.1). This suggests that some new
problems indeed emerged in this century.
It is important to record the number of papers retrieved according the causative
organisms: MRSA six, VRE ten, AGNB 14, Pseudomonas spp. 19, Acinetobacter
spp. 23 and fungi 13. Perhaps, against our expectation, AGNB organismsin
Table 12.2 New and older trends at the beginning of the twenty-first century
New trends Older trends

Emerging viral infections may increase Surveillance cultures mostly used after
bacterial and fungal outbreaks outbreaks occurred
Extensive use of molecular techniques proved Pathogenesis of outbreaks rarely clarified due to
that many outbreaks are polyclonal and lack of surveillance cultures
detected new genes responsible for antibiotic SDD still rarely used for control of outbreaks
resistance In general, endemic infections more common
than outbreaks
Emergence of new resistant clones Infection control measures usually enhanced
after outbreak occurred
The principle of SDD extended to other Mortality primarily attributed to underlying
antibiotics, e.g. vancomycin to prevent MRSA disease, with exception of NICU and direct
outbreaks injection of pathogen
SDD selective decontamination of the digestive tract; MRSA methicillin-resistant Staphylococcus
aureus; NICU neonatal intensive care unit
particular, opportunists such as Pseudomonas and Acinetobacterprevailed

significantly, for which there must be a reason. If we take MRSA as an example,
all around the world, this drug-resistant pathogen has been a primary focus
for nosocomial infection control and treatment for years. Thus, there are fewer
outbreaks. An extensive study from Germany that involved 139 ICUs showed that
cluster and single MRSA infections were significantly more common than actual
outbreaks (38 ICUs compared with 12, respectively) [7]. To our knowledge, there
were no similar studies for VRE and AGNB, but one would anticipate similar
findings and interpretation.
On the other hand, opportunistic pathogens such as Pseudomonas spp.,
Acinetobacter spp. and fungi often caused unexpected outbreaks, particularly in
immunocompromised patients. They originated from external sources and were
difficult to treat because of their resistance to multiple antibiotics.
Our search for specific features relevant to published outbreaks revealed some
new, and confirmed some older, trends (Table 12.2). Probably the best example of
how new viral infectionssuch as SARScan change the rate of bacterial and
fungal infections in ICUs came from the experience in China [10]. There was a
significant increase in the rate of MRSA and Candida spp. acquisition in an ICU
during the SARS period. It may be anticipated, therefore, that in the future, SARS
and influenza viral infections would lead to complex ICU outbreaks.
We pointed out earlier how using molecular techniques revealed that many
outbreaks were due to more than one clone [4]. Our analysis confirms this,
although the origin of different clones remained obscure in all reports in which
polyclonality was detected. However, we recently put forward a hypothesis that
microbial gut overgrowth is responsible for increased spontaneous mutation
leading to polyclonality and antibiotic resistance [42]. Furthermore, extensive use
of molecular techniques not only revealed a number of new genes responsible for
antibiotic resistance [18] but showed that genetically unrelated organisms readily
exchange antibiotic resistance genes [15, 31]. Yet further, a new trend is related to
the SDD concept. Two studies, one from Italy and one from Spain, reported the
use of enterally administered vancomycin to control and prevent, respectively,
MRSA outbreaks [5, 8]. This is further evidence that the principle of SDD can be
used with antimicrobials directed specifically to the causative organism. As early
as 1993 we reported how selective decontamination with nystatin successfully
controlled a Candida outbreak in an NICU [43].
Among older trends, surveillance cultures, or lack of them, are still prominent.
Even in 2009 there were authors responsible for infection control in hospitals and
ICUs who claimed that surveillance cultures of all patients with potential to
develop infection are difficult and very costly [44]. Some time ago (1994), we
expressed an alternative view in response to an identical attitude [45]. Needless to
say, lack of surveillance cultures not only delays the recognition of an outbreak
and its control but also precludes the understanding of the pathogenesis of
the majority of outbreaks. Surveillance cultures are also crucial for detecting
outbreaks of exogenous pathogenesis, i.e. without carriage. On the other hand, the
source of an exogenous outbreak is readily identified with molecular techniques.
Some of these outbreaks are striking, such as one from this analysis in which
Acinetobacter-contaminated parentally administered solutions were repeatedly
infused to patients, leading to a very high mortality rate of 50% [29].
In conclusion, new trends as well as old confirm what we indicated in the
previous edition of this book, which is that to control and prevent ICU outbreaks,
surveillance cultures and SDD should be integrated in routine infection-control
measures.
References
1. El-Masri MM, Williams KM, Fox-Wasylyshyn SM (2004) Severe acute respiratory
syndrome: another challenge for critical care nurses. AACN Clin Issues 15:150159
2. Burns SM (2009) H1N1 influenza is here. J Hosp Infect 73:200202
3. Vincent JL, Rello J, Marshall J et al (2009) International study of the prevalence and outcome
of infection in intensive care units. JAMA 302:23232329
4. Damjanovic V, Corbella X, van der Spoel JI, van Saene HKF (2005) Outbreaks of infection
in intensive care unitsusefulness of molecular techniques for outbreak analysis. In: van
Saene HKF, Silvestri L, de la Cal MA (eds) Infection control in the intensive care unit, 2nd
edn. Springer, Milan, pp 247296
Control 30:391399
6. Cassone M, Campanile F, Pantosti A et al (2004) Identification of a variant Rome clone of
methicillin-resistant Staphylococcus aureus with decreased susceptibility to vancomycin,
responsible for an outbreak in an intensive care unit. Microb Drug Res 10:4349
7. Gastmeier P, Sohr D, Geffers C et al (2002) Occurrence of methicillin-resistant
Staphylococcus aureus infections in German intensive care units. Infection 4:198202
9. Weese JS, Faires M, Rousseau J et al (2007) Cluster of methicillin-resistant Staphylococcus
aureus colonisation in a small animal intensive care unit. JAMA 231:13611364
10. Yap FH, Gomersall CD, Fung KS et al (2004) Increase in methicillin-resistant
Staphylococcus aureus acquisition rate and change in pathogen pattern associated with an
outbreak of severe acute respiratory syndrome. Clin Infect Dis 39:511516
11. Khan E, Sarwari A, Hasan R et al (2002) Emergence of vancomycin-resistant Enterococcus
faecium at a tertiary care hospital in Karachi, Pakistan. J Hosp Infect 52:292296
12. Peta M, Carretto E, Bdbarini D et al (2006) Outbreak of vancomycin-resistant Enterococcus
spp. on an Italian general intensive care unit. Clin Microbiol Infect 12:163169
13. Pearman JW (2006) 2004 Lowbury lecture: the Western Australian experience with
vancomycin-resistant enterococcifrom disaster to ongoing control. J Hosp Infect 63:1426
14. Delamare C, Lameloise V, Lozniewski A et al (2008) Glycopeptide-resistant Enterococcus
outbreak in an ICU with simultaneous circulation of two different clones. Pathol Biol 56:454460
15. Zhu X, Zeng B, Wang S et al (2009) Molecular characterization of outbreak-related strains of
vancomycin-resistant Enterococcus faecium from an intensive care unit in Beijing, China.
16. Se BY, Chun HJ, Yi HJ et al (2009) Incidence and risk factors of infections caused by
vancomycin-resistant Enterococcus colonization in neurosurgical intensive care unit patients.
J Korean Neurosurg Soc 46:123129
17. Gomez-Gil R, Romero-Gomez MP, Garcia-Arias A et al (2009) Nosocomial outbreak of
linezolid-resistant Enterococcus faecalis infection in a tertiary care hospital. Diag Microb
18. Brinas L, Lantero M, Zarazaga M et al (2004) Outbreak of SHV-5 beta-lactamase-producing
Klebsiella pneumoniae in a neonatalpediatric intensive care unit in Spain. Microb Drug Res
10:354358
19. vant Veen A, van der Zee A, Nelson J et al (2005) Outbreak of infection with a multiresistant
Klebsiella pneumoniae strain associated with contaminated roll boards in operating rooms.
J Clin Microbiol 43:49614967
20. Laurent C, Rodriguez-Villalobos H, Rost F et al (2008) Intensive care unit outbreak of
extended-spectrum beta-lactamase-producing Klebsiella pneumoniae controlled by cohorting
patients and reinforcing infection control measures. Infect Control Hosp Epidemiol 29:
517524
21. Manzur A, Tubau F, Pujol M et al (2007) Nosocomial outbreak due to extended-spectrum-
beta-lactamase-producing Enterobacter cloacae in a cardiothoracic intensive care unit. J Clin
Microbiol 45:23652369
22. Steppberger K, Walter S, Claros MC et al (2002) Nosocomial neonatal outbreak of Serratia
marcescensanalysis of pathogens by pulsed field gel electrophoresis and polymerase chain
reaction. Infection 30:277281
23. Alfizah H, Nordiah AJ, Rozaidi WS (2004) Using pulsed-field gel electrophoresis in the
molecular investigation of an outbreak of Serratia marcescens infection in an intensive care
unit. Singap Med J 45:214218
24. de Vries JJ, Bass WH, van der Ploeg K et al (2006) Outbreak of Serratia marcescens
colonization and infection traced to a healthcare worker with long-term carriage on the hands.
Infect Control Hosp Epidemiol 27:11531158
25. Dorsey G, Borneo HT, Sun SJ et al (2000) A heterogeneous outbreak of Enterobacter
cloacae and Serratia marcescens infections in a surgical intensive care unit. Infect Control
26. Bukholm G, Tannaes T, Kjelsberg AB et al (2002) An outbreak of multidrug-resistant
Pseudomonas aeruginosa associated with increased risk of patient death in an intensive care
unit. Infect Control Hosp Epidemiol 23:441446
27. Thuong M, Arvaniti K, Ruimy R et al (2003) Epidemiology of Pseudomonas aeruginosa and

risk factors for carriage acquisition in an intensive care unit. J Hosp Infect 53:274282
28. Alvarez-Lerma F, Maull E, Terradas R et al (2008) Mosturizing body milk as a reservoir of
Burkholderia cepacia: outbreak of nosocomial infection in a multidisciplinary intensive care
unit. Crit Care 12:R10
29. Menichetti F, Tascini C, Ferranti S et al (2000) Clinical and molecular epidemiology of an
outbreak of infusion-related Acinetobacter baumannii bacteremia in an intensive care unit. Le
Infezioni Medicina 1:2429
30. Podnos YD, Cinat ME, Wilson SE et al (2001) Eradication of multi-drug resistant
Acinetobacter from an intensive care unit. Surg Infect 2:297301
31. Valenzuela JK, Thomas L, Partridge SR et al (2007) Hospital gene transfer in a polyclonal
outbreak of carbapenem-resistant Acinetobacter baumannii. J Clin Microbiol 45:453460
32. Jamal W, Salama M, Dehrab N et al (2009) Role of tigecycline in the control of carbapenem-
resistant Acinetobacter baumannii outbreak in an intensive care unit. J Hosp Infect 72:
234242
33. Damjanovic V, Taylor N, van Saene HKF (2009) Origin of epidemic clones of Acinetobacter
in the critically ill. J Hosp Infect 73:285286
34. Rotimi VO, Jamal W, Salama M (2009) Control of Acinetobacter outbreaks in the intensive
care unit. J Hosp Infect 73:286287
35. Murphy N, Damjanovic V, Hart CA et al (1986) Infection and colonisation of neonates by
Hansenula anomala. Lancet 1:291293
36. Kalenic S, Jandrlic M, Vegar V et al (2001) Hansenula anomala outbreak at a surgical
intensive care unit: a search for risk factors. Eur J Epidemiol 17:491496
37. Pasqualotto AC, Sukiennik TC, Severo LC et al (2005) An outbreak of Pichia anomala
fungaemia in a Brazilian pediatric intensive care unit. Infect Control Hosp Epidemiol
26:553558
38. Maravi-Poma E, Rodriguez-Tudela JL, de Jalon JG et al (2004) Outbreak of gastric
mucormycosis associated with the use of wooden tongue depressors in critically ill patients.
39. Eroz G, Otag F, Erturan Z et al (2004) An outbreak of Dipodascus capitatus infection in the
ICU: three case reports and review of the literature. Jpn J Infect Dis 57:248252
40. Munoz P, Bouza E, Cuenca-Estrella M et al (2005) Saccharomyces cerevisiae fungemia: an
emerging infectious disease. Clin Infect Dis 40:16251634
41. DAgata EMC, Venkataraman L, De Girolami P et al (1999) Colonization with broad-
spectrum cephalosporin-resistant Gram-negative bacilli in intensive care units during a
nonoutbreak period: prevalence, risk factors, and rate of infection. Crit Care Med 27:
10901095
42. van Saene HK, Taylor N, Damjanovic V et al (2008) Microbial gut overgrowth guarantees
increased spontaneous mutation leading to polyclonality and antibiotic resistance in the
critically ill. Curr Drug Targets 9:419421
24:245259
44. Miranda LN, van der Heijden IM, Costa SF et al (2009) Candida colonisation as a source of
candaemia. J Hosp Infect 72:916
45. Damjanovic V, van Saene HKF, Weindling AM et al (1994) The multiple value of
surveillance cultures: an alternative view. J Hosp Infect 28:7175
Preventing Infection Using Selective
Decontamination of the Digestive Tract 13
and D. F. Zandstra
13.1 Introduction
Selective decontamination of the digestive tract (SDD) is an antimicrobial prophylaxis

designed to prevent or minimize endogenous and exogenous infections in critically
ill patients. The purpose of SDD is to preventor eradicate if initially present
the oropharyngeal and intestinal abnormal carrier state of potentially pathogenic
microorganisms (PPMs), mainly aerobic Gram-negative microorganisms, but also
methicillin-sensitive Staphylococcus aureus (MSSA), and yeasts, leaving the indige-
nous flora predominately undisturbed. The practice of SDD has four fundamental
features, termed the classic Stoutenbeeks tetralogy [1, 2] (Fig. 13.1, Table 13.1):
1. parenteral antibiotics given immediately on admission for 4 days to control
primary endogenous infections due to PPMs already present in the admission
flora;
2. enteral antimicrobials [polymyxin E, tobramycin, and amphotericin B (PTA)]
given throughout treatment in the intensive care unit (ICU) to control secondary
carriage and subsequent endogenous infections due to PPMs acquired in the unit;
3. health care workers hand hygiene throughout treatment in the ICU to control
exogenous infections due to transmission of ICU-associated microorganisms;
4. surveillance cultures of patients throat and rectum on admission and twice
weekly to monitor the efficacy of the maneuver.
SDD selectively targets the 15 PPMs and the high-level pathogens, such as
Streptococcus pyogenes. By design, SDD does not cover low-level pathogens,
including anaerobes, viridans streptococci, enterococci and coagulase-negative
L. Silvestri (&)
The four Types of Definitions of infection

components infection according to the
of SDD prevented criterion of carriage
by SDD
Caused by normal and

abnormal PPMs carried by
Parenteral Primary
patients in throat and/or gut
antibiotic endogenous
on admission to the ICU
Caused by abnormal PPMs

not carried by patients in
throat and/or gut on ICU
admission. PPM is acquired
Enteral Secondary during ICU stay, causing
antimicrobials endogenous secondary carriage
To control the
efficacy of enteral
antimicrobials
Causative abnormal PPM is

not carried by the patients
digestive tract and is
Hygiene Exogenous introduced directly into the
sterile internal organ
Surveillance
To classify infections
cultures according to the carrier state
To identify a resistance
problem
Fig. 13.1 Stoutenbeeks tetralogy of SDD and type of infection prevented/controlled by each
component. SDD selective decontamination of the digestive tract; PPM potentially pathogenic
microorganism; ICU intensive care unit
13 Preventing Infection Using Selective Decontamination of the Digestive Tract 205
Table 13.1 The full four-component protocol of selective decontamination of the digestive tract
Target PPMs and antimicrobials Total daily dose (divided 4 9 daily)
\5 years 512 years [12 years

1. Parenteral antimicrobials: normal PPMs: 150/kg 200/kg 4,000
cefotaxime (mg)
2. Enteral antimicrobials: abnormal PPMs
A. Oropharynx
1. AGNB: polymyxin E with tobramycin 2 g of 2% paste or gel
2. Yeasts: amphotericin B or nystatin 2 g of 2% paste or gel
3. MRSA: vancomycin 2 g of 4% paste or gel
B. Gut
1. AGNB: polymyxin E (mg) 100 200 400
with tobramycin (mg) 80 160 320
2. Yeasts: amphotericin B (mg) 500 1,000 2,000
or nystatin units 2 9 106 4 9 106 8 9 106
3. MRSA: vancomycin (mg) 2040/kg 2040/kg 5002,000
3. Hygiene with topical antimicrobials
4. Surveillance swabs of throat and
rectum on admission, Monday, Thursday
PPMs potentially pathogenic microorganisms; AGNB aerobic Gram-negative bacilli; MRSA
methicillin-resistant Staphylococcus aureus
staphylococci. The most important feature of SDD is the enteral administration of

nonabsorbable polymyxin E/tobramycin to eradicate the abnormal aerobic Gram
negative bacilli (AGNB). This results in decontamination of the digestive tract.
Critically ill patients are unable to clear these pathogens due to their underlying
disease. Intestinal overgrowth with AGNB causes systemic immunoparalysis [2].
The reason for the enteral administration of polymyxin E/tobramycin is that it
promotes recovery of systemic immunity and because preventing or eradicating
abnormal AGNB in throat and gut effectively controls aspiration and translocation
of these microorganisms into the lower airways and blood stream, respectively.
Enterally administered antimicrobials have been shown to be effective in con-
trolling secondary endogenous infections. However, their use does not affect pri-
mary endogenous and exogenous infections. The second component is adequate
parenteral administration of an antimicrobial to control primary endogenous
infections. Cefotaxime has been used in most trials to cover both normal and
abnormal pathogens. In adding enterally to parenterally administered antibiotics,
the original pre-1980s antibiotics remain useful, without the development of
antimicrobial resistance. Third, high standards of hygiene are indispensable for
reducing hand contamination and subsequent transmission from external sources.
Finally, surveillance samples of the throat and rectum are an integral component of
the SDD protocol. Knowledge of the carrier state allows compliance and efficacy
of this prophylactic protocol to be monitored.
13.2 Efficacy
After 25 five years of clinical research, SDD has been assessed in 63 randomized
controlled trials (RCTs) [365] and ten meta-analyses of RCTs only (Table 13.2)
[6675].
13.2.1 Carriage
SDD significantly reduced oropharyngeal carriage by 87% [odds ratio (OR) 0.13,
95% confidence interval (CI) 0.070.2] and rectal carriage due to Gram-negative
PPMs by 85% (OR 0.15; 95% CI 0.070.31) [72]. Gram-positive carriage was also
reduced, but not significantly. Additionally, fungal carriage was significantly
reduced by 68% (OR 0.32, 95% CI 0.190.53) [70].
13.2.2 Lower Airway Infection
All meta-analyses showed a significant reduction of lower respiratory tract infection.

The meta-analysis from the Italian Cochrane Centre demonstrated that enteral and
parenteral administration of antimicrobials for SDD reduced lower airway infections
by 82% (OR 0.28; 95% CI 0.200.38) [74]. Only four patients needed to be treated
with SDD to prevent one case of pneumonia. Moreover, lower airway infection due
to Gram-negative bacteria was reduced by 89% (OR 0.11, 95% CI (0.050.20) and
that due to Gram-positives by 48% (OR 0.52 95% CI 0.340.78) [72].
13.2.3 Bloodstream Infection
SDD significantly reduced bloodstream infections by 27% (OR 0.73, 95% CI

0.590.90), particularly those due to Gram-negative bacteria (OR 0.39; 95% CI
0.240.63) [71].
13.2.4 Fungal Infection
SDD including polyene, either amphotericin B or nystatin, significantly reduced

fungal infections by 70% (OR 0.30, 95% CI 0.170.53). Fungemia was reduced,
albeit not significantly, mainly due to the low event rates in test and control groups
(OR 0.89, 95% CI 0.164.95) [70].
Table 13.2 Efficacy of selective decontamination of the digestive tract assessed in ten meta-
analyses of randomized controlled trials only
First author Year Lower airway Bloodstream MODS: OR Mortality:

infection: OR (95% infection: OR (95% CI) OR (95%
CI) (95% CI) CI)
Vandenbroucke- 1991 0.12 (0.080.19) 0.92
Grauls [66] (0.451.84)
DAmico [67] 1998 0.35 (0.290.41) 0.80
(0.690.93)
Liberati [68] 2004 0.35 (0.290.41) 0.78
(0.680.89)
Safdar [69] 2004 0.82
(0.222.45)
Silvestri [70] 2005 0.89 (0.164.95)
Silvestri [71] 2007 0.63 (0.460.87) 0.74
(0.610.91)
Silvestri [72] 2008
Gram negative 0.07 (0.040.13) 0.36 (0.220.60)
Gram positive 0.52 (0.340.78) 1.03 (0.751.41)
Silvestri [73] 2009 0.71
(0.610.82)
Liberati [74] 2009 0.28 (0.200.38) 0.75
(0.650.87)
Silvestri [75] 2010 0.50 0.82
(0.340.74) (0.511.32)
OR odds ratio; CI confidence interval; MODS multiple organ dysfunction syndrome
13.2.5 Mortality
Mortality rate was an outcome measure in eight of the ten meta-analyses [6669,
71, 7375]. There was a consistent survival benefit in all meta-analyses that
assessed the full four-component SDD protocol providing the sample size was
large enough [67, 68, 71, 73, 74]. The Italian meta-analysis, which assessed only
RCTs in which the full SDD protocol was used, showed a mortality rate reduction
of 29% (OR 0.71, 95% CI 0.610.82) [73]. This effect achieved a 42% mortality
rate reduction in studies where SDD eradicated the carrier state (OR 0.58, 95% CI
0.450.77) [73]. Eighteen patients need to be treated with the full SDD protocol to
prevent one death [73, 74]. The meta-analyses of Vandenbroucke-Grauls and
Vandenbroucke [66], Safdar et al. [69], and Silvestri et al. [75] showed an impact
on mortality rate that was not significant due to the small sample size. Two Dutch
RCTs with the primary endpoint of mortality have been published. In the first [17],
the randomization unit was the ICU and not the patient and included about 1,000
patients. The risk of mortality was significantly reduced by 40% in the unit in
which SDD was administered to all patients (OR 0.6; 95% CI 0.40.8). The second
[19] is the largest study on SDD ever published and included about 6,000 patients.
The primary endpoint was mortality, whereas resistance was among the secondary
endpoints. The study compared SDD, selective oropharyngeal decontamination
(SOD), a modified SDD protocol without the gut component and the parenterally
administered antibiotic, and standard care. Both SDD and SOD significantly
reduced the odds of death compared with standard care [OR 0.83 (p = 0.02), and
0.86 (p = 0.045), respectively]. However, mortality rate reduction was higher,
albeit not significantly, in the SDD group than in the SOD group. These results
regarding SOD have been confirmed by a recent meta-analysis of the nine RCTs
using SOD and including 4,733 patients [76]. Although SOD has been shown to
significantly reduce the odds of pneumonia, the meta-analysis failed to demon-
strate any significant impact on survival (OR 0.93; 95% CI 0.811.07). Addi-
tionally, SOD has been shown to be associated with a 33% and SDD with a 45%
reduction in ICU-acquired Gram-negative bacteremia [77], explaining why SDD,
and not SOD, is associated with a significant mortality rate reduction.
13.2.6 Miscellaneous
One meta-analysis explored the efficacy of SDD in preventing multiple organ

dysfunction syndrome (MODS) [75]. Seven RCTs involving 1,270 patients
reported available information and demonstrated that SDD reduces MODS by 50%
(OR 0.50, 95% CI 0.340.74). Overall mortality rates for SDD versus control
patients were 18.7 and 22.9%, respectively, demonstrating a nonsignificant
reduction in the odds of death (OR 0.82, 95% CI 0.511.32), due to the small
sample size.
Another meta-analysis verified whether SDD reduced ventilator-associated
tracheobronchitis (VAT) [78]. Twelve RCTs involving 2,252 patients (1,102 SDD;
1,150 controls) provided useful information on VAT. There were 135 (12.25%)
patients with VAT in the SDD group and 234 (20.34%) in controls, indicating a
46% VAT reduction in the group receiving SDD (OR 0.54; 95% CI 0.420.69).
The efficacy of SDD in selected patient groups, such as burn patients and
patients receiving esophageal surgery, has been explored. The meta-analysis on
mortality rates in burn patients assessed three RCTs recruiting 440 patients [79].
There were 48 deaths: 15 (5.2%) in the SDD group and 33 (21.8%) in controls.
SDD significantly reduced the odds of death by 78% (OR 0.22; 95% CI
0.120.43). Three RCTs investigating gastroesophageal surgery were pooled in a
meta-analysis of 410 patients (198 SDD, 212 controls) [80]. Fifty-six patients
developed pneumonia: 15 (7.65%) in the SDD group and 41 (19.34%) in controls.
SDD significantly reduced the odds for pneumonia by 64% (OR 0.36; 95% CI
0.190.69; p = 0.0018). Interestingly, anastomotic leakage was significantly
reduced by SDD.
13.3 Safety
The use of parenterally administered antibiotics has been shown to lead to the
emergence of antimicrobial resistance, which has not been shown in RCTs of
SDD [81]. This may be explained by the fact that the addition of enterally
administered antibiotics to the parenterally administered antibiotics may have
kept the systemic agents useful. An intriguing aspect of 25 years of clinical
research in SDD is the experience that the pre-1980s antibiotics, such as cefo-
taxime, are still active as long they are combined with successful eradication of
AGNB from the gut.
Resistance was the endpoint of three RCTs of SDD [13, 17, 19]. A Klebsiella
pneumoniaeproducing extended-spectrum beta-lactamase was endemic in a
French hospital [13]: carriage and infection rates were 19.6 and 9%, respectively.
Once enterally administered antimicrobials were added to those administered
parenterally, there was a significant reduction in both carriage and infection (19.6
vs. 1%; 9 vs. 0%). A Dutch single-center RCT of about 1,000 patients reported that
carriage of AGNB resistant to imipenem, ceftazidime, ciprofloxacin, tobramycin,
and polymyxins occurred in 16% of patients receiving parenterally and enterally
administered antimicrobials compared with 26% of control patients who received
antibiotics parenterally only, with a relative risk of 0.6 (95% CI 0.50.8) [17]. The
largest multicenter RCT to date is also from The Netherlands and comprised about
6,000 patients [19]. The proportion of patients with AGNB shown in rectal swabs
that were not susceptible to the marker antibiotics was lower with SDD than with
standard care or SOD. For example, carriage of multi-drug-resistant Pseudomonas
aeruginosa was 0.4% in SDD versus 0.8% in SOD and 1.3% in the group receiving
standard care (p \ 0.005). Moreover, the study authors reported in a separate
analysis of the same RCT results on bacteremia and lower respiratory tract col-
onization due to highly resistant microorganisms (HRMO), namely aerobic Gram-
negative bacilli [82]. Bacteremia due to HRMO was significantly reduced by SDD
compared with SOD (OR 0.37; 95% CI 0.160.85). Lower respiratory tract
colonization due to HRMO was less with SDD (OR 0.58; 95% CI 0.430.78) than
with SOD (OR 0.65; 95% CI 0.490.87) compared with standard care. Therefore,
SDD was superior to SOD and to standard care in preventing antimicrobial
resistance.
In an ecological study [83] conducted during the study periods of the Dutch
RCT [19], an increase in resistance after discontinuation of SOD and SDD was
observed, which seems to contradict the reduction in resistance. However, that
ecological analysis has an important limitation, i.e., the use of a point-prevalence
survey in which all patients in the unit (whether enrolled in the SDD or SOD trial)
were included. Moreover, the average prevalence of AGNB resistant to ceftazi-
dime, tobramycin, and ciprofloxacin in the respiratory tract was significantly lower
during SDD/SOD than the pre- and post-intervention periods, and AGNB resis-
tance to ciprofloxacin and tobramycin in rectal swabs was significantly reduced
during SDD compared with standard care/SOD [84, 85].
The target microorganisms of SDD include PPMs belonging to the normal flora,
including S. pneumoniae and MSSA, as well as the opportunistic aerobic Gram-
negative bacilli, including Klebsiella, Acinetobacter, and Pseudomonas spp.
Methicillin-resistant S. aureus (MRSA), by design, is not covered by the original
SDD protocol, and hence, six randomized trials conducted in ICUs in which
MRSA was endemic at the time of the study showed a trend toward higher MRSA
infection rates in patients receiving SDD. These observations suggest that the
parenterally and enterally administered antimicrobials of the SDD protocol, i.e.,
cefotaxime, polymyxin, tobramycin, and amphotericin B, may select for and
promote MRSA. Under these circumstances, SDD requires the addition of oro-
pharyngeally and intestinally administered vancomycin. Two studies showed that
adding vancomycin to SDD is an effective and safe maneuver [86, 87].
SDD is not active against vancomycin-resistant enterococci (VRE). All SDD
randomized trials were undertaken in ICU and hospital settings without VRE
experience. SDD was evaluated in two observational studies undertaken in ICU
with a low VRE prevalence [87, 88]. In the Spanish study, VRE was imported into
the unit, but no change in policy was required, as extensive spread did not occur
[87]. In the American study, SDD was evaluated in a unit with a low incidence of
VRE, and the authors reported that SDD did not increase the incidence of VRE
carriage and infection [88].
13.4 Conclusions
SDD is the only evidence-based maneuver that prevents infection in the critically
ill. It significantly reduces lower respiratory tract infections, bloodstream
infections, multiple organ failure, mortality rates, and resistance if the full four-
component protocol is used. SOD only significantly reduces pneumonia but not
mortality rates. Moreover, the full SDD protocol significantly reduces intestinal
carriage of multi-drug-resistant aerobic Gram-negative microorganisms, thus
reducing the occurrence of ICU-acquired bacteremia.
References
1. Stoutenbeek CP, van Saene HKF, Miranda DR, Zandstra DF (1984) The effect of selective
decontamination of the digestive tract on colonization and infection rate in multiple trauma
patients. Intensive Care Med 10:185192
2. van Saene HKF, Petros AJ, Ramsay G, Baxby D (2003) All great truths are iconoclastic:
Care Med 29:677690
3. Abdel-Razek SM, Abdel-Khalek AH, Allam AM et al (2000) Impact of selective
gastrointestinal decontamination on mortality and morbidity in severely burned patients.
Ann Burns Fire Disasters 13:213215
4. Abele-Horn M, Dauber A, Bauernfeind A et al (1997) Decrease in nosocomial pneumonia in
ventilated patients by selective oropharyngeal decontamination (SOD). Intensive Care Med
23:18781895
5. Aerdts SJA, van Dalen R, Clasener HAL et al (1991) Antibiotic prophylaxis of respiratory
tract infection in mechanically ventilated patients. A prospective, blinded, randomized trial of
the effect of a novel regimen. Chest 100:783791
6. Arnow PA, Caradang GC, Zabner R et al (1996) Randomized controlled trial of selective
decontamination for prevention of infections following liver transplantation. Clin Infect Dis
22:9971003
7. Barret JP, Jeschke MG, Herndon DN (2001) Selective decontamination of the digestive tract
on severely burned pediatric patients. Burns 27:439445
8. Bergmans DCJJ, Bonten MJM, Gaillard CA et al (2001) Prevention of ventilator-associated
pneumonia by oral decontamination. A prospective, randomized, double-blind, placebo-
controlled study. Am J Respir Crit Care Med 164:382388
reduces Gram-negative pulmonary colonization but not systemic endotoxemia in patients
undergoing elective liver transplantation. Crit Care Med 22:4049
10. Blair P, Rowlands BJ, Lowry K et al (1991) Selective decontamination of the digestive tract: a
stratified, randomized, prospective study in a mixed intensive care unit. Surgery 110:303310
11. Boland JP, Sadler DL, Stewart W et al (1991) Reduction of nosocomial respiratory tract
infections in the multiple trauma patients requiring mechanical ventilation by selective
parenteral and enteral antisepsis regimen (SPEAR) in the intensive care [abstract]. 17th
congress of chemotherapy, Berlin, N0465
12. Bouter H, Schippers EF, Luelmo SAG et al (2002) No effect of preoperative selective gut
randomized, placebo-controlled study. Crit Care Med 30:3843
nosocomial multiresistant gram-negative bacilli. Study of an outbreak in an intensive care
unit. Ann Intern Med 110:873881
14. Camus C, Bellisant E, Sebille V et al (2005) Prevention of acquired infections in
intubated patients with the combination of two decontamination regimens. Crit Care Med
33:307314
15. Cerra FB, Maddaus MA, Dunn DL et al (1992) Selective gut decontamination reduces
nosocomial infections and length of stay but not mortality or organ failure in surgical
intensive care unit patients. Arch Surg 127:163169
16. Cockerill FR, Muller SR, Anhalt JP et al (1992) Prevention of infection in critically ill
patients by selective decontamination of the digestive tract. Ann Intern Med 117:545553
17. de Jonge E, Schultz M, Spanjaard L et al (2003) Effects of selective decontamination of the
18. de la Cal MA, Cerd E, Garcia-Hierro P et al (2005) Survival benefit in critically ill burned
patients receiving selective decontamination of the digestive tract. A randomized, placebo-
controlled, double-blind trial. Ann Surg 241:424430
19. de Smet AMGA, Kluytmans JA, Cooper BS et al (2009) Decontamination of the digestive
tract and oropharynx in ICU patients. N Engl J Med 360:2031
20. Diepenhorst GMP, van Ruler O, Besselink MGH et al (2011) Influence of prophylactic
probiotics and selective decontamination on bacterial translocation in patients undergoing
pancreatic surgery. Shock 35:916
21. Farran L, Llop J, Sans M et al (2008) Efficacy of enteral decontamination in the prevention of
anastomotic dehiscence and pulmonary infection in esophagogastric surgery. Dis Esophagus
21:159164
22. Ferrer M, Torres A, Gonzalez J et al (1994) Utility of selective decontamination in
mechanically ventilated patients. Ann Intern Med 120:389395
23. Finch RG, Tomlinson P, Holliday M et al (1991) Selective decontamination of the digestive
tract (SDD) in the prevention of secondary sepsis in a medical/surgical intensive care unit
[abstract]. 17th congress of chemotherapy, Berlin, N0471
24. Flaherty J, Nathan C, Kabins SA et al (1990) Pilot trial of selective decontamination for
prevention of bacterial infection in an intensive care unit. J Infect Dis 162:13931397
25. Gastinne H, Wolff M, Delatour F et al (1992) A controlled trial in intensive care units of
selective decontamination of the digestive tract with nonabsorbable antibiotics. N Engl J Med
326:594959
26. Gaussorgues Ph, Salord F, Sirodot M et al (1991) Efficacit de la dcontamination digestive
sur la survenue des bactrimies nosocomiales chez les patients sous ventilation mcanique et
recevant des btamimtiques. Ranimation Soins Intensive et Mdicine dUrgence 7:169174
27. Georges B, Mazerolles M, Decun J-F et al (1994) Decontamination digestive selective:
resultats dune etude chez le polytraumatise. Reanim Urgences 3:621627
28. Gosney M, Martin MV, Wright AE (2006) The role of selective decontamination of the
digestive tract in acute stroke. Age Aging 35:4247
29. Hammond JMJ, Potgieter PD, Saunders GL et al (1992) Double-blind study of selective
decontamination of the digestive tract in intensive care. Lancet 340:59
30. Hellinger WC, Yao JD, Alvarez S et al (2002) A randomized, prospective, double blinded
evaluation of selective bowel decontamination in liver transplantation. Transplantation
73:19041909
31. Jacobs S, Foweraker JE, Roberts SE (1992) Effectiveness of selective decontamination of the
digestive tract (SDD) in an ICU with a policy encouraging a low gastric pH. Clin Intensive
Care 3:5258
32. Kerver AJH, Rommes JH, Mevissen-Verhage EAE et al (1988) Prevention of colonization
and infection in critically ill patients: a prospective randomized study. Crit Care Med
16:10871093
33. Korinek AM, Laisne MJ, Nicolas MH et al (1993) Selective decontamination of the digestive
tract in neurosurgical intensive care unit patients: a double-blind, randomized, placebo-
controlled study. Crit Care Med 21:14661473
34. Krueger WA, Lenhart F-P, Neeser G et al (2002) Influence of combined intravenous and
mortality in critically ill surgical patients. A prospective, stratified, randomized, double-blind,
placebo-controlled clinical trial. Am J Respir Crit Care Med 166:10291037
35. Laggner AN, Tryba M, Georgopulos A et al (1994) Oropharyngeal decontamination with
gentamicin for long-stay ventilated patients on stress ulcer prophylaxis with sucralfate? Wien
Klin Wochenschr 106:1519
36. Lingnau W, Berger J, Javorsky F et al (1997) Selective intestinal decontamination in multiple
trauma patients: prospective, controlled trial. J Trauma 42:687694
37. Luiten EJT, Hop WCJ, Lange JF et al (1995) Controlled clinical trial of selective
decontamination for the treatment of severe acute pancreatitis. Ann Surg 222:5765
38. Martinez-Pellus AE, Merino P, Bru M et al (1993) Can selective digestive decontamination
avoid the endotoxemia and cytokine activation promoted by cardiopulmonary bypass? Crit
Care Med 21:16841691
39. Martinez-Pellus AE, Merino P, Bru M et al (1997) Endogenous endotoxemia of intestinal
origin during cardiopulmonary bypass. Role of type of flow and protective effect of selective
digestive decontamination. Intensive Care Med 23:12511257
40. Oudhuis GJ, Bergmans DG, Dormans T et al (2011) Probiotics versus antibiotic
decontamination of the digestive tract: infection and mortality. Intensive Care Med
37:110117
41. Palomar M, Alvarez-Lerma F, Jord R et al (1997) Prevention of nosocomial infection in
mechanically ventilated patients: selective digestive decontamination versus sucralfate. Clin
42. Pneumatikos I, Koulouras V, Nathanail C et al (2002) Selective decontamination of
subglottic area in mechanically ventilated patients with multiple trauma. Intensive Care Med
28:432437
43. Pugin J, Auckenthaler R, Lew DP et al (1991) Oropharyngeal decontamination decreases

incidence of ventilator-associated pneumonia. A randomized, placebo-controlled, double-
blind clinical trial. JAMA 265:27042710
44. Quinio B, Albanese J, Bues-Charbit M et al (1996) Selective decontamination of the
digestive tract in multiple trauma patients. A prospective double-blind, randomized, placebo-
controlled study. Chest 109:765772
45. Rayes N, Seehofer D, Hansen S et al (2002) Early enteral supply of Lactobacillus and fiber
versus selective bowel decontamination: a controlled trial in liver transplant recipients.
Transplantation 74:123128
46. Rios F, Maskin B, Saenz Valliente A et al (2005) Prevention of ventilation associated
pneumonia (VAP) by oral decontamination (OD). Prospective, randomized, double blind,
placebo-controlled study. ATS interventional conference, San Diego, USA, C95, Poster 608
(abstract) available at http://abstracts2view.com/ats05/view.php?nu=ATS5L-3301
47. Rocha LA, Martin MJ, Pita S et al (1993) Prevention of nosocomial infection in critically ill
patients by selective decontamination of the digestive tract. A randomized, double-blind,
placebo controlled study. Intensive Care Med 18:398404
48. Rodriguez-Roldan JM, Altuna-Cuesta A, Lopez A et al (1990) Prevention of nosocomial lung
infection in ventilated patients: use of an antimicrobial pharyngeal nonabsorbable paste. Crit
Care Med 18:12391242
49. Rolando N, Gimson A, Wade J et al (1993) Prospective controlled trial of selective parenteral
and enteral antimicrobial regimen in fulminant liver failure. Hepatology 17:196201
50. Rolando N, Wade JJ, Stangou A et al (1996) Prospective study comparing the efficacy of
prophylactic parenteral antimicrobials. With or without enteral decontamination, in patients
with acute liver failure. Liver Transpl Surg 2:813
51. Ruza F, Alvarado F, Herruzo R et al (1998) Prevention of nosocomial infection in a pediatric
intensive care unit (PICU) through the use of selective digestive decontamination. Eur J
Epidemiol 14:719727
52. Sanchez Garcia M, Cambronero Galache JA, Lopez Diaz J et al (1998) Effectiveness and cost
of selective decontamination of the digestive tract in critically ill intubated patients. A
randomized, double-blind, placebo-controlled, multicenter trial. Am J Respir Crit Care Med
158:908916
53. Schardey HM, Joosten U, Finke U et al (1997) The prevention of anastomotic leakage after
total gastrectomy with local decontamination. A prospective, randomized, double-blind,
placebo-controlled, multicenter trial. Ann Surg 225:172180
54. Smith SD, Jackson RJ, Hannakan CJ et al (1993) Selective decontamination in pediatric liver
transplants. A randomized prospective study. Transplantation 55:13061309
55. Stoutenbeek CP, van Saene HKF, Zandstra DF (1996) Prevention of multiple organ failure by
selective decontamination of the digestive tract in multiple trauma patients. In: Faist E, Baue
AE, Schildberg FW (eds) The immune consequences of trauma, shock and sepsis
mechanisms and therapeutic approach. Pabst Science Publishers, Lengerich, pp 10551066
56. Stoutenbeek CP, van Saene HKF, Little RA et al (2007) The effect of selective
decontamination of the digestive tract on mortality in multiple trauma patients: a
multicenter randomized controlled trial. Intensive Care Med 33:261270
57. Tetteroo GWM, Wagenvoort JHT, Castelein A et al (1990) Selective decontamination to
reduce gram-negative colonisation and infections after oesophageal resection. Lancet
335:704707
58. Ulrich C, Harinck-de Weerd JE, Bakker NC et al (1989) Selective decontamination of the
digestive tract with norfloxacin in the prevention of ICU-acquired infections: a prospective
randomized study. Intensive Care Med 15:424431
59. Unertl K, Ruckdeschel G, Selbmann HK et al (1987) Prevention of colonization and
respiratory infections in long-term ventilated patients by local antimicrobial prophylaxis.
60. Verwaest C, Verhaegen J, Ferdinande P et al (1997) Randomized, controlled trial of selective

digestive decontamination in 600 mechanically ventilated patients in a multidisciplinary
intensive care unit. Crit Care Med 25:6371
61. Wiener J, Itokazu G, Nathan C et al (1995) Randomized, double-blind, placebo-controlled
trial of selective digestive decontamination in a medical-surgical intensive care unit. Clin
62. Winter R, Humphreys H, Pick A et al (1992) A controlled trial of selective decontamination
of the digestive tract in intensive care and its effect on nosocomial infection. J Antimicrob
Chemother 30:7377
63. Yu J, Xiao YB, Wang XY (2007) Effects of preoperatively selected gut decontamination on
cardiopulmonary bypass-induced endotoxemia. Chin J Traumatol 10:131137
64. Zobel G, Kutting M, Grubbauer H-M et al (1991) Reduction of colonization and infection
rate during pediatric intensive care by selective decontamination of the digestive tract. Crit
Care Med 19:12421246
65. Zwaveling JH, Maring JK, Klompmaker IJ et al (2002) Selective decontamination of the
digestive tract to prevent postoperative infection: a randomized placebo-controlled trial in
liver transplant patients. Crit Care Med 30:12041209
66. Vandenbroucke-Grauls CMJ, Vandenbroucke JP (1991) Effect of selective decontamination
of the digestive tract on respiratory tract infections and mortality in the intensive care unit.
Lancet 338:859862
67. DAmico R, Pifferi S, Leonetti C et al (1998) Effectiveness of antibiotic prophylaxis in critically
ill adult patients: systematic review of randomised controlled trials. BMJ 316:12751285
1:CD000022
69. Safdar N, Said A, Lucey MR (2004) The role of selective digestive decontamination for
reducing infection in patients undergoing liver transplantation: a systematic review and meta-
analysis. Liver Transpl 10:817827
70. Silvestri L, van Saene HKF, Milanese M et al (2005) Impact of selective decontamination of
the digestive tract on fungal carriage and infection: systematic review of randomised
controlled trials. Intensive Care Med 31:898910
71. Silvestri L, van Saene HKF, Milanese M et al (2007) Selective decontamination of the
digestive tract reduces bacterial bloodstream infections and mortality in critically ill patients.
Systematic review of randomised, controlled trials. J Hosp Infect 15:187203
72. Silvestri L, van Saene HKF, Casarin A et al (2008) Impact of selective decontamination of
the digestive tract on carriage and infection due to Gram-negative and Gram-positive bacteria.
A systematic review of randomised controlled trials. Anaesth Intensive Care 36:324338
73. Silvestri L, van Saene HKF, Weir J, Gullo A (2009) Survival benefit of the full selective
digestive decontamination regimen. J Crit Care 24:e7e14
4:CD000022
75. Silvestri L, van Saene HKF, Zandstra DF et al (2010) Impact of selective decontamination of
the digestive tract on multiple organ dysfunction syndrome: systematic review of randomized
controlled trials. Crit Care Med 38:13701376
76. Silvestri L, van Saene HKF, Zandstra DF et al (2010) SDD, SOD or oropharyngeal
chlorhexidine to prevent pneumonia and to reduce mortality in ventilated patients: which
manoeuvre is evidence-based? Intensive Care Med 31:14361437
77. Oostdijk EAN, de Smet AMGA, Kesecioglu J, Bonten MJM, on behalf of the Dutch
SOD-SDD Trialsts Group (2011) The role of intestinal colonization with Gram negative
bacteria as a source for intensive care unit-acquired bacteremia. Crit Care Med 39:961966
78. Silvestri L, Milanese M, Taylor N et al (2010) Selective digestive decontamination reduces
ventilator-associated tracheobronchitis. Respir Med 104:19531955
79. Silvestri L, de la Cal MA, Taylor N et al (2010) Selective decontamination of the digestive
tract in burn patients: an evidence-based maneuver that reduces mortality. J Burn Care Res
31:372373
80. Silvestri L, van Saene HKF (2010) Selective digestive decontamination to prevent
pneumonia after esophageal surgery. Ann Thor Cardiovasc Surg 16(3):220221
81. Silvestri L, van Saene HKF (2006) Selective decontamination of the digestive tract does not
increase resistance in critically ill patient. Evidence from randomized controlled trials. Crit
Care Med 34:20272029
82. de Smet AM, Kluytmans J, Blok H et al (2010) Effects of selective digestive and selective
oropharyngeal decontamination on bacteraemia and respiratory tract colonization with highly
resistant micro-organisms. Clin Microbiol Infect 16(suppl 2):S98
83. Oodstijk EAN, de Smet AMGA, Blok HEM et al (2010) Ecological effects of selective
Med 181:452457
84. Zandstra DF, Petros AJ, Taylor N et al (2010) Withholding selective decontamination of the
digestive tract from critically ill patients must now surely be ethically questionable given the
vast evidence base. Crit Care 14:443
85. Petros AJ, Taylor N, V Damjanovic et al (2010) Worlds apart; proof that SDD works. Am J
Control 30:391399
88. Bhorade SM, Christensen J, Pohlman AS et al (1999) The incidence of and clinical variables
associated with vancomycin-resistant enterococcal colonization in mechanically ventilated
patients. Chest 115:10851091
Part IV
Infections on ICU
Lower Airway Infection
14
J. Almirall, A. Liapikou, M. Ferrer
and A. Torres
14.1 Definition
Lower respiratory tract infections (RTI) in intubated patients include

ventilator-associated tracheobronchitis (VAT) and ventilator-associated pneumo-
nia (VAP). Both are hospital-acquired infections that occur within 48 h after
intubation [1, 2]. Diagnostic criteria for VAT and VAP overlap in terms of clinical
signs and symptoms. In contrast to VAT, VAP requires the presence of new and
persistent pulmonary infiltrates on a chest radiograph, which may be difficult to
interpret in some critically ill patients, and two or more of the following criteria:
fever ([38.3C) or hypothermia; leukocyte count [10,000/ll; purulent tracheo-
bronchial secretions, or a reduced partial pressure of oxygen in arterial blood
(PaO2)/fraction of inspired oxygen (FiO2) ratio C15% according to the US centers
for disease control and prevention definitions. patients with a clinical pulmonary
infection score [6 are also considered to have pneumonia [3].
The apparent crude incidence of VAT ranges from 3 to 10%, but it is difficult to
determine the exact incidence and importance of VAT for several reasons. The major
reason is that to confirm the absence of infiltrates on a chest radiograph, a computed
tomography (CT) scan is required. VAT is probably an intermediate process between
lower respiratory tract colonization and VAP. Postmortem studies show a continuum
between bronchitis and pneumonia in mechanically ventilated (MV) ICU patients
[4]. VAP that occurs during the first 4 days of MV is defined as early onset in order to
differentiate it from late-onset VAP, which develops thereafter.
A. Torres (&)
Servei de Pneumologia i Allrgia Respiratria,
Hospital Clnic, Barcelona, Spain
220 J. Almirall et al.
The term ventilator-associated pneumonia, however, is a misnomer, as the MV

is not the main risk factor for lung colonization and pneumonia. The endotracheal
tube (ETT) seems to play the most important role in the pathogenesis of VAP, as it
creates a direct conduit for bacteria to reach the lower airways and greatly impairs
host defenses. Interestingly, studies demonstrate that MV could also increase the
risk of pneumonia. Indeed, lungs become highly susceptible to bacterial coloni-
zation when injurious ventilatory settings are applied, i.e., with high tidal volumes
and low positive end expiratory pressures (PEEP).
Therefore, either ETT-associated pneumonia or ventilation-acquired pneumo-
nia are better terms to describe pneumonia in tracheally intubated and MV
patients, as they emphasize the role of ETT and MV in the pathogenesis of such
pneumonia. The term ventilation-acquired pneumonia would allow physicians and
scientists to maintain the current acronym VAP [5].
14.2 Pathogenesis
Tracheally intubated patients can be colonized via exogenous and endogenous

bacterial sources. When bacteria gain access to the lower respiratory tract in
healthy, nonintubated patients, colonization is prevented by several defense
mechanisms, such as cough, cilia, mucous clearance, polymorphonuclear leuko-
cytes, macrophages and their respective cytokines, antibodies [immunoglobulin
(Ig)M, IgG, IgA], and complement factors. Critically ill patients are already at
high risk of infection because of the illness, comorbidities, and malnutrition.
In MV patients, the tracheal tube may encourage aspiration by bypassing normal
defenses, allowing secretions to pool in the upper part of the trachea. It also creates
a direct conduit for bacteria to reach the airways, impairs cough, compromises
mucociliary clearance, and facilitates bacterial adhesion to the airways through
cuff-related injury to the tracheal mucosa. When endotracheal tubes are inserted
nasally instead of orally, sinusitis is significantly more likely to occur through
blockage of the sinus ostia. The occurrence of nosocomial sinusitis has been
associated with VAP.
High-volume, low-pressure, endotracheal tube cuffs, commonly used during
prolonged MV, are not leakproof, and micro- and macroaspiration of bacteria-
laden oropharyngeal secretions often occurs. Patients are colonized from exoge-
nous bacterial sources via the hands and apparel of healthcare personnel, con-
taminated aerosols, and invasive devices such as tracheal aspiration catheters and
fiberoptic bronchoscopes (FOB). Pathogens are also acquired from the patients
endogenous flora, though there is still controversy regarding the primary source of
infection (oropharynx, stomach). It is well acknowledged, however, that in
critically ill patients, oral flora quickly shifts to a predominance of aerobic
Gram-negative pathogens Pseudomonas aeruginosa and methicillin-resistant
Staphylococcus aureus (MRSA). Following bacterial aspiration and colonization
of the proximal airways, the occurrence of VAP mainly depends on the size of the
inoculum, functional status, exposure to antibiotics, and potential host defenses.
14 Lower Airway Infection 221
14.3 Epidemiology
Nosocomial pneumonia accounts for 31% of all nosocomial infections, and a large
majority (83%) of patients who develop nosocomial pneumonia are mechanically
ventilated. The exact incidence of VAP is difficult to obtain due to overlapping
lower RTIs and the difficulty in diagnosing VAP correctly. The incidence of VAP
ranges from 9 to 67% of patients on MV. The rate of VAP, expressed as the total
number of episodes of VAP/1,000 ventilator days, ranges from 5 to 16 [6]. VAP
can increase the time on a ventilator by 10 days, length of ICU stay by 6 days, and
length of total hospital stay by 11 days.
Disease incidence depends greatly on the type of population studied, the
presence or absence of risk factors for colonization by multi-drug-resistant
pathogens, and the type and intensity of preventive strategies applied. Tracheal
intubation and MV are the main risk factors for VAP during the first week of
ventilation (risk assessed at approximately 3% per day in the first week of MV).
A one-day point-prevalence study conducted in 1,417 intensive care units (ICUs)
in Western Europe reported that VAP was the most common ICU-acquired
infection and MV was associated with a threefold increased risk of developing
pneumonia [7]. Studies conducted in several countries in the European Union
have shown varying incidence density ranging from approximately 925 cases/
1,000 ventilation days [6]. Epidemiological studies on a large United States
database with medical, surgical, and trauma patients have shown a VAP inci-
dence of 9.3%.
Hospital mortality rate of patients with VAP is significantly higher than that
of patients without VAP. Crude VAP mortality rates range between 20 and
50%, depending on comorbidities, illness severity, pathogens, and quality of
antibiotic treatment [1]. Ventilated ICU patients with VAP appear to have a
two- to tenfold higher risk of death compared with patients without pneumonia.
However, several patients with VAP die and not because of VAP. However,
mortality rates vary from one study to another, and the prognostic impact is
debated. It is well recognized that one-third to one-half of all VAP deaths are
directly attributable to the disease. Mortality rates are higher when VAP
associated with bacteremia, especially with P. aeruginosa or Acinetobacter
spp., medical rather than surgical illness, and treatment with ineffective anti-
biotic therapy [2].
VAP is associated with higher medical care costs. Patients who develop VAP
during a hospital stay remain longer in the ICU and the hospital, and the increased
level of care and need for additional invasive procedures drastically increases
healthcare costs. It has been reported that each case of VAP is associated with
additional hospital costs of $20000 to more than US $40000. Infection with MRSA
increases hospital costs by an additional $7731 per patient. These data emphasize
the need for prevention and better outcomes [8].
14.4 Etiologic Agents
The etiological cause of VAP is usually identified via semiquantitative

microbiologic analysis of tracheal aspirates with or without initial microscopic
evaluation. When VAP is diagnosed using a microbiologic strategy following
clinical suspicion of lung infection, samples from the lower respiratory tract are
collected and quantitative cultures performed. Pathology studies clearly show
that the sensitivity of microbiological studies is drastically reduced when
antibiotics are administered. Therefore, new antibiotics should be administered
after sampling. Specimens can be obtained noninvasively via a tracheal suction
catheter or invasively through an FOB. When an FOB is used, pathogens from
the lower respiratory tract are retrieved mainly through bronchoalveolar lavage
(BAL) or protected specimen brush (PSB). Several modifications of these
techniques have been developed, such as mini-BAL and blind PSB sampling.
During pneumonia, pathogens colonize the lower respiratory tract at concen-
trations of 105106 colony-forming units/milliliter (CFU/ml). With regard to
sample size, the commonly accepted diagnostic threshold for PSB, BAL, and tra-
cheal aspirates are 103, 104105, and 105106 CFU/ml, respectively. Most of the
current debate regarding VAP diagnosis still concerns invasive versus. noninvasive
sampling techniques. Five randomized clinical trials attempted to demonstrate dif-
ferences in outcome between techniques; only one study showed significant survival
benefit using invasive sampling techniques [9].
Studies in the 1990s confirmed the association between oral bacterial coloni-
zation and nosocomial pneumonia in MV patients. In addition, patients in the ICU
have higher mean plaque scores than patients in non-ICU control groups. Patho-
gens isolated from plaque of these ICU patients included MRSA. These findings
suggest that dental plaque may also provide a reservoir for pathogenic bacteria that
contribute to VAP.
The most common microorganisms implicated as causative agents of VAP are
P. aeruginosa (24%), S. aureus (20%), and Enterobacteriaceae (14%) [1012].
Increasing resistance of S. aureus to methicillin/oxacillin has been reported for
many years, reaching almost 60% in recent studies [13]. Multiple etiologic agents
are often present. All bacteria implicated in the VAP etiology are reported in
Table 14.1.
Several differences in the etiology of early- and late-onset pneumonia can be
recognized, with the former mainly caused by pathogens with enhanced
antibiotic susceptibility and better outcome, such as Haemophilus influenzae
and S. pneumonia. Anaerobic bacteria play a minor role in VAP pathogenesis.
Theoretically, patients who develop VAP within 4 days may have aspirated
oropharyngeal contents colonized by anaerobic bacteria, but the need to
administer antianaerobic drugs has not been clearly established. In general,
viruses and fungi are potential causes of VAP only in immunosuppressed
patients.
Table 14.1 Causative agents of ventilator-associated pneumonia (VAP)
Kollef [8] n = 398 Agbath [9] n = 313 Kollef [10] n = 93

Gram-positive
MSSA 35 (8.8) 68 (21.7) 15 (16.1)
MRSA 59 (14.8) 25 (8.0) 10 (10.7)
Streptococcus pneumoniae 24 (7.7) 6 (6.4)
Streptococcus spp. 13 (4.2)
Gram-negative
Pseudomonas aeruginosa 57 (14.3) 43 (13.7) 19 (20.4)
Haemophilus influenzae 52 (16.6) 6 (6.4)
Enterobacteriaceae 38 (9.5) 64 (20.4) 15 (16.1)
Acinetobacter baumannii 8 (2.0) 10 (3.2) 6 (6.4)
MRSA methicillin-resistant Staphylococcus aureus; MSSA methicillin-sensitive Staphylococcus
aureus
14.5 Risk Factors
A number of papers using both univariate and multivariate statistical techniques

highlight the risk factors associated with VAP. Knowledge of these risk factors is
crucial in implementing effective preventive measures. These risk factors can be
modifiable or nonmodifiable conditions (Table 14.2). More importantly, several
identified risk factors have been modified in studies aiming at reducing VAP
incidence. These include enteral feeding, ventilator-circuit manipulation, patient
positioning, MV modes, and strategies for stress-ulcer prophylaxis. Recent
guidelines classify recommendations for preventative interventions of modifiable
risk factors [2]. Presumed relationships between identified risk factors, preventive
strategies, and VAP pathogenesis are shown in Fig. 14.1.
14.6 Preventive Strategies
The high morbidity and mortality rates of VAP and the costs of the disease, both in
terms of treatment and increasing hospital length of stay, have led to efforts to
reach consensus in control measures and prevention. Many hospitals have devel-
oped and implemented evidence-based prevention protocols and educational
programs for physicians and nurses. These strategies have often improved quality
of care and reduced VAP incidence. When North American epidemiological data
from the 2008 National Healthcare Safety Network (NHSN) report are compared
with data from the 2003 National Nosocomial Infections Surveillance (NNIS),
pneumonia incidence densities are slightly lower overall, suggesting that new
preventive strategies applied in the meantime have had a positive effect [13].
Table 14.2 Risk factors for ventilator-associated pneumonia (VAP)
Modifiable risk factors Nonmodifiable risk factors

Supine patient position Age [60 years
Large-volume gastric aspiration COPD/ARDS/pulmonary disease
Colonization of the ventilator circuit Organ failure
Low endotracheal cuff pressure Coma/impaired consciousness
Staff hand infection Tracheostomy
Nasotracheal intubation Reintubation
Oropharyngeal colonization Intracranial pressure monitor
Histamine type 2 (H2) antagonists and antacids Length of stay in the ICU
Duration of intubation and mechanical
ventilation [2 days
Prior antibiotics
Enteral nutrition
Therapeutic interventions
Use of sedative and paralytic agents
COPD chronic obstructive pulmonary disease; ARDS acute respiratory distress syndrome;
ICU intensive care unit
14.6.1 Ventilator and VAP Bundles
Preventive strategies have focused on reducing/avoiding cross-transmission, pul-

monary aspiration across the cuff, and bacterial load in the oropharynx. Several
strategies with proven efficacy in reducing MV-related morbidity and mortality
rates have been grouped as a ventilator bundle and could bring about a 45%
reduction in VAP rates [14]. The interventions are recommended by the Institute
for Healthcare Improvement (IHI) and include:
1. elevating the head of the bed by 3045;
2. daily sedation vacations and assessment of readiness for extubation;
3. peptic ulcer disease prophylaxis;
4. deep venous thrombosis prophylaxis.
Although the aforementioned bundle was not specifically designed to prevent
VAP, effects of body position, sedation vacation, and assessment of readiness for
extubation have generated significant reduction in VAP rates. The bundle was
subsequently implemented specifically to address VAP prevention, and two
additional strategies were incorporated: (1) daily oral use of chlorhexidine; (2)
subglottic secretion drainage.
14.6.2 Endotracheal Intubation
Intubation and MV is undoubtedly associated with increased risk of VAP

and therefore should be avoided whenever possible. Noninvasive positive-
pressure ventilation (NPPV) is an attractive alternative for patients with acute
Pathogenesis Risk factors Preventive strategies
Malnutrition Ensure appropriate

nutritional support ;
Poor oral hygiene clean oral cavity; daily oral
use of chlorhexidine ;
Prior antibiotic avoid unnecessary antibiotic
administration administration;
Dry mouth prevent dehydration
Bacterial colonization,
(oropharynx/stomach/ Gastrica alkalization Avoid unnecessary stress-
sinuses/subglottic ulcer prophylaxis
space/ventilator circuit
condensate)
Avoid long-term placement
Nasogastric tube of nasogastric tube; interrupt
enteral nutrition for 8 h
every day;
use oral intubation; try
Nasal intubation noninvasive positive-
pressure ventilation;
Accumulation of circuit routinely drain circuit
condensate condensate
Maintain semirecumbent
position
Supine positioning Maintain oral hygiene
Nasogastric tube Use continuous subglottic
suctioning
Large gastric volumes Avoid unplanned
extubations
Aspiration of
contaminated Patient/ventilator circuit Routinely drain circuit
secretions/circuit manipulation condensate
condensate/aerosols into
lower airways
Accumulation of circuit Use a heat and moisture

condensate exchanger
Reintubation Ensure adequate

endotracheal tube cuff
pressures
Extubate as soon as
VAP clinically indicated
Fig. 14.1 Relationship between pathogenesis, risk factors, and preventive strategies for
ventilator-associated pneumonia (VAP)
exacerbations of chronic obstructive pulmonary disease (COPD) or acute hyp-

oxemic respiratory failure and should be used whenever possible in selected
(immunosuppressed patients) with pulmonary infiltrates, fever, and respiratory
failure and to facilitate difficult weaning. Reintubation should be avoided, if
possible, as it increases the risk of VAP [15]. Orotracheal intubation should be
preferred over nasotracheal intubation to prevent nosocomial sinusitis and thus
reduce the risk of VAP.
Specific strategies, such as improved methods of sedation and the use of pro-
tocols to facilitate and accelerate weaning, have been recommended to reduce
intubation and MV duration but are dependent on adequate ICU staffing. Daily
interruption or lightening of sedation, in particular, can decrease time on MV, as
well as avoiding paralytic agents, which is also recommended so as not to depress
defence mechanisms.
14.6.3 Tracheal Tube, Ventilatory Circuit, and Gas Conditioning
Most endotracheal tubes used in the ICU have high-volume, low-pressure

(HVLP) cuffs. The internal volume of standard HVLP cuffs can exceed the
internal diameter of the trachea by up to 40%, so when inflated, HVLP cuffs
seal the trachea without being stretched, and their internal pressure closely
reflects pressure exerted against the tracheal wall. Nevertheless, longitudinal
folds invariably form, and bacteria-laden oropharyngeal secretions easily leak
along these folds, increasing risks for airways infection and pneumonia. Cuffs
made of new materials such as polyurethane have been developed. During
inflation, these cuffs form smaller folds and can prevent or greatly reduce the
aspiration of secretions past the cuff. Leakage of oropharyngeal contents past
the ETT cuff has also been reduced with a new endotracheal tube that contains
a separate dorsal lumen, which opens into the subglottic region and allows
continuous aspiration of subglottic secretions (CASS tube) This strategy has
significantly reduced the incidence of pneumonia, particularly early-onset VAP,
and should be used if available [16]. The internal pressure of the endotracheal
tube cuff pressure must also be maintained between 2530 cm H2O, particu-
larly when no PEEP is applied, to prevent leakage of contaminated secretions
past the cuff into the lower airways and tracheal injury. Patients who require
prolonged endotracheal intubation or bedside percutaneous dilation tracheos-
tomy for prolonged MV are also at risk of developing swallowing dysfunctions
that may predispose to aspiration and the subsequent development of nosoco-
mial pneumonia [17].
The ventilatory circuit can become colonized and facilitate bacterial inocula-
tion. The frequency of ventilator circuit change does not affect the incidence of
VAP, but the condensate fluid collected in the ventilator circuit can increase the
risk of exogenous and endogenous bacterial colonization. Therefore, the inad-
vertent flushing of contaminated condensate into the lower airway should be
avoided through careful emptying of ventilator circuits.
There are no consistent data showing reduced VAP incidence [2] and better
outcome using either heat and moisture exchangers (HME) or heated humidifiers
(HH). Neither humidification strategy can be recommended as a pneumonia pre-
vention tool at this stage; however, inspiratory gases should be delivered at body
temperature or slightly below and at the highest relative humidity in order to
prevent heat and moisture loss from the airways and, more importantly, change in
rheologic properties of secretions and impairment of mucociliary clearance.
14.6.4 Gastric Colonization and Body Position
Gastric sterility is maintained in an acidic environment. In critically ill patients,

use of antacids for stress-ulcer prophylaxis, and enterally administered nutrition
alkalinizes gastric contents and facilitates bacterial colonization of the stomach.
Retrograde colonization of the oropharynx and pulmonary aspiration past
the ETT cuff causes bacterial colonization of the lower respiratory tract
and pneumonia. Guidelines recommend elevating the head of a patients bed
3045, especially during enteral feeding, to reduce gastroesophageal reflux and
incidence of nosocomial pneumonia [2]. Differences between the semirecumbent
and supine positions have been reported in one randomized clinical study. Drak-
ulovic et al. [18] showed that the semirecumbent position (458) lowered the risk
for onset of nosocomial pneumonia by 78% in comparison with completely supine
position (0), reducing the gastrooropharyngeal route of pulmonary infection.
14.6.5 Enterally Administered Nutrition
Enterally administered nutrition in supine patients is a risk factor for VAP devel-
opment through increased risk of aspiration of gastric contents. Residual volume
should be carefully monitored and, in the case of consistently large volumes, the use
of agents that increase gastrointestinal (GI) motility (e.g., metoclopramide). When
necessary, enterally administered nutrition should be withheld to reduce aspiration
risk. Enterally administered nutrition acidification and postpyloric tube placement
and nutrition suspension 8 h daily (intermittent nutrition) are strategies that should
reduce gastric colonization and risk of gastroesophageal reflux, although investi-
gators have reported inconsistent results [19]. However, the effectiveness of such
interventions awaits validation in clinical trials. Nevertheless, intubated patients
should be kept in a semirecumbent position (3045) to prevent aspiration, especially
when receiving enterally administered nutrition.
14.6.6 Stress-Ulcer Prophylaxis
As mentioned above, gastric sterility is maintained in an acidic environment within

the stomach. A gastric pH [4 facilitates bacterial colonization mostly due to
Gram-negative bacteria. However, the majority of critically ill patients are at a
higher risk for GI bleeding during MV; hence, stress-ulcer prophylaxis is essential.
Antacids, histamine-2-receptor antagonists (H2 blockers), and proton-pump
inhibitors (PPI) are usually administered to prevent GI lesions. Sucralfate, an
alternative gastroprotective agent, does not change gastric acidity and prevents GI
bleeding, protecting gastric mucosa. Several randomized clinical trials and
meta-analyses investigated the rates of VAP using sucralfate versus agents that
alkalinize gastric juice (antacids, H2 blockers, PPI) with conflicting results. An
additional risk for GI bleeding using sucralfate has also been found [20]. Thus, the
use of sucralfate as VAP-preventive strategy should only be recommended in
patients with low risk of GI bleeding.
14.6.7 Oropharyngeal and Digestive Tract Colonization
Progression from colonization to tracheobronchitis and pneumonia is a dynamic

process, and identifying the different entities depends on the specificity of
diagnostic tools. Oropharyngeal colonization, either present on admission or
acquired during ICU stay, has been identified as an independent risk factor for
the development of ICU-acquired pneumonia caused by enteric Gram-negative
bacteria and P. aeruginosa [21]. In tracheally intubated patients, oral flora rap-
idly shifts from a predominance of aerobic Gram-positive bacteria and anaerobes
to a majority of aerobic Gram-negative pathogens. Oropharyngeal decontami-
nation can be achieved through topical administration of antiseptics, such as
chlorhexidine. Using chlorhexidine in cardiac postsurgical patients and patients
requiring MV for at least 48 h has been shown to reduce the incidence of VAP.
Despite the reduction in nosocomial pneumonia, no survival benefits were
demonstrated.
Selective oropharyngeal decontamination (SOD) and subglottic decontamina-
tion can be obtained via topical administration of nonabsorbable antibiotics, the
most common being polymyxin E, tobramycin/gentamicin, and amphotericin B,
all of which provide antimicrobial activity against all aerobic Gram-negative
pathogens. An additional short course of systemic third-generation cephalosporins,
such as cefotaxime or ceftriaxone, has also been used for selective digestive
decontamination (SDD) to prevent early infections caused by H. influenzae and
S. pneumonia. Several randomized clinical trials and meta-analyses have shown
reduced bacterial colonization and VAP; however, effects on ICU length of stay
and mortality rates are inconsistent. Several concerns have dampened the enthu-
siasm for the SDD preventive strategy: possible emergence of antibiotic-resistant
bacteria; lack of consistent survival benefits demonstrated by randomized clinical
trials, and, ultimately, increased healthcare costs. A large randomized clinical trial
involving 5,939 patients shows an absolute reduction in mortality of 2.5 and 3.5
percentage points with SOD and SDD, respectively, without evidence of increased
emergence of antibiotic resistance [22]. Currently, they are not recommended for
routine use, especially in patients who may be colonized with multi-drug-resistant
pathogens.
14.6.8 Probiotics
Probiotics are viable microorganisms that colonize the host GI tract by adhering
to the intestinal mucosa and compete with the adhesion of pathogens to epithelial
binding sites, thus creating an unfavorable local milieu for pathogen coloniza-
tion. Probiotic products have been shown to be of some benefit in the following
diseases: acute infectious diarrhea in children, necrotizing enterocolitis in very-
low-birth-weight infants, allergic atopic dermatitis prevention in children, and
prevention of relapses of ulcerative colitis. In critically ill patients, studies
demonstrate that oral administration of a probiotic Lactobacillus preparation
delayed respiratory tract colonization with P. aeruginosa and resulted in a
reduced rate of ventilator-associated pneumonia caused by P. aeruginosa. Also,
Morrow et al. [23] found that patients treated with Lactobacillus were signifi-
cantly less likely to develop microbiologically confirmed VAP compared with
patients treated with placebo (40.0 vs. 19.1%, p = 0.007). A meta-analysis of
five randomized controlled trials concluded that probiotic administration is
associated with lower incidence of VAP [24]. Future studies need to be designed
with standardization of the probiotic product and dosing (both daily dose and
therapy duration).
14.6.9 Bacterial Biofilm
Bacterial biofilm is a highly structured, matrix-enclosed bacterial community.

Sessile bacteria encased within the matrix express genes in a different pattern from
their planktonic counterpart to achieve a survival advantage in a hostile environ-
ment. Studies show evidence that following tracheal intubation, bacterial biofilm is
formed early within the internal surface of the endotracheal tube. These sessile
communities develop resistance to antibiotics, to cellular and humoral immune
defenses, and are the cause of persistent infection. Certain bacteria, such as
Pseudomonas spp., appear to be more capable of forming biofilms, especially in
the presence of abnormal airway mucosa, such as that which as exists in patients
with cystic fibrosis. Bacteria from within the biofilm can be dislodged mechani-
cally through tracheal suction catheters, bronchoscope, or airflow and can ulti-
mately increase the risk for VAP. Coating medical devices, such as intravascular
and urinary catheters, with antimicrobial agents such as silver is a widely applied
method to reduce the incidence of device-associated infections. In the last decade,
several in vitro and in vivo laboratory studies have tested the efficacy and safety of
silver-coated ETTs, demonstrating reduced bacterial colonization of the ETT
without associated adverse effects. However, results from animal studies have also
emphasized that the antibacterial effect may not last beyond 2448 h, mainly due
to mucus accumulation within the ETT. A large randomized clinical trial com-
paring bactericidal effects of silver-coated tracheal tubes to standard tubes showed
a relative VAP risk reduction of 36% and greatest efficacy within the first 10 days
of MV [12]. Silver-coated ETT is an attractive approach to decreasing risk of
pneumonia, and further studies should be performed to improve antimicrobial

efficacy and assess limitations of the strategy.
14.7 Prognostic Factors
Medical conditions predisposing patients to serious infections, such as COPD,

immunosuppression, chronic heart failure, chronic hepatopathy, and chronic renal
failure can have an impact on the severity of VAP episodes. The presence of specific
factors may be associated with poorer outcomes in VAP patients, such as older age,
duration of ventilation before enrolment, presence of neurologic disease on admis-
sion, failure of the PaO2/FiO2 ratio to improve by day 3, acute renal failure, and
shock. But the most important prognostic factor associated with mortality is
appropriate initial antibiotic treatment. The percentage of inadequate treatment
ranges between 22 and 73% in the literature. Multi-drug-resistant microorganisms,
such as P. aeruginosa, Acinetobacter spp., and MRSA, are the more common
pathogens that are not susceptible to initial antibiotic therapy. Susceptibility to
antibiotics of microorganisms that cause VAP varies between patient populations,
hospitals, and ICUs. The most recent American Thoracic Society guidelines [2] list
the following risk factors for colonization and infection with multi-drug-resistant
bacteria: antibiotic treatment within the last 90 days; current hospitalization or
within the last 90 days of[5 days duration; high frequency of multi-drug-resistant
organism in the hospital/unit; presence of risk factors for healthcare-associated
pneumonia (hospitalization for C2 days in the preceding 90 days; residence in a
nursing home or extended care facility; home infusion therapy; chronic dialysis
within the last 30 days; home wound care; family members carriers of multi-drug-
resistant bacteria); immunosuppressive disease and/or treatment.
VAP onset is an important issue regarding the associated mortality risk for ICU
patients. Late-onset VAP has the worst prognosis in comparison with early-onset
pathogens. Typically, late-onset VAP is caused by high-risk microorganisms, and
hospital mortality rates can be as high as 65% when VAP is caused by
P. aeruginosa, Acinetobacter spp. or Stenotrophomonas maltophilia. Bacteremia has
been associated with increased mortality rates in patients with community-acquired
pneumonia, but less information is available for bacteremic episodes of VAP.
Most prognostic factors are included in several scores designed to stratify
patients according to disease severity on ICU admission. Some examples are the
acute physiology and chronic evaluation (APACHE) score versions I, II, III, and
IV; and the simplified acute physiology score (SAPS) versions I, II, and III. When
patients are infected, the following scores can be used: sequential organ failure
assessment (SOFA), a tool that evaluates six organs and has been recognized as a
valuable prognostic scoring system, multiple organ dysfunction score (MODS) and
the organ dysfunction and/or infection (ODIN) score. However, no score has been
developed to assess severity in VAP patients at the time of diagnosis. In the ICU
setting, attending physicians daily have to confront patients in whom a pulmonary
infection can complicate their already critical situation.
A predisposition, infection, response, organ failure (PIRO)-based model could

be useful for assessing severity and stratifying mortality rate. This four-variable
score is based on the patients predisposition to the disease, gravity of the insult,
hosts response, and related organ dysfunction [25]. The VAP PIRO score could be
useful in daily practice, as it classifies patients according to their mortality risk
with only one measurement on the day of the VAP diagnosis. This simple tool has
been tested in different situations and has proven efficiency in assessing VAP
severity and predicting ICU mortality rate. It has also shown worse outcomes in
patients with a score of [2.
References
1. Chastre J, Fagon JY (2002) Ventilator-associated pneumonia. Am J Respir Crit Care Med
165:867903
2. Niederman MS, Craven DE, Bonten MJ (2005) American Thoracic Society and Infectious
Diseases Society of America (ATS/IDSA) Guidelines for the management of adults with
hospital-acquired, ventilator-associated, and healthcare-associated pneumonia. Am J Respir
3. Garner JS, Jarvis WR, Emori TG et al (1998) CDC definitions for nosocomial infections
1998. Am J Infect Control 16:128140
4. Nseir S, Favory R, Jozefowicz E, VAT Study Group et al (2008) Antimicrobial treatment for
ventilator-associated tracheobronchitis: a randomized, controlled, multicenter study. Crit
Care 12:R62
5. Torres A, Ewig S, Lode H et al (2009) Defining, treating and preventing hospital acquired
pneumonia: European perspective. Intensive Care Med 35:929
6. National Nosocomial Infections Surveillance (NNIS) (2004) System Report, data summary
from January 1992 through June 2004, issued October 2004. Am J Infect Control 32:470485
7. Vincent JL, Bihari DJ, Suter PM et al (1995) The prevalence of nosocomial infection in
intensive care units in Europe. Results of the European prevalence of infection in intensive
care (EPIC) study. EPIC international advisory committee. JAMA 274:639644
8. Rello J, Ollendorf DA, Oster G et al (2002) Epidemiology and outcomes of ventilator-
associated pneumonia in a large US database. Chest 122:21152121
9. Fagon JY, Chastre J, Wolff M et al (2000) Invasive and noninvasive strategies for
management of suspected ventilator-associated pneumonia. Ann Intern Med 132:621630
10. Kollef MH, Morrow LE, Niederman MS et al (2006) Clinical characteristics and treatment
patterns among patients with ventilator-associated pneumonia. Chest 129:12101218
11. Agbaht K, Diaz E, Muoz E et al (2007) Bacteremia in patients with ventilator-associated
pneumonia is associated with increased mortality: a study comparing bacteremic vs
nonbacteremic ventilator-associated pneumonia. Crit Care Med 35:20642070
12. Kollef MH, Afeas B, Anzuelo A et al (2008) NASCENT investigation group silver-coated
endotracheal tubes and incidence of ventilator-associated pneumonia: the NASCENT
randomized trial. JAMA 300:805813
13. Rosenthal VD, Maki DG, Jamulitrat S, INICC members et al (2010) International nosocomial
infection control consortium (INICC) report, data summary for 20032008, issued June 2009.
Am J Infect Control 38(2):95104.e2
14. Resar R, Pronovost P, Haraden C et al (2005) Using a bundle approach to improve ventilator
care processes and reduce ventilator-associated pneumonia. Jt Comm J Qual Patient Saf
31:243248
15. Torres A, Gatell JM, Aznar E et al (1995) Re-intubation increases the risk of nosocomial
pneumonia in patients needing mechanical ventilation. Am J Respir Crit Care Med 152:
137141
16. Valles J, Artigas A, Rello J et al (1995) Continuous aspiration of subglottic secretions in

preventing ventilator-associated pneumonia. Ann Intern Med 122:179186
17. Romero CM, Marambio A, Larrondo J et al (2010) Swallowing dysfunction in nonneurologic
critically ill patients who require percutaneous dilatational tracheostomy. Chest 137:
12781282
18. Drakulovic MB, Torres A, Bauer TT et al (1999) Supine body position as a risk factor for
nosocomial pneumonia in mechanically ventilated patients: a randomised trial. Lancet 354:
18511858
19. Heyland DK, Drover JW, Dhaliwal R, Greenwood J (2002) Optimizing the benefits and
minimizing the risks of enteral nutrition in the critically ill: role of small bowel feeding.
JPEN J Parenter Enteral Nutr 26:S51S55
20. Cook DJ, Fuller HD, Guyatt GH et al (1994) Risk factors for gastrointestinal bleeding in
critically ill patients. N Engl J Med 330:377381
21. Bonten MJM, Bergmans DCJJ, Ambergen AW et al (1996) Risk factors for pneumonia, and
colonization of respiratory tract and stomach in mechanically ventilated ICU patients. Am J
23. Morrow LE, Kollef MH, Casale TB (2010) Probiotic prophylaxis of ventilator-associated
pneumonia: a blinded, randomized, controlled trial. Am J Respir Crit Care Med 182(8):
10581064
24. Siempos II, Ntaidou TK, Falagas ME (2010) Impact of the administration of probiotics on the
incidence of ventilator-associated pneumonia: a meta-analysis of randomized controlled
trials. Crit Care Med 38(3):954962
25. Lisboa T, Diaz E, Sa-Borges M et al (2008) The ventilator-associated pneumonia PIRO
score: a tool for predicting ICU mortality and health-care resources use in ventilator-
associated pneumonia. Chest 134:12081216
Bloodstream Infection in the ICU
Patient 15
J. Valles and R. Ferrer
15.1 Introduction
Hospital-acquired infections (HAI) occur in 510% of patients admitted to

hospitals in the Unites States and remain a leading cause of morbidity and mor-
tality [1]. The endemic rates of HAI vary markedly between hospitals and between
areas of the same hospital. Patients in intensive care units (ICUs), representing
815% of hospital admissions, experience a disproportionately high percentage of
HAI compared with patients in noncritical care areas [28]. Patients admitted to
ICUs account for 45% of all HA pneumonias and bloodstream infections (BSIs),
although critical care units comprise only 510% of all hospital beds [3]. The
severity of the underlying disease, invasive diagnostic and therapeutic procedures
that breach normal host defenses, contaminated life-support equipment, and the
prevalence of resistant microorganisms are critical factors in the high rate of
infection in the ICUs [9]. On the other hand, 40% of patients admitted to the ICU
present infections acquired in the community, and 17% of them present BSI [10].
The incidence rate of patients with community-acquired (CA) BSI admitted in a
general ICU is about nine to ten episodes per 1,000 admissions [11, 12] repre-
senting 3040% of all episodes of BSI in a medicalsurgical ICU (Fig. 15.1).
In this chapter, we discuss the characteristics and prognosis of BSI in the ICU,
including hospital- and CA episodes.
J. Valls (&)
Critical Care Center, Hospital Sabadell,
Sabadell, Barcelona, Spain
234 J. Valles and R. Ferrer
70
60
50
40
30
20
10
0
2000 2001 2002 2003 2004 2005 2006 2007 2008 2009
HAI-ICU HAI-non ICU CA
Fig. 15.1 Distribution of bloodstream infections (BSI) in the medicalsurgical intensive care
unit (ICU) of Hospital Sabadell (20002009) HAI-ICU, hospital-acquired BSI in ICU; HAI-non
ICU, hospital-acquired BSI in wards; CA, community-acquired BSI
15.2 Hospital-Acquired Bloodstream Infections in the ICU
15.2.1 Epidemiology
HA BSI in the ICU is defined in a patient with a clinically significant blood culture
positive for a bacterium or fungus and that is obtained more than 72 h after
admission or previously, if it is directly related to a invasive manipulation on
admission in the ICU (e.g., urinary catheterization or insertion of intravenous line)
[13]. Patients in the ICU not only have higher endemic rates of HAI than patients
in general wards, but the distribution of their infections also differs. The two most
important HAI in general wards are urinary tract and surgical wound infections,
whereas in the ICU, lower respiratory tract infections and BSI are the most
frequent [14]. This distribution is related to the widespread use of mechanical
ventilation and intravenous catheters. Data compiled through the national noso-
comial infections surveillance system (NNIS) of the centers for disease control and
prevention (CDC) in the USA revealed that bloodstream infections accounted for
almost 20% of HAI in ICU patients, 87% of which were associated with a central
line [15]. A recent nationwide surveillance study in 49 US hospitals (SCOPE)
reported that 51% of HA BSIs occurred in the ICU [16]. Studies conducted in
critically ill patients show that the incidence rate of nosocomial BSI in the ICU
ranges from 27 to 68 episodes per 1,000 admissions [1721] (Table 15.1),
depending on the type of ICU (surgical, medical, coronary care unit), severity of
patients illness, use of invasive devices, and the length of ICU stay. These
infection rates among ICU patients are as much as five to ten times higher than
those recorded for patients admitted to general wards.
15 Bloodstream Infection in the ICU Patient 235
Table 15.1 Rates of hospital-acquired bloodstream infections (BSIs) in the intensive care
unit (ICU)
Year Type of ICU Episodes of nosocomial First author

BSIs per 1,000 admissions
1994 Medicalsurgical 67.2 Rello [17]
1994 Surgical 26.7 Pittet [19]
1996 Adult, multicenter study 41 Brun-Buisson [20]
1997 Adult, multicenter study 36 Valls [18]
2006 Adult, multicenter study 68 Garrouste-Orgas [21]
15.2.2 Risk Factors
Conditions that predispose an individual to BSI include not only the patients
underlying conditions but also therapeutic, microbial, and environmental factors.
Illnesses that have been associated with an increased BSI risk include hematologic
and nonhematologic malignancies, diabetes mellitus, renal failure requiring dial-
ysis, chronic hepatic failure, immune deficiency syndromes, and conditions
associated with the loss of normal skin barriers, such as serious burns and pressure
ulcers. In the ICU, therapeutic maneuvers associated with an increased risk of HA
BSI include procedures such as placement of intravascular and urinary catheters,
endoscopic procedures, and drainage of intra-abdominal infections. Several risk
factors have been associated with the acquisition of BSI by specific pathogens.
Coagulase-negative staphylococci are mainly associated with central venous line
infection and with the use of intravenously administered lipid emulsions. Candida
spp. infections are related to exposure to multiple antibiotics, hemodialysis,
isolation of Candida spp. from sites other than the blood, azotemia, and the use of
indwelling catheters [22]. In an analysis of risk factors for HA candidemia in our
ICU, we found that exposure to more than four antibiotics during the ICU stay
[odds ratio (OR) 4.10], parenterally administered nutrition (OR 3.37), previous
surgery (OR 2.60), and the presence of solid malignancy (OR 1.57) were the
variables that were independently associated with the development of Candida
spp. infection [23].
15.2.3 Microbiology
The spectrum of microorganisms that invade the bloodstream in patients with HAI
during their stay in the ICU has been evaluated in several studies. Although almost
any microorganism can produce BSI, staphylococci and Gram-negative bacilli
account for the vast majority of cases. However, among the staphylococci, coag-
ulase-negative staphylococci (CNS) have become a clinically significant agent of
BSIs in the ICU [17, 18, 24, 25]. The ascendance of this group of staphylococci has
increased the interpretative difficulties for clinicians, as a high number of CNS
Table 15.2 Microorganisms causing nosocomial bloodstream infection in adult intensive care
units
Reference Gram-positive Gram-negative Fungi Polymicrobial

microorganisms microorganisms episodes (%)
Rello [17] 44.1% CNS 40.5% 5.4% 9.9
S. aureusa P. aeruginosab Candida
Enterococcus E. coli spp.
spp. Enterobacter spp.
Pittet [19] 51.0% CNS 39.0% 4.8% 21
S. aureusa Enterobacter spp. Candida
Enterococcus Klebsiella spp. spp.
spp. S. marcescensc
Valls[18] 49.8% CNS 32.6% 4.4% 12.7
S. aureusa P. aeruginosab Candida
Enterococcus A. baumanniid spp.
spp. K. pneumoniaee
Jamal [27] 46.8% CNS 36.6% 17.6% 9.8
S. aureusa Enterobacter spp. Candida
Enterococcus S. marcescensc spp.
spp. K. pneumoniaee
Garrouste-Orgas 52.5% ECN 29.3% 6.6% 11.6
[21] S. aureusa Enterobacter spp. Candida
Enterococcus P. aeruginosab spp.
spp. Other
CNS coagulase-negative staphylococci
a
Staphylococcus
b
Pseudomonas
c
Serratia
d
Acinetobacter
e
Klebsiella
isolations represent contamination rather than true BSI. The increased importance
of CNS BSI seems to be related to the high incidence of multiple invasive devices
used in critically ill patients and to the multiple antimicrobial therapies used for
Gram-negative infections in ICU patients, which results in selection of Gram-
positive microorganisms. The change in the spectrum of organisms causing HA
BSIs in an adult ICU is confirmed by Edgeworth et al. [26], who analyzed the
evolution of HA BSIs over 25 years in the same ICU. Between 1971 and 1990, the
frequency of isolation of individual organisms changed little, with S. aureus,
P. aeruginosa, Escherichia coli, and Klebsiella pneumoniae predominating.
However, between 1991 and 1995, the number of BSIs doubled, largely due to the
increased isolation of CNS, Enterococcus spp., and intrinsically antibiotic-resistant
Gram-negative organisms, particularly P. aeruginosa and Candida spp.
The leading pathogens among cases of HA BSIs in the ICU are Gram-positive
microorganisms, representing nearly half of the organisms isolated [1719, 21, 27]
(Table 15.2). CNS, S. aureus, and enterococci are the most frequent Gram-positive
bacteria in all studies, and CNS is isolated in 2030% of all episodes of BSI.
Table 15.3 Major sources of hospital-acquired bloodstream infection in the ICUs
Type of infection Rello [17] Pittet [19] Valls [18] Edgewort [26] Garrouste-Orgas
(%) (%) (%) (%) [21] (%)
Intravenous catheter 35 18 37.1 62 20.2
Respiratory tract 10 28 17.5 3 16.3
Intra-abdominal 9 NA 6.1 6.9 NA
Genitourinary tract 3.6 5.4 5.9 2.4 2.5
Surgical wound 8 8 2.4 3 9.9
Other 7 14.5 2.9 12.9
Unknown origin 27 20 28.1 22.4 32.7
Gram-negative bacilli are responsible for 3040% of BSI episodes, and the
remaining cases are mostly due to Candida spp. Polymicrobial episodes are
relatively common, representing about 10%. Anaerobic bacteria are isolated in
fewer than 5% of cases. Among Gram-positive BSIs, the incidence of pathogens is
similar in the different ICUs, with CNS being the most frequently isolated
organism and S. aureus the second commonest pathogen in all studies. Only the
incidence of strains with antibiotic resistance, such as methicillin-resistant
Staphylococcus aureus (MRSA) or vancomycin-resistant enterococci (VRE), dif-
fers substantially according to the characteristics of individual institutions and
depending on whether they become established as endemic nosocomial pathogens
in the ICU. On the other hand, the Gram-negative species isolated from HA BSIs
in ICUs of different institutions show marked variability. The relative contribution
of each Gram-negative species to the total number of isolates from blood varies
from hospital to hospital and over time. The antibiotic policy of the institution may
induce the appearance of highly drug-resistant microorganisms and the emergence
of endemic nosocomial pathogens, in particular, Pseudomonas spp, Acinetobacter
spp., and Enterobacteriaceae, with extended-spectrum beta-lactamase (ESBL).
15.2.4 Sources
The vast majority (70%) of nosocomial BSIs in the ICU are secondary bactere-
mias, including the BSIs related to intravascular catheter infection, and the
remaining 30% are bacteremias of unknown origin. Table 15.3 summarizes the
sources of nosocomial bacteremias in the ICU reported in several series [1719,
21, 26]. As shown, intravascular catheter-related infections and respiratory tract
infections are the leading sources of secondary episodes. The source of nosocomial
BSIs varies according to microorganism. Coagulase-negative staphylococci and S.
aureus commonly complicate intravenous-related infections, whereas Gram-
negative bacilli are the main etiology for secondary BSIs following respiratory
tract, intra-abdominal, and urinary tract infections. Among bacteremias of
unknown origin, most are caused by Gram-positive microorganisms, mainly CNS,
and may originate also in device-related infections not diagnosed at the time of
BSI development.
15.2.5 Systemic Response
The host reaction to invading microbes involves a rapidly amplifying polyphony of

signals and responses that may spread beyond the invaded tissue. Fever or
hypothermia, chills, tachypnea, and tachycardia often herald the onset of the
systemic inflammatory response to microbial invasion, also called sepsis. BSI and
fungemia have been simply defined as the presence of bacteria or fungi in blood
cultures, and four stages of systemic response of increasing severity have been
described: the systemic inflammatory response syndrome (SIRS), which is iden-
tified by a combination of simple and readily available clinical signs and symp-
toms (i.e., fever or hypothermia, tachycardia, tachypnea, changes in blood
leukocyte count); sepsis, in patients in whom the SIRS is caused by documented
infection; severe sepsis, when patients have dysfunction of the major organs;
septic shock, which describes patients with hypotension and organ dysfunction in
addition to sepsis [28]. The presence of organisms in the blood is one of the most
reliable criteria for characterizing a patient presenting with SIRS as having sepsis
or one of its more severe presentations, such as severe sepsis or septic shock.
In a multicenter study, Brun-Buisson et al. [20] analyzed the relationship
between BSI and severe sepsis in adults ICUs and general wards in 24 hospitals in
France. Of the 842 episodes of clinically significant BSI recorded, 162 (19%)
occurred in patients hospitalized in ICUs. Three hundred and seventy-seven epi-
sodes (45%) of BSIs were HA, and their incidence was 12 times greater in ICUs
than in wards. The frequency of severe sepsis during BSI differed markedly
between wards and ICUs (17% vs. 65%, p \ 0.001). HA episodes in the ICU
represented an incidence rate of 41/1,000 admissions, and the incidence rate of
severe sepsis among patients with HA BSI in the ICU was 24 episodes per 1,000
admissions. A multicenter study reported by our group [18] analyzed 590 HA BSIs
in adult ICUs of 30 hospitals in Spain and classified their systemic response as
sepsis in 371 episodes (62.8%), severe sepsis in 109 (18.5%), and septic shock in
the remaining 110 (18.6%). Episodes of BSI associated with intravascular cathe-
ters showed the lowest rate of septic shock (12.8%); episodes of BSI secondary to
lower respiratory tract, intra-abdominal, or genitourinary tract infections showed
the highest incidence of severe sepsis and septic shock. In the study by Brun-
Buisson et al. [20] involving patients hospitalized in ICUs, intravascular catheter-
related BSI was also associated with a lower risk of severe sepsis [OR 0.2; 95%
confidence interval (CI) 0.10.5; p \ 0.01). Systemic response may differ
according to the microorganism causing the episode. Gram-negative and Candida
spp. have been associated with a higher incidence of severe sepsis and septic shock
[18], whereas CNS caused the lowest incidence of septic shock. In the French
multicenter study, episodes caused by CNS were also associated with a reduced
risk of severe sepsis (OR 0.2; p = 0.02) relative to other microorganisms [20].
15.2.6 Prognosis
HA BSIs remain a leading cause of morbidity and mortality in critically ill

patients. The crude mortality rate related to HA BSIs in ICU patients ranges from
20 to 60%, and the mortality rate directly attributable to BSI infection ranges from
14 to 38% [1721]. Although one-third of deaths occur within the first 48 h after
symptom onset, death can occur 14 or more days later. Late deaths are often due to
poorly controlled infection, complications during ICU stay, or failure of multiple
organs. Bueno-Cavanillas et al. [29] analyzed the impact of HAI on the mortality
rate in an ICU. Overall crude relative risk (RR) of mortality was 2.48 (95% CI
1.474.16) in patients with a HAI compared with noninfected patients, and 4.13
(95% CI 2.118.11) in patients with BSI. In a matched, risk-adjusted multicenter
study in 12 ICUs, Garrouste-Orgas et al. [21] found that HA BSI was associated
with a threefold increase in mortality.
The risk of dying is influenced by the patients prior clinical condition and the
rate at which complications develop. Analysis using prognostic stratification
systems (such as Acute Physiology and Chronic Health Evaluation (APACHE) or
the Simplified Acute Physiological Score (SAPS) II) indicate that factoring in the
patients age and certain physiologic variables results in more accurate estimates
of the risk of dying. Variables associated with the high-care fatality rates include
acute respiratory distress syndrome (ARDS), disseminated intravascular coagula-
tion (DIC), renal insufficiency, and multiple organ dysfunction (MOD). Microbial
variables are less important, although high-care fatality rates have been observed
for patients with BSI due to P. aeruginosa and Candida spp., and for patients with
polymicrobial BSI [21].
15.2.7 Prevention
Indwelling vascular catheters are a leading source of BSIs in critically ill

patients. More than 250,000 vascular-catheter-related BSIs (CR-BSI) occur
annually in the USA [3032], resulting in substantial morbidity and mortality
rates and costs [3335]. Despite the publication of clinical practice guidelines
[30] on managing and preventing intravascular catheter-related infection,
CR-BSI are common. According to the NNIS system of the CDC, the median
rate of all types of CR-BSI ranges from 1.8 to 5.2 episodes per 1,000 catheter
days. In Spain, the mean rate of CR-BSI in the National Study of Nosocomial
Infections Surveillance in the ICU (ENVIN-UCI) in 2006 was five episodes per
1,000 catheter days [36]. In our medicalsurgical ICU in 2006, central venous
catheters (CVC) were used in 83% of patients, and the incidence of CR-BSI was
5.8 episodes per 1,000 catheter-days.
Pronovost et al. [37] implemented an evidence-based intervention in 108 ICUs to
reduce CR-BSI, designating a team leader for each hospital instructed in the different
interventions and responsible for disseminating this information among their col-
leagues. The intervention consisted of five evidence-based procedures recommended
by the CDC: hand washing, using full-barrier precautions during CVC insertion,
cleaning the skin with chlorhexidine, avoiding the femoral site if possible, and
removing unnecessary catheters. A checklist was used to ensure adherence to
infection-control practices. Three months after implementing the intervention, their
median rate of CR-BSI had decreased from 2.7/1,000 catheter days at baseline to 0/
1,000 catheter days (p \ 0.002), and their mean rate had decreased from 7.7/1,000
catheter days at baseline to 1.4/1,000 catheter days (p \ 0.002). This improvement
was maintained throughout the 18-month study period.
In 2007 in our ICU, we implemented a similar multiple-system intervention
applying evidence-based measures and reduced the incidence of catheter-relate
BSI from 6.7/1,000 catheter days to 2.4/1,000 catheter days (RR 0.36; 95% CI
0.160.80; p = 0.015), with a 20% reduction in the incidence of HA BSIs in the
ICU [38] (Fig. 15.1).
15.3 Community-Acquired Bloodstream

Infections in the ICU
15.3.1 Epidemiology
CA BSI is defined as infection that develops in a patient prior to if the bacteremia

develops within the first 48 h of hospital and ICU admission and is not associated
with any procedure performed after admission. CAI represent an important reason
for ICU admission. Severe CA pneumonia and intra-abdominal infections are the
most frequent, and approximately 20% also present bacteremia. A few epidemiol-
ogic studies focusing solely on CA BSI on ICU admission are available. Data from a
multicenter study reported a CA BSIs rate of 10.2/1,000 ICU admissions [39].
15.3.2 Microbiology
In CA BSIs in patients admitted to the ICU, the incidence of Gram-positive is

similar to Gram-negative microorganisms, and almost 10% are polymicrobial
episodes. E. coli, Streptococcus pneumoniae, and S. aureus are the leading
pathogens, and the prevalence of these microorganisms is related to the main
sources of BSIs found in these patients, such as urinary tract and pulmonary tract
infections and those of unknown origin [11, 12, 39] (Table 15.4).
15.3.3 Sources
Among CA BSIs, lower respiratory tract, intra-abdominal, and genitourinary

infections represent [80% of episodes of bacteremia admitted in the ICU
(Table 15.4). Approximately 2029% of episodes are of unknown origin, including
mainly meningococcal and staphylococcal infections [11, 12, 39].
Table 15.4 Microorganisms and sources of community-acquired bloodstream infections

admitted to the intensive care unit
Reference Sources (%) Microorganisms (%)

Forgacs [11] Pulmonary 38.5 S. pneumoniaea 32.3
Genitourinary 23.0 E. colib 27.2
c
Endocarditis 8.0 S. aureus 13.5
Biliary tract 5.9 Other GNB 14.2
Other 11.1 Other GPC 8.2
Unknown origin 20.0 Other 14.2
Valls [39] Pulmonary 20.0 E. colib 28.1
a
Abdominal 20.1 S. pneumoniae 17.9
Genitourinary 19.8 S.aureusc 14.9
Other 10.3 Other GNB 18.6
Unknown origin 29.2 Other GPC 9.5
Other 11.07
GNB Gram-negative bacilli; GPC Gram-positive cocci
a
Streptococcus
b
Escherichia
c
Staphylococcus
The incidence of severe sepsis and septic shock in patients with CA BSIs is higher
than in HA episodes, in part because the severity of the systemic response is the
reason for ICU admission. In a multicenter French study, 74% of CA BSI episodes
presented severe sepsis or septic shock at admission [20]. In a multicenter Spanish
study carried out in 30 ICUs, the incidence of severe sepsis and septic shock was
75%. In that study, Gram-negative microorganisms and urinary tract and intra-
abdominal infections were associated more frequently with septic shock [39].
15.3.5 Prognosis
Patients admitted to the ICU with CA BSI present a crude mortality rate close to
40%, compared with 18% in patients admitted to general wards [12, 39, 40]. This
is due in part to the severity of systemic response (severe sepsis and septic shock)
and associated complications, which is the reason for admission [12, 39]. The
appropriateness of empiric antimicrobial treatment is the most important variable
influencing the outcome of these patients [12, 39]. The incidence of inappropriate
antibiotic treatment in CA BSIs ICU patients ranges between 15% and 20% in two
studies, and the mortality rate among patients with inappropriate empiric antibiotic
treatment is [70% [12, 39, 41].
15.4 Health-Care-Associated Bloodstream

Infections in the ICU
15.4.1 Epidemiology
Patients residing in the community and who are receiving care at home, living in
personal-care and rehabilitation centers, receiving chronic dialysis, and receiving
chemotherapy in physicians offices may present BSIs, which have traditionally
been categorized as CA infections. However, the difference between community-
or HA BSIs has become less clear. Therefore, some investigators propose a new
classificationhealthcare-associated infectionsthat are distinct from CAI and
HAI [42, 43]. Health-care-associated BSI has been defined when a positive blood
culture is obtained from a patient at the time of or within 48 h of hospital
admission if the patient fulfilled any of the following criteria:
1. received intravenous therapy at home, received wound care or specialized
nursing care, or had self-administered intravenously administered medical
therapy;
2. attended a hospital hemodialysis clinic or received chemotherapy intravenously;
3. was hospitalized in an acute-care hospital for C2 days in the 90 days before the
BSI;
4. resided in a personal- or long-term-care facility [43].
According to this new classification, approximately 4050% of patients
admitted with BSI traditionally defined as CA should be classified as having
healthcare-associated BSI [42, 43]. In a large US database, healthcare-associated
BSI accounted for more than half of all BSIs. If patients with healthcare-associated
BSI had been included in the CA BSI category, according to the traditional
classification scheme, they would have accounted for approximately 60% of CA
BSI patients who needed hospitalization [44].
There are no studies available regarding the importance of healthcare-
associated BSI in ICU patients using this new classification. However, recent
studies suggest that healthcare-associated BSIs are less frequent than HA- and CA
BSI in critically ill patients. In a multicenter study carried out in three hospitals in
Spain, 1,157 BSI episodes were studied: 50% were CA, 26% HA, and 24%
healthcare-associated BSI [45]. In patients admitted to the ICU, 60% of BSIs were
HA, 30% CA, and 10% healthcare associated.
We conducted a multicenter study in 28 ICUs in Spain (unpublished data)
analyzing 1,590 BSI episodes and confirmed the low incidence of healthcare-
associated BSI in ICU patients compared with patients admitted to conventional
wards. The most frequent BSIs were HA (77%), CA (21%), and healthcare
associated (8%). Compared with patients with CA episodes, patients with
healthcare-associated BSI were older and more likely to have severe comorbidi-
ties, such as congestive heart failure, peripheral vascular disease, chronic renal
disease, and cancer. This high number of comorbidities and the patients basal
condition may be the reason for a lower ICU admission rate of these patients due
Table 15.5 Pathogens most frequently found in bloodstream infections through epidemiologic
type of infection in a Spanish multicenter study [45]
Pathogen Total CBSI HBSI HCBSI P value

n = 1,157 (%) n = 581 (%) n = 295 (%) n = 281 (%)
Escherichia coli 472 (40.8) 308 (53) 61 (20.7) 103 (36.7) \0.001
MSSA 86 (7.4) 25 (4.3) 31 (10.5) 30 (10.6) \0.001
CN staphylococci 62 (5.3) 4 (0.7) 49 (16.6) 9 (3.2) \0.001
Streptococcus 80 (6.9) 64 (11) 5 (1.7) 11 (3.9) \0.001
pneumoniae
Pseudomonas 63 (5.4) 9 (1.5) 27 (9.2) 27 (9.6) \0.001
aeruginosa
Klebsiella 41 (3.5) 23 (4) 9 (3.1) 9 (3.2) NS
pneumoniae
Candida spp 12 (1) 0 (0) 11 (3.7) 1 (0.3) \0.001
Polymicrobial 72 (6.2) 23 (3.9) 26 (8.8) 23 (8.1) 0.01
CBSI community-acquired bloodstream infection; HBSI hospital-acquired bloodstream infection;
HCBSI healthcare-associated bloodstream infection; MSSA methicillin-sensitive Staphylococcus
aureus; CN coagulase-negative
to a possible limitation of therapeutic effort by physicians in wards or emergency

areas prior to the patients admission to the ICU. Consequently, the number of
patients with healthcare-associated infections admitted in the ICU may be reduced.
15.4.2 Microbiology
The pathogens responsible for healthcare-associated BSI and their susceptibility

patterns are similar to HAI. In a prospective observational study of 504 patients
with BSIs, Friedman et al. [43] found that S. aureus was the most common
pathogen in patients with healthcare-associated and HA bacteremia. MRSA was
present in 19% of healthcare-associated BSI and 20% in HA episodes. Moreover,
ampicillinsulbactam and ciprofloxacin resistance occurred with similar frequency
in Enterobacteriaceae isolated from patients with healthcare-associated and HA
BSI. A US multicenter study found similar results in the frequency of S. aureus
and MRSA in healthcare-associated and HA BSIs [44]. However, the distribution
of pathogens in both groups was not identical. Whereas E. coli and Proteus spp.
were identified more frequently in patients with healthcare-associated infections,
fungal organisms were more prevalent in patients with HA BSI. In our experience
[45], the distribution of pathogens isolated from patients with healthcare-
associated BSI shares more similarity with HA than with CA BSIs (Table 15.5). In
addition, episodes with healthcare-associated BSI had a higher prevalence of
MRSA infections (5%) than CA (0.2%) (OR 30.4; 95% CI 3.9232.4; p \ 0.001)
or HA (0.7%) (OR 7.7; 95% CI 1.734.1; p = 0.001) episodes. A study in Spain

found a high MRSA prevalence (16.8; 95% CI 14.918.8) among residents in
community long-term-care facilities and was isolated from 15.5% of nasal swabs
and 59% of decubitus ulcers [46].
15.4.3 Sources
Urinary tract infection, intravascular-device-related BSI, gastrointestinal-related

bacteremia, and respiratory tract infections are the most frequent sources of
healthcare-associated BSIs [43, 45]. Patients with healthcare-associated and those
with HA BSIs have similar frequencies of intravascular-device-related bacteremia.
In a multicenter study of ICU patients in Spain, 5060% with healthcare-

associated BSI presented septic shock. The frequency of severe sepsis and septic
shock was similar to CA and HA BSI (non-ICU-acquired) that originated in
conventional wards because they were the main causes for ICU admission
(unpublished data).
15.4.5 Prognosis
Friedman et al. [43] found a mortality rate at follow-up greater in patients with
healthcare-associated BSI (29% vs. 16%; p = 0.019) or HA infection (37% vs.
16%; p \ 0.001) than in patients with CA infection. Similar results were found by
Shorr et al. [44], where the mortality risk was significantly higher if BSI was
acquired in the hospital or associated with previous healthcare exposure. Consis-
tent with these reports, we found a significantly higher mortality rate at follow up
in groups with HA BSI (27.3%) and healthcare-associated BSI (27.5%) than in
those with CA BSI (10.4%) (p \ 0.001). Among patients with CA and healthcare-
associated BSIs, a multivariate analysis, adjusted for age and comorbidities,
showed healthcare-associated BSI (OR 2.4; 95% CI 1.53.7; p \ 0.001) as an
independent factor associated with mortality [45].
15.5 Treatment
BSIs are among the most serious infections causing severe sepsis or septic shock
acquired by hospitalized patients requiring intensive care. The mainstay of therapy
for patients with bacteremia remains antimicrobial therapy associated with a
optimal management of consequences of bacteremia, such as shock or metastatic
suppurative complications and surgical treatment, such as debridement or drainage
of abscesses and removing intravascular devices if necessary [47]. Appropriate

antimicrobial therapy has been shown to reduce mortality rates among patients
with bacteremia and, when initiated early, to have a favorable effect on outcome in
critically ill patients [44]. The initial antimicrobial therapy is necessarily empiric
based on targeting the most likely etiologic pathogens. However, inappropriate
empirical treatment occurs in 30% of cases and is more frequent in the following
circumstances: HAI, healthcare-associated infection, prior administration of anti-
biotics, and presence of multi-drug-resistant pathogens.
The distribution of pathogens associated with CA BSIs is relatively uniform.
However, an increase in the incidence of infections due to antibiotic-resistant
microorganisms, such as CA MRSA and infections caused by ESBL-producing
E. coli or K. pneumoniae have been reported in most countries, and these cir-
cumstances must be considered when empiric antibiotic treatment is initiated.
HA and healthcare-associated BSIs are associated with an increased incidence of
resistant microorganisms, such as MRSA, ESBL-producing Enterobacteriaceae,
Acinetobacter baumannii, and P. aeruginosa. In these cases, the appropriateness
of empirical treatment is more difficult, especially in ICU patients. In these
cases, and despite guideline recommendations, identifying the local flora pre-
dominant in each area before initiating empirical antibiotic treatment is
indispensable.
15.5.1 Treatment of Severe Sepsis and Septic Shock
Patients with BSI who develop severe sepsis or septic shock require additional
treatment: early goal-directed therapy (EGDT) [48], corticosteroids for refractory
septic shock [49], recombinant human activated protein C or drotrecogin alfa
(activated) for multiorgan failure [50], and lung-protective ventilation strategies
[51] have all been associated with survival benefits. These and other therapeutic
advances led to the development of the surviving sepsis campaign (SSC) guide-
lines [52]. To improve care for patients with sepsis, the SSC and the Institute for
Healthcare Improvement recommend implementing two sepsis bundles: the
resuscitation bundle and the management bundle (Table 15.6):
1. The resuscitation bundle includes lactate determination, early cultures and
antibiotics, and EGDT. This bundle describes seven tasks that should begin
immediately and must be accomplished within the first 6 h of presentation of
severe sepsis or septic shock. Some items may not be completed if the clinical
conditions described in the bundle do not prevail in a particular case, but
clinicians must assess for these elements.
2. The management bundle includes optimization of glycemic control and respi-
ratory inspiratory plateau pressure and determination of the need for cortico-
steroids or drotrecogin alfa (activated). Efforts to accomplish these goals should
begin immediately, and these items must be completed within 24 h of pre-
sentation of severe sepsis or septic shock.
Table 15.6 Surviving sepsis campaign bundles
Resuscitation bundle (first 6 h)

1. Measure serum lactate
2. Obtain blood cultures prior to antibiotic within 3 h from time of presentation for emergency
department (ED) admissions and 1 h for non-ED intensive care unit (ICU) admission
3. Administer broad-spectrum antibiotics within 3 h from time of presentation for ED
admissions and 1 h for non-ED ICU admission
4. In the event of hypotension and/or lactate [36 mg/dl
a. deliver an initial minimum dose of 20 ml/kg crystalloid (or colloid equivalent)
b. apply vasopressors for hypotension not responding to initial fluid resuscitation to
maintain mean arterial pressure (MAP) 65 mmHg
5. In the event of persistent hypotension despite fluid resuscitation (septic shock) and/or lactate
[36 mg/dl
a. achieve central venous pressure (CVP) of 8 mmHg
b. achieve central venous oxygen saturation (ScvO2) of 70%
Management bundle (first 24 h)
1. Administer low-dose steroids for septic shock in accordance with a standardized ICU policy
2. Administer drotrecogin alfa (activated) in accordance with a standardized ICU policy
3. Maintain glucose control higher than the lower limit of normal but \150 mg/dl
4. Maintain inspiratory plateau pressures [30 cm H2O for mechanically ventilated patients
Several studies have evaluated bundled care, and in a meta-analyses [53]

including all those studies, done by the National Institutes of Health (NIH) in the
United States, bundled care was associated with a consistent and significant
increase in survival rate (OR 1.91; CI 1.492.45; p \ 0.0001). Moreover, an
international effort to implement bundled care improved survival (hospital mor-
tality rates decreased from 37.0 to 30.8% in two years) [54]. According to these
results, bundled care is warranted in all patients with BSI and severe sepsis and
septic shock.
15.6 Summary
BSIs are among the most serious infections causing severe sepsis or septic shock in
hospitalized patients requiring intensive care. Nosocomial BSIs account for almost
20% of HAI in critically ill patients, [80% of which are associated with a central
line. These infection rates among ICU patients are as much as five to ten times
higher than those recorded for patients admitted to general wards. CA infections
represent an important reason for ICU admission: severe pneumonia and urinary
tract and intra-abdominal infections are the most frequent CA infections requiring
ICU admission, and approximately 20% of these patients will also present
bacteremia, which associated with a high incidence of severe sepsis and septic
shock. A new classification scheme for BSIshealthcare-acquired BSIshas been
proposed to distinguish between infections occurring among outpatients having
recurrent or recent contact with the healthcare system and patients with true
CA infections. According to this classification, approximately 4050% of patients
admitted to general hospital wards classified as having CA BSIs should be classified
as having healthcare-associated BSI. However, the rate of healthcare-acquired
BSI among critically ill patients in the ICU seems to be lower (\20% of
community-acquired infections) than among patients admitted to conventional
wards. BSIs in critically ill patients are associated with greater hospital mortality
rates. Clinical efforts should therefore be aimed at improving severe sepsis and
septic shock management, reducing the incidence of inadequate antimicrobial
treatment, and preventing BSI episodes associated with intravascular devices.
References
1. Wenzel RP (1990) Organization for infection control. In: Mandell GL, Douglas RG Jr,
Bennett JE (eds) Principles and practice of infectious diseases. Churchill Livingstone, New
York, pp 21762180
2. Weinstein RA (1991) Epidemiology and control of nosocomial infections in adult intensive
care units. Am J Med 91(Suppl 3B):179S184S
3. Wenzel RP, Thompson RL, Landry SM et al (1983) Hospital-acquired infections in intensive
care unit patients: an overview with emphasis on epidemics. Infect Control 4:371375
4. Maki DG (1989) Risk factors for nosocomial infection in intensive care: devices vs nature
and goals for the next decade. Arch Intern Med 149:3035
5. Donowitz LG, Wenzel RP, Hoyt JW (1982) High risk of hospital-acquired infection in the
ICU patient. Crit Care Med 10:355357
6. Brown RB, Hosmer D, Chen HC et al (1985) A comparison of infections in the different ICUs
within the same hospital. Crit Care Med 13:472476
7. Daschner F (1985) Nosocomial infections in intensive care units. Intensive Care Med
11:284287
8. Trilla A, Gatell JM, Mensa J et al (1991) Risk factors for nosocomial bacteremia in a large
Spanish teaching hospital: a case-control study. Infect Control Hosp Epidemiol 12:150156
9. Massanari RM, Hierholzer WJ Jr (1986) The intensive care unit. In: Bennett JV, Brachman
PS (eds) Hospital infections. Little, Brown and Company, Boston, pp 285298
10. Ponce de Len-Rosales S, Molinar-Ramos F, Domnguez-Cherit G et al (2000) Prevalence of
infections in intensive care units in Mexico: a multicenter study. Crit Care Med 28:13161321
11. Forgacs IC, Eykyn SJ, Bradley RD (1986) Serious infection in the intensive therapy unit: a
15-year study of bacteraemia. Q J Med 60:773779
12. Valls J, Ochagava A, Ru M et al (2000) Critically ill patients with community-acquired
bacteremia: characteristics and prognosis. Intensive Care Med 26(Suppl. 3):S222
13. Garner JS, Jarvis WR, Emori TG et al (1988) CDC definitions for nosocomial infections. Am
J Infect Control 16:128140
14. Trilla A (1994) Epidemiology of nosocomial infections in adult intensive care units. Intensive
Care Med 20:S1S4
intensive care units in the United States. Crit Care Med 27:887892
16. Wisplinghoff H, Bischoff T, Tallent SM et al (2004) Nosocomial bloodstream infections in
US Hospitals: analysis of 24, 179 cases from a prospective nationwide surveillance study.
17. Rello J, Ricart M, Mirelis B et al (1994) Nosocomial bacteremia in a medical-surgical

intensive care unit: epidemiologic characteristics and factors influencing mortality in 111
episodes. Intensive Care Med 20:9498
18. Valls J, Len C, Alvarez-Lerma F et al (1997) Nosocomial bacteremia in critically ill patients:
a multicenter study evaluating epidemiology and prognosis. Clin Infect Dis 24:387395
19. Pittet D, Tarara D, Wenzel RP (1994) Nosocomial bloodstream infection in critically
ill patients. Excess length of stay, extra costs, and attributable mortality. JAMA
271:15981601
20. Brun-Buisson C, Doyon F, Carlet J et al (1996) Bacteremia and severe sepsis in adults: a
multicenter prospective survey in ICUs and wards of 24 hospitals. Am J Respir Crit Care
Med 154:617624
21. Garrouste-Orgeas M, Timsit JF, Tafflet M et al (2006) Excess risk of death from intensive
care unit-acquired nosocomial bloodstream infections: a reappraisal. Clin Infect Dis
42:11181126
22. Wenzel RP (1995) Isolation of Candida species from sites other than the blood. Clin Infect
Dis 20:15311534
23. Daz E, Villagr A, Martnez M et al (1998) Nosocomial candidemia risk factors. Intensive
Care Med 24(Suppl 1):S143
24. Towns ML, Quartey SM, Weinstein MP et al (1993) The clinical significance of positive
blood cultures: a prospective, multicenter evaluation, abstr. C-232. In: Abstracts of the 93rd
general meeting of the American society for microbiology. American Society for
Microbiology, Washington, DC
25. Weinstein MP, Towns ML, Quartey SM et al (1997) The clinical significance of positive
blood cultures in the 1990s: a prospective comprehensive evaluation of the microbiology,
epidemiology, and outcome of bacteremia and fungemia in adults. Clin Infect Dis
24:584602
26. Edgeworth JD, Treacher DF, Eykyn SJ (1999) A 25-year study of nosocomial bacteremia in
an adult intensive care unit. Crit Care Med 27:14211428
27. Jamal WY, El-Din K, Rotimi VO et al (1999) An analysis of hospital-acquired bacteraemia in
intensive care unit patients in a university hospital in Kuwait. J Hosp Infect 43:4956
28. Levy MM, Fink MP, Marshall JC et al (2003) SCCM/ESICM/ACCP/ATS/SIS International
sepsis definitions conference. Intensive Care Med 29:530538
29. Bueno-Cavanillas A, Delgado-Rodrguez M, Lpez-Luque A et al (1994) Influence of
nosocomial infection on mortality rate in an intensive care unit. Crit Care Med 22:5560
30. OGrady N, Patchen Dellinger A, Gerberding J et al (2002) Guidelines for the prevention of
intravascular catheter-related infections. CID 35:12811307
31. Mermel LA (2000) Prevention of intravascular catheter-related infections. Ann Intern Med
132:391402
32. Warren DK, Cosgrove SE, Diekema DJ et al (2006) A multicenter intervention to prevent
catheter-associated bloodstream infections. Infect Control Hosp Epidemiol 27:662669
33. Renaud B, Brun-Buisson C et al (2001) Outcomes of primary and catheter-related
bacteremia. Am J Respir Crit Care Med 163:15841590
34. Blot SI, Depuydt P, Annemans L et al (2005) Clinical and economic outcomes in critically
ill patients with nosocomial catheter-related bloodstream infections. Clin Infect Dis
41:15911598
35. Warren DK, Quadir WW, Hollenbeak CS et al (2006) Attributable cost of catheter-associated
bloodstream infections among intensive care patients in a nonteaching hospital. Crit Care
Med 34:20842089
36. Alvarez-Lerma J, Palomar M, Olaechea P et al (2007) National study of control of
nosocomial infection in intensive care units. Evolutive report of the years 20032005 (Article
in Spanish). Med Intensiva 31:617
37. Pronovost P, Needham D, Berebholtz S et al (2006) An intervention to decrease catheter-
related bloodstream infections in the ICU. NEJM 355:27252732
38. Peredo R, Sabatier C, Villagr A et al (2010) Reduction in catheter-related bloodstream

infections in critically ill patients through a multiple system intervention. Eur J Clin
Microbiol Infect Dis 29(9):11731177
39. Valls J, Rello J, Ochagava A et al (2003) Community-acquired bloodstream infection in
critically ill adult patients: impact of shock and inappropriate antibiotic therapy on survival.
Chest 123:16151624
40. Cartn JA, Garca-Velasco G, Maradona JA (1988) Bacteriemia extrahospitalaria en adultos.
Anlisis prospectivo de 333 episodios. Med Clin (Barc) 90:525530
41. Ibrahim EH, Sherman G, Ward S et al (2000) The influence of inadequate antimicrobial
treatment of bloodstream infections on patient outcomes in the ICU setting. Chest
118:146155
42. Siegman-Igra Y, Fourer B, Orni-Wasserkauf R et al (2002) Reappraisal of community-
acquired bacteremia: a proposal of a new classification for the spectrum of acquisition of
bacteremia. Clin Infect Dis 34:14311439
43. Friedman ND, Kaye KS, Stout JE et al (2002) Health care-associated bloodstream infections
in adults: a reason to change the accepted definition of community-acquired infections. Ann
44. Shorr AF, Tabak YP, Killian AD et al (2006) Healthcare-associated bloodstream infection: a
distinct entity? Insights from a large US database. Crit Care Med 34:25882595
45. Valls J, Calbo E, Anoro E et al (2008) Bloodstream infections in adults: importance of
healthcare-associated infections. J Infect 56:2734
46. Manzur A, Gavalda L, Ruiz de Gopegui E et al (2008) Prevalence of methicillin-resistant
Staphylococcus aureus and factors associated with colonization among residents in
community long-term-care facilities in Spain. Clin Microbiol Infect 14:867872
47. Dellinger RP, Levy MM, Carlet JM et al (2008) Surviving sepsis campaign: international
guidelines for management of severe sepsis and septic shock. Crit Care Med 36:296327
48. Rivers E, Nguyen B, Havstad S et al (2001) Early goal-directed therapy in the treatment of
severe sepsis and septic shock. N Engl J Med 345:13681377
49. Annane D, Sebille V, Charpentier C et al (2002) Effect of treatment with low doses of
hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA
288:862871
50. Bernard GR, Vincent JL, Laterre PF et al (2001) Efficacy and safety of recombinant human
activated protein C for severe sepsis. N Engl J Med 344:699709
51. The Acute Respiratory Distress Syndrome Network (2000) Ventilation with lower tidal
volumes as compared with traditional tidal volumes for acute lung injury and the acute
respiratory distress syndrome. N Engl J Med 342:13011308
52. Dellinger RP, Carlet JM, Masur H et al (2004) Surviving sepsis campaign guidelines for
management of severe sepsis and septic shock. Intensive Care Med 30:536555
53. Barochia AV, Cui X, Vitberg D et al (2010) Bundled care for septic shock: an analysis of
clinical trials. Crit Care Med 38:668678
54. Levy MM, Dellinger RP, Townsend SR et al (2010) The surviving sepsis campaign: results of
an international guideline-based performance improvement program targeting severe sepsis.
Infections of Peritoneum,
Mediastinum, Pleura, Wounds, 16
and Urinary Tract
G. Sganga, G. Brisinda, V. Cozza

and M. Castagneto
16.1 Introduction
Intra-abdominal infections (IAIs) are defined as an inflammatory response of the

peritoneum to microorganisms and their toxins, which results in purulent exudate
in the abdominal cavity [15]. They have two major manifestations: generalized
peritonitis and IA abscess. Peritonitis remains a potentially fatal disease and still
represents a challenge for surgeons [3]. Although a greater understanding of the
pathophysiology of IAIs, improvement in critical care, and timely surgical and/or
radiological intervention have reduced the mortality rate associated with severe
peritonitis, that rate remains unacceptably high, ranging from 3% in localized
abscess to 10% in localized peritonitis, 32% in diffuse suppurative peritonitis, and
7080% in complicated mixed infections [15]. In an effort to improve treatment
results, especially of IAIs resulting from anastomotic leakage or perforation of the
gastrointestinal tract (GIT), new surgical techniques have been introduced.
16.2 Definition and Classification
Many attempts have been made to classify peritonitis in general and secondary
peritonitis in particular, which include a large variety of different pathological
conditions ranging in severity from a local problem to a devastating disease.
A simplified version is reported in Table 16.1. This differentiates the relatively
G. Sganga (&)
Istituto di Clinica Chirurgica, Universit Cattolica del Sacro Cuore,
Rome, Italy
252 G. Sganga et al.
Table 16.1 Classification of intra-abdominal infections (IAIs)
Primary peritonitis
Diffuse bacterial peritonitis in absence of GIT disruption
Spontaneous peritonitis in children
Spontaneous peritonitis in adults
Peritonitis in patients receiving continuous peritoneal dialysis
Tuberculous and other granulomatous peritonitis
Secondary peritonitis
Localized or diffuse peritonitis originating from a defect in GIT
Acute perforation peritonitis (GIT perforation, intestinal ischemia, pelviperitonitis, and other
forms)
Postoperative peritonitis (anastomotic leak, accidental perforation, and devascularization)
Posttraumatic peritonitis (after either blunt or penetrating abdominal trauma)
Tertiary peritonitis
Late peritonitis-like syndrome due to disturbance in the immune response
Peritonitis without evidence of pathogens
Peritonitis with fungi
Peritonitis with low-grade pathogenetic bacteria
GIT gastrointestinal tract
rare forms of primary peritonitis, which usually respond to medical treatment, and
tertiary peritonitis, which does not respond to any treatment, from the commonly
occurring secondary peritonitis that mandates surgical intervention and antibiotic
therapy [2]. IAIs include the following pathological conditions:
1. Infections of single organs (cholecystitis, appendicitis, diverticulitis, cholan-
gitis, pancreatitis, salpingitis, etc.), which may or may not be complicated by
peritonitis, even in the absence of perforation.
2. Peritonitis, classified as primary, secondary or tertiary.
3. Intra-abdominal abscesses, classified on the basis of their location and anatomic
configuration.
The term complicated IAI (C-IAI) is used to indicate infections that, originating
in an organ cavity, extend into the peritoneal space and form an abscess or peri-
tonitis. Resolution of this type of infection requires surgical treatment and per-
cutaneous drainage, as well as systemic antibiotic therapy [6]. They are divided
into two types: community c-IAIs, which can be mild or serious; and hospital c-
IAIs, which usually occur as postoperative infections. The nature of severe IAIs
makes it difficult to precisely define the disease, to assess its severity, and to
evaluate and compare therapeutic progress. Both the anatomical source of infec-
tion and, to a greater degree, the physiological compromise it inflicts, affects the
outcome of IAI.
16 Infections of Peritoneum, Mediastinum, Pleura, Wounds, and Urinary Tract 253
16.3 Microbiology
Primary peritonitis is a monomicrobial, aerobic infection [2] caused by Strep-

tococcus pneumoniae. The underlying risk factor most frequently encountered is
the presence of cirrhosis and ascites. Bacteriological cultures grow Escherichia
coli, whereas anaerobes are rare, and their presence, or a mixed flora, suggests
secondary peritonitis [2]. The latter represents a polymicrobial infection (average
of four isolates per patient, with the most frequent combination being E. coli and
Bacteroides fragilis) after a spontaneous or traumatic breach in a viscous-con-
taining microorganism, or due to postoperative breakdown of GIT anastomosis
[7, 8].
The quantity and variety of microorganisms increase progressively the more the
lesion is in a distal part of the GIT: the proximal anatomical regions (stomach and
duodenum are usually sterile) usually contain aerobic coliform flora with a small
anaerobic component (\104 CFU). In the more distal regions (for example, the
colon), the intestine contains a higher concentration of bacteria (1012 obligatory
anaerobes and 108 facultative anaerobes in 1 g of feces). After colon perforation,
the peritoneal cavity can be invaded by [400 different bacterial species, but only
some of these are directly responsible for infectious processes. Prolonged hospital
stay, especially in ICU, repeated surgery, and administration of systemic or
intraluminal antibiotic treatment, may drastically modify the patients flora ecol-
ogy, resulting in colonization of the proximal GIT with abnormal microorganisms
(fungi, Gram-negative bacteria of low pathogenicity), which may be found in
tertiary peritonitis [9], in critically ill patients, in ICU infections, and in multiple
organ failure (MOF) [2].
16.4 Inflammatory Response in Peritonitis
The outcome of peritonitis depends on the results of a struggle between the

patients systemic and local defenses [systemic inflammatory response syndrome
(SIRS)] on one hand, and the nature, volume, and duration of bacterial con-
tamination on the other. The exact events that follow invasion of the peritoneal
cavity with bacteria and adjuvants of infection (blood, bile, barium sulfate), and,
subsequently, its translymphatic spread, have been clearly defined in recent
studies [14, 10, 11]. Microorganisms and their products stimulate the hosts
cellular defenses to activate several inflammatory mediators, which are respon-
sible for sepsis. Bacterial peritonitis appears to induce an intense compartmen-
talized inflammatory process [10]. A key element is cytokines [proinflammatory:
tumor necrosis factor (TNF), interleukin (IL)-1 and -8; anti-inflammatory: IL-6
and -10], and host-produced pleomorphic immunoregulatory peptides [10, 11].
A trigger, such as a microbial toxin, stimulates production of TNF and IL-1,
which in turn promote endothelial-cell-leukocyte adhesion, release of proteases
and arachidonate metabolites, and clotting activation. IL-1 and TNF are syner-
gistic and share many biological effects, and their inhibition improves organ
function and survival. IL-8 may have an especially important role in perpetuating
tissue inflammation. IL-6 and -10, which are perhaps counterregulatory, inhibit
the generation of TNF, augment the action of acute phase reactants and
immunoglobulins, and inhibit T-lymphocyte and macrophage function [11].
Cytokines are measurable in the systemic circulation and in peritoneal exudates,
and the magnitude of the phenomena is negatively correlated with outcome [2,
3]. Most peritoneal cytokines probably derive from macrophages exposed to
bacterial endotoxin [12]. Recent findings suggest that female sex hormones play
a critical role in maintaining the immune response after trauma hemorrhage by
suppressing the production of TNF-a and preventing increased mortality rates
from subsequent sepsis [12]. Other potential sources are direct translocation of
cytokines through the GIT barrier or production by traumatized tissues [13].
Timely therapeutic intervention is crucial to abort the ensuing, self-perpetuating,
SIRS, sequential MOF, and death [2, 1012].
16.5 Diagnosis
16.5.1 Clinical Diagnosis
Clinical diagnosis of intra-abdominal sepsis is usually assisted by the presence

of systemic signs of inflammation (such as fever, neutrophil leukocytosis) and
by local abdominal symptoms (defense, rigidity and respiratory hypomobility,
absence of peristalsis, paralytic ileum), possibly with the assistance of some
diagnostic imaging techniques [abdominal X-ray, ultrasonography (US), com-
puted tomography (CT), magnetic resonance imaging (MRI)]. The typical
presentation of secondary peritonitis is spontaneous acute abdominal pain,
exacerbated by breathing and movement, accompanied by abdominal distension,
fever, nausea or vomiting and irregular bowel sounds. Diffused pain is sug-
gestive of generalized peritonitis, whereas localized pain is more indicative of a
process involving an organ or its immediate environment. Diagnosis is often
difficult, especially in elderly or immunocompromised patients, because the
clinical presentation is less precise; or in those with altered conscience state or
sedated for mechanical ventilation [14]. Acute abdominal complications are
considered infrequent events in the ICU, although they are characterized by
serious prognosis when they do appear. The absence of clinical signs and
typical symptoms, with consequent delay in diagnosis and treatment, has
assigned to IAIs the role of silent offender [15, 16]. Occult IAI can also
manifest as renal dysfunction or elevated bilirubin and transaminases. Blood
culture is often negative, and the absence of bacteremia in a surgery patient
with fever increases the probability of IAI. Polymicrobial or anaerobic bac-
teremia also suggests the presence of an intraperitoneal infection. Differential
diagnosis of unexpected difficulty in breathing or supraventricular arrhythmia
occurring 34 days after abdominal surgery must include suture dehiscence or
other intra-abdominal infectious pathology, and their diagnosis should be
excluded through appropriate imaging techniques [17]. Biohumoral data are

generally not very specific and do not permit diagnosis, but they can contribute
to suspected clinical diagnosis of IAI [18]. In trauma or surgery patients who
have undergone operations for secondary peritonitis of community origin, the
evolution of an IAI following surgery can manifest with general worsening of
the patients clinical condition. This may be due to the appearance of intra- or
extra-abdominal complications or to the development of organ (single or
multiple) insufficiency as a consequence of early serious damage due to peri-
tonitis [19].
16.5.2 Diagnostic Imaging Techniques
Radiographic imaging is the definitive diagnostic tool for patients with suspected
IAI and can usually identify the problem prior to any planned intervention. White
abdomen usually indicates the presence of gas in the peritoneum, an intestinal
obstruction, or signs of intestinal ischemia. Studies with contrast medium using
hydrosoluble agents can reveal a fold break. Contrast fluid in a drain or fistula can
help delineate the anatomy of a complex infection and help verify the adequacy of
abscess drainage. US, and especially CT, scans play a principal role in the diagnosis
and therapeutic strategy. US has the advantage of being easily available, can be
carried out on the patient in bed, and repeated as often as necessary because of its
innocuous effect on the patient. It permits investigation of the entire abdominal
pelvic cavity, revealing even mild intraperitoneal effusion (\500 ml), detection of
the distribution of the peritoneal cavity or part of it (localized effusion), thus
assisting in hypothesizing about the nature of the effusion based on its possible
homogeneous aspects, and finally help guide fine-needle percutaneous aspiration
[20]. Abdominal CT is the reference standard for evaluating the abdomen in criti-
cally ill patients [21]. Most causes of secondary peritonitis can be diagnosed
promptly with CT using intravenously administered iodinated contrast fluid and
opacification of the digestive tube (using hydrosoluble contrast medium injected
orally or rectally) Although CT is an invaluable diagnostic instrument, it usually
involves moving a potentially unstable patient from one ward to another. Further-
more, iodinated contrast medium can worsen renal function. Renal insufficiency and
paralytic ileum are both contraindications for CT, so the risks must be taken into
account when deciding whether to carry out this diagnostic technique. One alter-
native in a very unstable patient is diagnostic peritoneal lavage, which consists of
injecting 1 l of normal saline solution or lactated ringers through a catheter into the
peritoneum and subsequently observe the characteristics of the fluid returning from
the lavage. This can evidence the presence of bacteria, leukocytes, bile, enteral
content, or blood in cases of acute intestinal ischemia [22]. Scintigraphy has a very
limited role in IAI diagnosis in critically ill patients because it is not specific enough
and cannot provide an accurate enough image to guide drainage [23, 24]. MRI has a
very high diagnostic accuracy in evaluating acute intra-abdominal abscess [22].
16.6 Management
16.6.1 Standards of Care
Source control is defined as any and all physical measures necessary to eradicate
a focus of infection as well as influence factors that maintain infection and thus
promote microbial growth or impair host antimicrobial defenses [7, 25]. Primary
peritonitis is essentially a disease that is managed with antibiotics and not sur-
gery [26]. There is no reliable evidence that cefotaxime is the treatment of
choice for spontaneous bacterial peritonitis, although many authors have sug-
gested this. Furthermore, results indicate that 4 g/day cefotaxime may be as
effective as 8 g/day in terms of reducing mortality rates and symptoms resolu-
tion, and that treatment for 10 days is no more effective than treatment for
5 days. Goals for managing secondary peritonitis are summarized in Table 16.2.
Several studies have identified E. coli and B. fragilis as the main target
organisms for antibiotic therapy [2, 27]. The current practice of early empirical
antibiotic administration, targeted to these bacteria, is well established. However,
issues concerning drug choice and timing, the need for surgical cultures, and the
duration of postoperative administration are controversial. Despite several pub-
lished options, antibiotic therapy for secondary peritonitis is simple. The
emerging concepts suggest that less in terms of number of drugs and treatment
duration is better [2]. Furthermore, recent studies suggest that monotherapy with
a single broad-spectrum antibiotic that includes full activity against E. coli may
be equal or superior to polytherapy with multiple drug combinations [2730].
The surgical strategy depends on the source of the infection [31, 32], the degree
of peritoneal contamination, the patients clinical condition, and concomitant
disease. Moreover, early goal-directed therapy provides significant benefits with
respect to outcome in patients with severe sepsis and MOF [33]. Ideally, a severe
IAI should be cured with a single surgical procedure; unfortunately, infection
often persists or recurs. Traditionally, severe peritonitis has been treated by
performing a midline laparotomy to identify and eliminate the source of infec-
tion. In certain instances, complete control of the infective focus is not feasible
during the first operation [2]. Whereas eliminating the focus and reducing con-
tamination are accepted as conditions of successful treatment, surgical proce-
dures differ for treating residual infection. The following major approaches have
been developed: (1) continuous peritoneal lavage; (2) planned relaparotomy; and
(3) open treatment by laparostomy. Continuous peritoneal lavage takes the whole
concept of lavage to an extreme, with the hypothesis being that continual IA
irrigation will enhance removal of bacteria and their products and improve the
time to resolution [34]. Various forms of peritoneal lavage are routinely used to
manage patients with peritonitis. There is little evidence that supports this
approach in either the clinical and scientific literature; moreover, it has been
documented that lavage damages mesothelial cells, dilutes agents that are
involved in peritoneal defense, and may spread previously contained infection.
Table 16.2 Principles for managing peritonitis and indications for staged abdominal repair
1. Supportive measures
Combat hypovolemia and shock and maintain adequate tissue oxygenation
Treat with antibiotics bacteria not eliminated by surgery
Support failing organ systems
Provide adequate nutrition
2. Operative treatment
Repair and/or control infection source
Evacuate bacterial inoculum, pus, and adjuvants
Treat abdominal compartment syndrome
Prevent or treat persistent and recurrent infection or verify both repair and purge
3. Staged abdominal repair
Critical patient condition due to hemodynamic instability, precluding definitive repair
Excessive peritoneal edema (abdominal compartment syndrome, pulmonary, cardiac, renal, or
hepatic dysfunction, decreased visceral perfusion) preventing abdominal closure without
tension
Intra-abdominal pressure [15 mmHg
Massive abdominal wall loss
Impossible to eliminate or to control the source of infection
Incomplete debridement of necrotic tissue
Uncertainty of viability of remaining bowel
Uncontrolled bleeding (the need for packing)
In cases of IAI, a single operation may not be sufficient to achieve source

control, thus necessitating reexploration [5]. The planned relaparotomy approach
involves reoperations at fixed intervals, irrespective of the patients clinical
condition, to prevent development of further septic fluid collections, so pre-
cluding their systemic effects. Adverse effects of planned relaparotomies are
frequent and include damage to abdominal wall structures and IA viscera. Open
management facilitates frequent reexploration and, by treating the entire peri-
toneal cavity as one large infected collection, continuous exposure for maximal
drainage. Furthermore, it serves to reduce the high IA pressure caused by per-
itoneal edema associated with fluid resuscitation and inflammation, thus obvi-
ating the deleterious systemic consequences of abdominal compartment
syndrome [35]. GIT fistulas and abdominal wall defects have plagued simple
open management; these complications should be minimized by introducing
temporary abdominal closure devices, such as artificial mesh-zipper techniques.
Tertiary peritonitis develops late in the postoperative course, presents clinically
as sepsis, and is associated with a sterile peritoneal cavity or particular micro-
biology. Further antimicrobial administration and surgical interventions may
contribute to peritoneal superinfection with yeasts and other low-level pathogens
[2]. The low virulence of these organisms, which represent a markernot the
causeof tertiary peritonitis reflects the global immunodepression of the
affected patients.
Fig. 16.1 Percutaneous drainage of a pancreatic pseudocyst
16.6.2 Managing Intra-Abdominal Abscess
Abscesses represent a relatively successful outcome of peritonitis. They may be

visceral or nonvisceral; extra- or intraperitoneal. Nonvisceral abscesses arise
after resolution of diffuse peritonitis in which a localized area of suppuration
persists or after a GIT perforation that is effectively localized by peritoneal
defenses. Percutaneous US- or CT-guided drainage is the method of choice for
single abscess. Although retrospective studies attribute no lesser mortality or
morbidity rates to percutaneous drainage versus surgical drainage [36], the for-
mer represents a minimally invasive procedure that can spare the patient the
unpleasantness of another open abdominal operation (Fig. 16.1). Moreover, a
recent study has shown that percutaneous catheter drainage should be considered
as the initial therapy for patients with culture-positive peripancreatic fluid col-
lections [37].
16.7 Acute Pancreatitis and Infection of Pancreatic Necrosis
Acute pancreatitis (AP) is an inflammatory process of variable severity

(Table 16.3) ranging from a mild self-limiting form with interstitial edema of the
pancreas to a severe form with extensive pancreatic and peripancreatic necrosis
and hemorrhage [38]. Morbidity and mortality rates associated with AP are sub-
stantially higher when necrosis is present (mortality rate 25.8% in severe AP
(SAP) and 1.5% in mild disease) [38], especially when the area of necrosis is also
infected [39].
Table 16.3 Terminology in pancreatitis
Acute interstitial A mild, self-limited form of pancreatitis, characterized by interstitial

pancreatitis edema and an acute inflammatory response without necrosis, local
complications, or systemic manifestations, such as organ failure
Necrotizing A severe form of pancreatitis characterized by locoregional tissue
pancreatitis necrosis and systemic manifestations, such as pulmonary, renal, or
cardiac failure; a diffuse or focal area of nonviable pancreas, typically
associated with peripancreatic fat necrosis
Sterile necrosis Acute pancreatitis leading to tissue necrosis without supervening
infection
Infected necrosis Acute pancreatitis with locoregional tissue necrosis complicated by
bacterial or fungal infection
Acute fluid Occurring early in the course of acute pancreatitis, located in or near the
collections pancreas, and lacking an epithelial lining or a defined wall of granulation
or fibrous tissue
Pancreatic pseudocyst Pancreatic or peripancreatic fluid collection with a well-defined wall of
granulation tissue and fibrosis; absence of epithelial lining. One common
complication of pseudocyst is the development of infection
Pancreatic abscess Circumscribed collection of pus, usually in proximity to the pancreas,
containing little or no pancreatic necrosis, arising as a consequence of
narcotizing pancreatitis or pancreatic trauma
16.7.1 Pathophysiology
AP pathogenesis remains poorly understood. A number of factors can apparently

initiate this process, including obstruction or overdistention of the pancreatic duct,
exposure to ethanol and other toxins, hypertriglyceridemia, hypercalcemia,
increased permeability of the pancreatic duct, and hyperstimulation of the gland.
These diverse factors initiate inappropriate activation of zymogens. Ischemia of
the organ appears to transform mild edematous AP into severe hemorrhagic/nec-
rotizing forms [40, 41]. The AP mechanisms and resulting complications are
reported in Fig. 16.2. The necrotic pancreas becomes secondarily infected 4060%
of the time, usually with Gram-negative bacteria translocated from the GIT [42,
43]. Enteric organisms predominate, and polymicrobial infection is common.
Fungal infection is recognized with increased frequency, and the presence of
Candida spp. is associated with increased mortality rates [44].
16.7.2 Presentation and Classification
SAP is diagnosed if three or more of Ransons criteria are present, if the acute
physiology and chronic health evaluation (APACHE) II score is C8, or if one or
more of the following are present: shock, renal insufficiency, pulmonary insuffi-
ciency. Pancreatic glandular necrosis is usually associated with necrosis of
Fig. 16.2 Mechanisms of acute pancreatitis and resulting complications. Acute pancreatitis can
be triggered by several events, resulting in inflammation of the parenchyma. Ischemia of the
organ appears to transform mild edematous pancreatitis into severe necrotizing forms of the
disease. Pseudocysts can form if pancreatic juices and debris leak into the peripancreatic spaces.
Necrotic parenchyma becomes secondarily infected 4060% of the time, usually with Gram-
negative bacteria translocated from the GIT. Alternatively, the necrotic pancreas may release
toxic factors into the peripancreatic spaces, peritoneal cavity, or systemic circulation, leading to
local or systemic complications. Computed tomography scans shown a pancreatic pseudocyst,
and lack of enhancement representing necrosis of the head, body, and tail of the pancreas
(arrows)
peripancreatic fat and, by definition, represents SAP. The risk of infected necrosis
increases with the amount of glandular necrosis and the time from AP onset,
peaking at 3 weeks. Overall mortality rate in SAP is approximately 30%; as long
as necrotizing AP remains sterile, the mortality rate is approximately 10%, and it
at least triples if there is infected necrosis. In addition, patients with sterile necrosis
and high severity-of-illness scores accompanied by MOF, shock, or renal insuf-
ficiency have significantly higher mortality rates. Deaths occur in either of two
phases: early deaths (12 weeks after AP onset) are due to MOF caused by the
release of inflammatory mediators and cytokines; late deaths result from systemic
or local infection. Local infections, so-called secondary pancreatic infections, are
represented by pancreatic abscess, infected pseudocyst, and infected necrosis. By

definition, pancreatic abscess and infected pancreatic pseudocysts are associated
with little or no necrosis; in contrast, patients with diffuse or focal areas of non-
viable pancreatic parenchyma, often associated with peripancreatic fat necrosis,
are categorized as having infected or non-infected pancreatic necrosis.
16.7.3 Recognition
Dynamic intravenous contrast-enhanced abdominal CT diagnoses pancreatic

necrosis radiographically. Because the normal pancreatic microcirculation is dis-
rupted during necrotizing AP, affected portions of the pancreas do not show
normal contrast enhancement. The lack of normal contrast enhancement may be
better detected several days after initial clinical presentation. Contrast-enhanced
CT is the gold standard for the noninvasive diagnosis of pancreatic necrosis, with
an accuracy of [90% when there is [30% glandular necrosis [45]. Sterile and
infected necrosis can be difficult to distinguish clinically, as both may produce
fever, leukocytosis, and severe abdominal pain. Distinction is important, because
mortality rates among patients with infected necrosis without intervention is nearly
100%. The bacteriologic status of the pancreas may be determined by CT-guided
fine-needle aspiration of pancreatic and peripancreatic tissue or fluid [46]. This
aspiration method is safe and accurate (sensitivity 96%, specificity 99%) and is
recommended for patients with necrotizing AP whose clinical condition deterio-
rates or fails to improve despite aggressive supportive care. US-guided aspiration
may have a lower sensitivity and specificity, but it can be performed at the bedside
[47]. Surveillance aspiration may be repeated weekly, as clinically indicated.
16.7.4 Nonsurgical Management of Pancreatic Necrosis
Early studies of antibiotics in patients with AP failed to demonstrate a significant

benefit because the studies included both patients with interstitial edematous AP
and with pancreatic necrosis. As the development of infected necrosis substantially
increases mortality rates among patients with necrotizing AP, preventing infection
is critical. Because pancreatic infection occurs primarily as a result of bacterial
spread from the colon, studies have shown benefits from both systemic antibiotics
and selective gut decontamination [4854]. Therapy should begin as soon as the
diagnosis of necrotizing AP is made and should continue for at least 24 weeks. In
this respect, intravenous administration of cefuroxime is a reasonable balance
between efficacy and cost. Furthermore, imipenem has been recommended on the
basis of studies of antibiotic penetration into pancreatic tissue. To meet increased
metabolic demands and to rest the pancreas, total parenteral nutrition (TPN) is
frequently used in patients with pancreatic necrosis [55, 56]. This does not hasten
AP resolution, however. In two recent randomized studies, patients with SAP
received either TPN or enteral nutrition (EN) within 48 h of illness onset; EN was
Table 16.4 Possible indications for endoscopic retrograde cholangiopancreatography and

associated procedures in managing pancreatitis
Acute phase ERC: diagnosis of biliary pancreatitis in severe cases with an uncertain
etiology
ES: treatment of severe cases of gallstones, biliary sludge, and other related
pathologies, such as ampullary carcinoma
Persisting ES: for bile duct stones in unresolving mild pancreatitis
pancreatitis ERP: disruption of main pancreatic duct or planning surgical intervention
Complications ERP: assessment of pseudocysts, ascites, and fistulas prior to surgery
Stent or NBC: pseudocysts, fistulas, and liquid necrosis in selected cases
ERC: endoscopic pseudocystogastrostomy, or endoscopic
pseudocystoduodenostomy in selected cases
Convalescent ERCP: establishment of diagnosis in idiopathic cases
phase ERC with bile duct cannulation: CCK-stimulated bile collection for crystal
analysis
ES: alternative to cholecystectomy in high risk cases
Recurrent ES: sphincter of Oddi dysfunction or stenosis
pancreatitis Stent: trial therapy for pancreas divisum
ERC endoscopic cholangiography; ERP endoscopic retrograde pancreatography; ES endoscopic
sphincterotomy; NBC nasobiliary catheter
well tolerated, had no adverse clinical effects, and resulted in significantly fewer
total and infectious complications [57, 58]. Acute phase-response scores and
disease-severity scores were significantly improved after EN. Studies of endo-
scopic retrograde cholangiopancreatography (ERCP) and biliary sphincterotomy
performed within 72 h of admission in patients with gallstone AP and choledo-
cholithiasis showed an improved outcome in the group of patients who presented
with clinically SAP. Improvement was attributed to the relief of pancreatic ductal
obstruction and from reduced biliary sepsis. In the presence of pancreatic ductal
disruption, a frequent occurrence in pancreatic necrosis, the introduction of
infectiontransforming sterile to infected pancreatic necrosisby incidental
pancreatography during ERCP may theoretically occur. Therefore, ERCP should
be reserved for patients in whom biliary obstruction is suspected (Table 16.4).
16.7.5 Interventions for Pancreatic Necrosis and Secondary

Pancreatic Infections
Timing and type of intervention for patients with pancreatic necrosis are contro-
versial. As the mortality rate from sterile necrosis is approximately 10% and
surgical intervention has not been shown to lower this figure, most investigators
recommend supportive medical therapy [45]. Conversely, infected necrosis is
considered uniformly fatal without intervention. Aggressive surgical pancreatic
debridement, so-called necrosectomy, remains the standard of care if drainage is
Table 16.5 Results of surgical treatment for secondary pancreatic infections
Conventional Open or semiopen Lesser sac lavage

drainage (%) procedures (%) (%)
Small-bowel 5 14 4
complications
Large-bowel 8 13 2
complications
Bleeding 14 16 8
Pancreatic fistulas 16 17 17
Persistent or recurrent 4477 8 1640
sepsis
Reoperation 3055 30
Mortality rate 4076 1620 2532
undertaken and may require multiple abdominal explorations. Necrosectomy

should be undertaken soon after confirmation of infected necrosis [38]. The benefit
of surgery in patients with MOF and sterile necrosis remains unproved, although
this scenario is frequently cited as an indication for surgical debridement. In
addition, the longer surgical intervention can be delayed after necrosis onset, the
better survival is, probably because of improved demarcation between viable and
necrotic tissue at the time of operation. The role of delayed, necrosectomy, after
MOF resolution, in patients with sterile necrosis also remains controversial.
16.7.6 Surgical Procedures After Debridement
There are three main types of surgical drainage after pancreatic debridement: con-
ventional drainage, open or semiopen procedures, and closed procedures with lesser
sac lavage [5971]. Conventional drainage involves necrosectomy with placement
of Penrose/sump drainage. Open or semiopen management involves necrosectomy
and either scheduled repeated laparotomies or open packing, which leaves the
abdominal wound exposed for frequent dressing changes. Closed management
involves necrosectomy with extensive intraoperative lavage of the pancreatic bed;
furthermore, the abdomen is closed over large-bore drains for continuous high-
volume postoperative lavage of the lesser sac. Results of these procedures are
reported in Table 16.5. The principal reason for failure of the conventional surgical
approach is the development of recurrent postoperative infection and sepsis. In an
extensive review [72] of [1,100 cases of severe pancreatic infections, 76% of
postoperative deaths were found to be attributable to persistent or recurrent infection
after surgical drainage. A closer examination of the reasons behind the demonstrated
failure of Penrose/sump drainage to provide adequate control of postoperative
infection unearths the fact that unscheduled re-explorations for recurrent sepsis are
necessary in one-third to one half of cases drained in this fashion [73, 74]. Leaving
the abdomen open eliminates the need for repeated laparotomy; packing may be
changed in ICU. Repeated debridement and manipulation of the abdominal viscera
with the open and semiopen techniques results in a high rate of postoperative local
complications (pancreatic fistulas, small- and large-bowel complications, bleeding
from pancreatic bed). Pancreatic or GIT fistulas occur in up to 40% of patients after
surgical necrosectomy and often require additional surgery for closure. The mor-
tality rate from debridement with the open or semiopen technique is approximately
15%. In patients who underwent lesser sac lavage, 1640% required reoperation for
persistent peripancreatic sepsis, and mortality rates varied from 5 to 50%, averaging
30% [72, 75].
16.7.7 Alternative Debridement Methods
Alternative methods for debriding pancreatic necrotic material have recently been
described [7678] but require considerable technical expertise, and the precise role
of these techniques in managing pancreatic necrosis will be better defined. One
study described the successful treatment of infected necrosis by aggressive irri-
gation and drainage through large-bore percutaneous catheters in 34 patients.
Pancreatic surgery was completely avoided in 47% of patients; in nine patients,
sepsis was controlled, and elective surgery was later performed to repair external
pancreatic fistulas related to catheter placement. Nine patients required immediate
surgery when percutaneous therapy failed, and four (12%) died [78]. Recently,
endoscopic drainage of sterile or infected pancreatic necrotic material has been
reported [79]. Complete necrosis resolution without the need for surgery was
achieved in 25 of 31 patients (81%). Surgical intervention was required more
commonly for acute complications of endoscopy than for drainage failure. Lap-
aroscopic approach to SAP treatment has been recently reported [77]. Advantages
are the possibility of exploring, irrigating, decompressing, and draining the pan-
creas, and performing postoperative lavage via the drainage tubes. Pathological
disease extent can be determined and appropriate treatment approaches thus
planned by laparoscopic exploration. The laparoscopic technique creates less
trauma in the early treatment of SAP but, at least theoretically, patients with early
severe multiorgan disease without retroperitoneal infection may become victims of
a later infection introduced surgically [80, 81].
16.8 Mediastinal Infections
Infection of the mediastinal space is a life-threatening condition with extremely

high mortality rates if recognized late or treated improperly. Etiologic factors
responsible for the development of mediastinitis include esophageal perforation
resulting from instrumentation, foreign bodies, penetrating or blunt trauma,
spontaneous disruption, leakage from an esophageal anastomosis, tracheobronchial
perforation, and mediastinal extension from an infectious process originating in

the pulmonary parenchyma, pleura, chest wall, vertebrae, or great vessels [8284].
Furthermore, infection may extend from the head and neck downward into the
mediastinum: this is described as descending necrotizing mediastinitis because the
infection uses the fascial planes in the neck to gain access to the mediastinum [85
87]. It is narcotizing, as the infection is often polymicrobial in etiology and
includes gas-producing organisms. The potential spaces that can allow infections
from the head or neck to enter the mediastinum include the carotid and prever-
tebral spaces. Mediastinitis occurs most often after median sternotomy for car-
diovascular surgery [8890]. Risk factors for its development include the use of
bilateral internal mammary artery grafts, diabetes mellitus, emergency surgery,
external cardiac compression, obesity, postoperative shock (especially when
multiple blood transfusions are required) prolonged bypass and operating-room
time, reexploration following initial surgery, sternal wound dehiscence, and sur-
gicaltechnical factors (such as excessive use of electrocautery, bone wax, or
paramedian sternotomy). These infections are relatively rare in the era of antibiotic
use. In developing countries, mediastinitis still is a common devastating potential
complication of head and neck infections. The mortality rate varies from 1947%,
and in the presence of comorbid conditions for patients presenting with established
infections, it may be as high as 67% [89].
16.8.1 Microbiology of Mediastinitis
This is often a mixed infection, with facultative and strict anaerobes acting
together. Streptococcus spp. are the most common facultative organisms (70% of
cases), whereas Bacteroides spp. are the most common strict anaerobes [91]. Other
organisms implicated include Pseudomonas aeruginosa and Fusobacterium,
Peptostreptococcus, and Staphylococcus spp. Mixed Gram-positive and Gram-
negative infections account for about 40% of cases. Isolated Gram-negative
infections rarely occur. Mediastinitis not associated with cardiac surgery is usually
due to anaerobic organisms. Mixed Gram-negative and Gram-positive infections
as well as Candida spp. are more common after esophageal perforation or head
and neck surgery [82, 92].
16.8.2 Diagnosis
Mediastinitis is manifested clinically by fever, tachycardia, leukocytosis, and pain

that may be localized to the chest, back, or neck. Mediastinitis presents within a
spectrum that ranges from the subacute patient through the fulminate critically ill
who requires immediate intervention in order to prevent death [89]. The typical
postoperative patient will present with fever, high pulse rate, and complaints
suggestive of a sternal WI. Approximately two-thirds of patients present these
symptoms within 14 days following surgery [90]. When mediastinitis is due to
esophageal perforation from either trauma or instrumentation, the patient usually

presents with neck pain and subcutaneous emphysema because the most common
perforation site is at the level of the cricopharyngeal muscle [92]. The diagnosis of
mediastinitis is often a clinical one. No single laboratory investigation can confirm
the diagnosis. Complete blood count shows leukocytosis, often with a left shift on
white blood cell (WBC) differential. Hematocrit decreases if bleeding has
occurred. Platelet count increases in early stages of sepsis or decreases as sepsis
worsens, or disseminated intravascular coagulation occurs. Bacteremia is found in
almost 60% of patients with postoperative mediastinitis. Mediastinal pacing wires
should be sent for culture if they are still present and no longer needed. It has been
found that pacing-wire culture had a sensitivity of 75%, a specificity of 83%, a
positive predictive value (PPV) of 12%, and a negative predictive value (NPV) of
99% [89, 90]. This suggests that sterile cultures argue against a diagnosis of
postoperative mediastinitis. Delays in the diagnosis of mediastinitis greatly
increase morbidity and mortality rates. The condition is typically recognized
through high clinical suspicion in susceptible populations, but radiology studies
are often helpful in diagnosis. CT scans are more accurate in identifying airfluid
levels and pneumomediastinum [85]; furthermore, it may demonstrate sternal
separation and substernal fluid collections in postoperative mediastinitis. The later
the scans are done following surgery, the more accurate the results. If performed
after the second postoperative week, CT scans have a sensitivity and specificity of
almost 100%. Aspiration of substernal fluid collections under CT guidance may
prove helpful in establishing a diagnosis as well as giving practical information
from Gram stain and culture [93]. MRI is poorly suited as a diagnostic modality in
mediastinitis. Postoperative patients may have sternal wires, vascular clips,
metallic valves, and pacing wires that contraindicate MRI. It is also logistically
difficult to perform an MRI in an intubated, critically ill patient. Furthermore,
esophageal dye studies are the most useful in cases of suspected esophageal
perforation.
16.8.3 Treatment
Treating mediastinitis requires correcting the inciting cause and aggressive sup-
portive therapies. Mediastinitis may result in airway compromise, and protecting
the airway is often vital. Appropriate, well-directed antibiotic therapy is crucial to
successful treatment. In postoperative mediastinitis, because up to 20% of
organisms cultured from infected sternotomy sites will be methicillin-resistant S.
aureus (MRSA), and because another 20% will be Gram-negative organisms,
antibiotic coverage must be very broad and deep to include Pseudomonas spp.
While awaiting results of stains and cultures, a regime of vancomycin, ceftazi-
dime, and a quinolone should provide adequate coverage. If septic shock is
present, some would substitute an aminoglycoside for the quinolone, although
close watch of drug levels is required to prevent renal damage. Therapy duration is
usually prolonged, ranging from weeks to months. Recently, it has suggested that
46 weeks of therapy is adequate for most patients. EN should be introduced

immediately with a duodenal feeding tube. Recent data suggest that the use of diets
formulated with various anti-inflammatory compounds to include omega-3 long-
chain fatty acids and arginine provide clinically important benefits for critically ill
patients with sepsis [94]. Mediastinitis due to esophageal perforation should be
initially treated with broad-spectrum antibiotics that have good anaerobic cover-
age. Piperacillintazobactam and vancomycin provide good empirical coverage
while awaiting culture results. Patients with true anaphylaxis to penicillin can be
treated with quinolone and clindamycin in place of the piperacillintazobactam.
The role of antifungal therapy is controversial, although amphotericin B has been
used. In patients with mediastinal infections in continuity with empyema, sub-
phrenic abscess or neck abscess, drainage with tube thoracostomy, or percutaneous
drainage is frequently successful. In patients who do not respond to the initial
measures or in whom mediastinitis occurs from most other causes, thorough
debridement of necrotic and infected tissue is necessary in conjunction with sur-
gical drainage [89]. Controversy exists about whether the cervical or transthoracic
approach is best. In some case series, the combination of the two has been asso-
ciated with a lower mortality rate. Surgical treatment is required in severe infec-
tions. A simple sternal dehiscence without infection can be effectively treated by
rewiring the sternum; this usually yields excellent long-term results. More
advanced mediastinitis can be treated with extensive debridement followed by
irrigation. Many surgeons prefer to leave the wound open for subsequent
debridement efforts. In this case, the wound is packed, and closure via a plastics
procedure is delayed until wound cultures have been consistently negative. Ven-
tilator weaning and risk of bleeding from the exposed heart and vessels are dis-
advantages of the open approach that change thoracic mechanics. However, the
most common cause of recurrent mediastinitis is inadequate wound-site debride-
ment and sterilization during the first procedure. Advanced postoperative medi-
astinitis is best treated with extensive debridement, followed by delayed closure
using either muscle or omental flap plastics procedures [95]. Oesophageal perfo-
ration was usually fatal; the standard approach has been to operate immediately.
In recent years, several case series have addressed medical treatment of esophageal
perforation. Shaffer et al. [92] recommend considering nonoperative treatment in
absence of crepitus, pneumothorax, pneumoperitoneum, or intraperitoneal
extravasation, in esophageal disruptions that are well contained within the medi-
astinum or a pleural loculus; in instrumental perforations in which the patient was
nil per os for the procedure and the perforation was detected early; in patients who
are clinically stable; and for perforations with a long delay before diagnosis, such
that the patient develops tolerance for the perforation without the need for surgery
[92]. Surgery is recommended for Boerhaave syndrome, large perforations with
widespread mediastinal contamination, perforations associated with preexisting
conditions such as achalasia and cancer, perforation of the IA portion of the
esophagus, perforations with pneumothorax, perforations with retained foreign
bodies, and unstable patients with shock or other signs of systemic sepsis.
16.9 Pleural Effusion and Empyema
Pleural effusion (PE) is defined as an abnormal accumulation of fluid in the pleural

space: approximately 300 ml of fluid is required for the development of costo-
phrenic angle blunting seen on upright chest X-ray; at least 500 ml of effusion is
necessary for detection on clinical examination. It is an indicator of a pathologic
process that may be of primary pulmonary origin or an origin related to another
organ system or to systemic disease. It may occur in the setting of acute or chronic
disease (Table 16.6) and is not a diagnosis in and of itself [9699]. PEs is clas-
sified as either transudate or exudate based on fluid protein and lactate dehydro-
genase (LDH) concentrations. Normal pleural fluid has the following
characteristics: pH 7.607.64, protein content \2% (12 g/dl), fewer than 1,000
WBCs per cubic millimeter, glucose content similar to that of plasma, and LDH
level \50% of plasma. Pleural fluid is exudative if one or more of the following
conditions is met and transudative if none are met: ratio of pleural fluid and serum
protein levels [0.5, ratio of pleural fluid and serum LDH levels [0.6, and pleural
fluid LDH level more than two-thirds of the upper limit for serum LDH levels.
Empyema is a pyogenic or suppurative infection of the pleural space. It is exu-
dative in nature, and represents the most common exudative type of PEs [100,
101]. Empyema may be classified into three categories based on the chronicity of
the disease process. The acute phase is characterized by PE of low viscosity and
cell count. The transitional or fibrinopurulent phase, which can begin after 48 h, is
characterized by an increase in WBCs in PE: this begins to loculate and is asso-
ciated with fibrin deposition on visceral and parietal pleura and progressive lung
entrapment. The chronic phase occurs after as little as 12 weeks and is associated
with an ingrowth of capillaries and fibroblasts into the pleural rind and inexpan-
sible lung. Empyema may occur by direct contamination of the pleural space
through wounds of the chest, by hematologic spread, by direct extension from lung
parenchymal infections, by rupture of an intrapulmonary abscess or infected
cavity, or by extension from the mediastinum. Most often, empyemas are the
results of a primary infectious process in the lung.
16.9.1 Frequency and Mortality/Morbidity
In industrialized countries, relative annual incidence of PE is estimated to be 320

per 100,000 people. Approximate PE annual incidences are based on major
underlying disease processes, such as congestive heart failure, bacterial pneumo-
nia, malignancy, pulmonary embolus, cirrhosis with ascites, pancreatitis, collagen
vascular disease, and tuberculosis. PE morbidity and mortality rates are directly
related to cause, disease stage at the time of presentation, and biochemical findings
in the pleural fluid. Morbidity and mortality rates of patients with pneumonia and
PE are higher than those of patients with pneumonia alone.
Table 16.6 Etiology of pleural effusion (PE)
PE characteristic Diseases
Transudative Congestive heart failure (most common transudative PE); hepatic cirrhosis
with and without ascites; nephrotic syndrome; peritoneal dialysis/continuous
ambulatory peritoneal dialysis; hypoalbuminemia (e.g., severe starvation);
glomerulonephritis; superior vena cava obstruction; urinothorax
Exudative Malignant disorders (metastatic disease to the pleura or lungs, primary lung
cancer, mesothelioma, Kaposi sarcoma, lymphoma, leukemia); infectious
diseases (bacterial, fungal, parasitic, and viral infections; infection with
atypical organisms such as Mycoplasma, Rickettsiae, Chlamydia,
Legionella); GIT diseases (pancreatic disease, Whipple disease, IA abscess,
esophageal perforation, abdominal surgery, diaphragmatic hernia, endoscopic
variceal sclerotherapy); collagen vascular diseases (rheumatoid arthritis,
systemic lupus erythematosus, drug-induced lupus syndrome, immunoblastic
lymphadenopathy, Sjgren syndrome, familial Mediterranean fever, Churg
Strauss syndrome, Wegener granulomatosis); benign asbestos effusion;
Meigs syndrome (benign solid ovarian neoplasm associated with ascites and
pleural effusion); drug-induced primary pleural disease (nitrofurantoin,
dantrolene, methysergide, bromocriptine, amiodarone, procarbazine,
methotrexate, ergonovine, ergotamine, oxprenolol, maleate, practolol,
minoxidil, bleomycin, interleukin-2, propylthiouracil, isotretinoin,
metronidazole, mitomycin); injury after cardiac surgery (Dressler syndrome);
uremic pleuritis; yellow nail syndrome; ruptured ectopic pregnancy;
electrical burns
Exudative/ Pulmonary embolism; hypothyroidism; pericardial disease (inflammatory or
transudative constrictive); atelectasis; trapped lung (usually a borderline exudate);
sarcoidosis (usually an exudate); amyloidosis
Miscellaneous Hemothorax; following coronary artery bypass graft surgery; after lung or
liver transplant; Milk of calcium pleural effusion; ARDS; systemic
cholesterol emboli; iatrogenic misplacement of lines or tubes into the
mediastinum or the pleural space; radiation pleuritis; necrotizing sarcoid
granulomatosis; ovarian hyperstimulation syndrome; postpartum pleural
effusion (immediate or delayed); rupture of a silicone bag mammary
prosthesis; rupture of a benign germ cell tumor into the pleural space (e.g.,
benign mediastinal teratoma; syphilis; echinococcosis
GIT gastrointestinal tract; IA intra-abdominal; ARDS acute respiratory distress syndrome
16.9.2 Symptoms and Physical Findings
Clinical manifestations of PE are variable and often related to the underlying

disease process. The most commonly associated symptoms are progressive
dyspnea, cough (typically nonproductive), and pleuritic chest pain. Dyspnea is the
most common clinical symptom at presentation and usually indicates a large
effusion [96, 98]. Chest pain may be mild or severe; it is typically described as
sharp or stabbing, is exacerbated with deep inspiration, and is pleuritic. Pain may
be localized to the chest wall or referred to the ipsilateral shoulder or upper
abdomen, usually because of diaphragmatic involvement. Other signs and symp-

toms occurring with PEs are associated more closely with the underlying disease
process. An acute febrile episode, purulent sputum production, and pleuritic chest
pain may occur in patients with PE associated with aerobic bacterial pneumonia
[99]. Physical findings are variable and depend on the PE volume. Generally,
findings are undetectable for effusions \300 ml. Four main types of fluids in the
pleural space are serous fluid (hydrothorax), blood (hemothorax), lipid (chylo-
thorax), and pus (pyothorax or empyema). PE develops in 3040% of patients with
bacterial pneumonia. Those with bacterial pneumonia, especially that caused by S.
pneumoniae, have a high predilection for complications: these can include bac-
teremia and multilobar involvement. Moreover, Gram-negative and anaerobic
organisms are common cause of empyema [100].
16.9.3 Diagnosis
The initial step in analyzing pleural fluid is to determine whether PE is a transudate

or an exudate. Clinical presentation should direct the biochemical and micro-
biological studies of pleural fluid. The minimal amount of pleural fluid needed for
basic diagnostic purposes is 20 ml; if possible, 60 ml should be obtained for
potential diagnostic studies. If the clinical presentation is highly suggestive of
transudative PE, protein and LDH levels should be determined initially. If the
patient has undergone diuretic therapy, the pleural albumin level should be
determined simultaneously. Concomitant serum total protein, LDH, and, if indi-
cated, serum albumin levels should be measured. If transudative effusion is
diagnosed, no further tests are needed. Exudative PE requires further laboratory
investigation. Cytological analysis is strongly recommended for patients with a
history of undiagnosed exudative PE, suspected malignancy or Pneumocystis
carinii infection, or exudative PE with normal fluid glucose and amylase levels
[100, 101]. Additional studies should be requested on the basis of the gross
appearance of the pleural fluid or when a specific condition is suspected. Gross
appearance and results of certain laboratory studies may provide useful diagnostic
information (Table 16.7). Laboratory results can aid in narrowing the differential
diagnosis of exudative PE (Table 16.8).
16.9.4 Imaging Studies
This step is the most important in the PE evaluation. Common imaging studies
used to confirm PE are chest radiography, US, and CT scan [96101]. Chest
radiography is the primary diagnostic tool because of its availability, accuracy,
and low cost. It can be used to determine the cause of PE (enlarged cardiac
silhouette, underlying lung, parenchymal disease). The most common radiologic
appearance is blunting of the costophrenic angle and/or sulci. Upright postero-
anterior or anteroposterior radiographs may not show lateral costophrenic angle
Table 16.7 Clinical significance of pleural fluid characteristics
Characteristic Significance
Bloody Most likely an indication of malignancy in the absence of trauma;
can also indicate pulmonary embolism, infection, pancreatitis,
tuberculosis, mesothelioma, or spontaneous pneumothorax
Turbid Possible increased cellular or lipid content
Yellow or whitish, turbid Presence of chyle, cholesterol, or empyema
Brown Rupture of amoebic liver abscess into the pleural space (amebiasis
with a hepatopleural fistula)
Black Aspergillus involvement of pleura
Yellow green with debris Rheumatoid pleurisy
Highly viscous Malignant mesothelioma (due to increased levels of hyaluronic
acid), long-standing pyothorax
Putrid odor Anaerobic infection of pleural space
Ammonia odor Urinothorax
Purulent Empyema
Yellow and thick, with Effusions rich in cholesterol (longstanding chyliform effusion, e.g.,
metallic sheen tuberculous or rheumatoid pleuritis)
blunting until 250500 ml of fluid is present. Lateral radiographs show blunting

of the posterior costophrenic angle and the posterior gutter when as little as
175200 ml of fluid is present. Bilateral decubitus radiographs are recommended,
especially with larger effusions. PE location can help in differential diagnosis.
Atypical chest radiographic presentations are possible. When an airfluid level is
present in the pleural space, the following must be considered: bronchopleural
fistula, pneumothorax, trauma, presence of gas-forming organisms, diaphrag-
matic hernia, fluid-filled bullae or lung cysts and rupture of the esophagus into
the pleural space. Diaphragmatic hernias can be excluded or confirmed with GIT
contrast material administration. US can be used to detect as little as 550 ml of
pleural fluid, with 100% sensitivity for effusions C100 ml [100]. US helps
identify loculated PE and differentiation of pleural fluid from pleural fibrosis,
thickening, and parenchymal consolidation. It can help localize the diaphragm if
pleural or parenchymal disease obscures it. Unlike CT, US is rapid and available
at the bedside. Chest CT scanning permits simultaneous imaging of the entire
pleural space, pulmonary parenchyma, and mediastinum. CT scans reveal early-
stage pleural abnormalities. Contrast-enhanced scans can depict multiple locu-
lations and localizing PEs; differentiate between lung consolidation versus PE
and cystic versus solid lesions; necrotic areas; pleural thickening, nodules,
masses, or rounded atelectasis; and peripheral lung abscess versus loculated
empyema; and tumoral extent [99, 100].
Table 16.8 Clinical significance of biochemical and cytological pleural fluid characteristics
Amylase Can be elevated in acute pancreatitis, pancreatic pseudocyst,
esophageal rupture, malignancy, and ruptured ectopic pregnancy.
Pancreatic pseudocyst has the highest amylase levels
(frequently [100,000 UI). Determination of the amylase isoenzyme
level is useful in distinguishing PE caused by pancreatic disease from
effusions caused by esophageal rupture or nonpancreatic disease
Glucose A low PE glucose level is \60 mg/dl. Differential diagnosis includes
malignancy, rheumatoid pleurisy, complicated parapneumonic PE,
empyema, hemothorax, ChurgStrauss syndrome, and occasionally
lupus pleuritis
pH If \7.20, suggests empyema, complicated parapneumonic PE,
esophageal rupture, rheumatoid pleuritis, malignancy, hemothorax,
tuberculous pleuritis, lupus pleuritis, or urinothorax. Arterial pH
influences pleural fluid pH; therefore, acidemia must be ruled out
before any of the above causes are considered. With parapneumonic
PE, indications for tube thoracostomy include a pH \7.0, glucose
level \40 mg/dl, and positive finding with Gram stains or cultures.
To use the pH criteria for chest tube placement in systemic acidosis,
the PE pH should be at least 0.3 U less than the arterial pH. A low
pleural fluid pH almost always is associated with a high pCO2
LDH Indicator of the degree of pleural inflammation. The higher the value,
the more inflamed the pleural surface. High concentrations
([1,000 IU/l) occur with complicated parapneumonic PE
RBC and total WBC RBC counts [100,000 mm3 suggest trauma, malignancy, pulmonary
count embolism, injury after cardiac surgery, asbestos pleurisy, esophageal
rupture, pancreatitis, tuberculous pleurisy, and thoracic
endometriosis. The total PE leukocyte count is virtually never
diagnostic. Neutrophilic predominance indicates an acute
inflammatory process near the time of thoracentesis. Significant
eosinophilia occurs when the ratio of pleural fluid and total pleural
fluid counts is [10%; the most common cause is air or blood in the
pleural space. The differential diagnosis of pleural fluid eosinophilia
includes pneumothorax, hemothorax, pulmonary infarction, prior
thoracentesis, benign asbestos effusion, drug use, parasitic diseases,
fungal infections, and ChurgStrauss syndrome; in the absence of
these, eosinophilia with pneumonia and pleural effusion is a good
prognostic sign, because such effusions rarely become infected.
Significant basophilia (counts [10%) is distinctly uncommon;
however, if present, it suggests leukemic pleural infiltration
Lymphocyte count Lymphocytes indicate a long-standing chronic effusion. Pleural fluid
lymphocytosis ([50%) suggests malignant disease, particularly
lymphoma; however, other conditions (e.g., chronic rheumatoid
pleurisy, chronic fungal infection, yellow nail syndrome, chylothorax,
trapped lung, benign asbestos PE, sarcoidosis) must be considered.
The presence of an undiagnosed exudative effusion with
lymphocytosis is an indication for closed pleural biopsy
(continued)

Mesothelial cell count These cells line the pleural cavities, and their absence simply
indicates diffuse pleural injury or fibrosis. These cells predominate in
transudative PE
Plasma cell and A large number of plasma cells in the pleural fluid suggest multiple
macrophage counts myeloma with pleural involvement. A small number of plasma cells
are of no particular diagnostic value. The presence of macrophages
has no diagnostic value.
PE plural effusion; LDH lactose dehydrogenase; RBC red blood cells; WBC white blood cells
16.9.5 Invasive Diagnostic Procedures
After the presence of PE is established, the cause should be identified. This step
can be critical because unnecessary invasive procedures cause morbidity and
mortality. When a decision is made to investigate the cause, thoracentesis is the
first-line invasive diagnostic procedure [100]. Thoracentesis also can be used as a
therapeutic modality. It is the least invasive procedure and it is relatively safe. For
stable and asymptomatic patients in whom PE most likely is caused by viral
pleurisy, manifestation of a systemic disease, thoracic or abdominal surgery, or
childbearing, thoracentesis may not be indicated, or it can be deferred. In this
situation, therapy for the specific cause should be initiated, and if no improvement
occurs after a few days, diagnostic thoracentesis should be performed. Thora-
centesis is also indicated in cases in which the specific cause of PE is unknown or
has never been investigated or when the thickness of the free pleural fluid level
is [10 mm on the lateral decubitus radiograph. In addition, thoracentesis is
indicated if the patient has respiratory compromise, hemodynamic instability, or
massive effusion with contralateral mediastinal shift. After thoracentesis, and
regardless of its success, chest radiography is recommended to rule out a sub-
sequent pneumothorax. Pneumothorax is the most common complication (inci-
dence 320% with unguided thoracentesis, 27% with US guidance) and is
operator dependent. Other complications include subcutaneous hematoma, infec-
tion of the pleural space or soft tissue overlying the thoracentesis site, pain at the
site, cough, chest pain, hemothorax, vasovagal reflex, reexpansion pulmonary
edema, hypovolemia, hypoxemia, splenic or hepatic laceration, hemoperitoneum,
and adverse reactions to local anaesthetics. Definite indications to tube thoracos-
tomy include empyema, hemothorax, large pneumothorax, and parapneumonic PE.
16.9.6 Treatment
Most commonly, PE is an incidental finding in a stable patient. Patients with a toxic

condition, respiratory distress, or cardiovascular instability require emergency
medical services more frequently. As with any other life-threatening condition,
direct initial management is airway stabilization to ensure adequate oxygenation and

ventilation [98]. On the basis of presentation, patients with PE may be stable,
requiring hospital admission; stable, not requiring hospital admission; or unstable.
Stable patients who do not require admission include those in whom the clinical
circumstances clearly explain PE and/or prior investigations of the cause were
performed, in whom PEs are typical of the disease or are asymptomatic, and in
whom diagnostic or therapeutic thoracentesis is not required. Stable patients
requiring admission include most patients with PE thicker than 10 mm on the lateral
decubitus radiograph. Such patients include those with no prior history of PE, those
with parapneumonic PE who do not appear to have a toxic condition, and those with
a prior history of PE who have a change in their usual symptoms or effusion.
Although these patients are not in acute respiratory distress, diagnostic thoracentesis
is imperative. For parapneumonic PE, delay in diagnostic thoracentesis and anti-
biotic therapy can be detrimental. Simple parapneumonic PE have a great potential
to become complicated or empyemas. Antimicrobial therapy alone is not sufficient
for complicated parapneumonic PE or empyemas; they require tube thoracostomy
and antibiotics [98]. Unstable patients include those with a toxic appearance,
respiratory distress, or cardiovascular compromise due to PE. The initial treatment
focus should be to stabilize the airway and circulation. Infected pleural fluid with
bronchopleural fistula is considered a medical emergency. Bronchopleural fistula
should be suspected when a patient with PE produces a larger amount of sputum than
that expected from associated pulmonary disease. The presence of an airfluid level
in the pleural space on upright radiographs suggests bronchopleural fistula. These
patients require immediate diagnostic thoracentesis and antibiotics [98, 99]. In any
patient with chest-penetrating or nonpenetrating trauma, hemothorax should be
suspected. Traumatic hemothorax is an indication for the insertion of a large-bore
chest tube. Antibiotics and diuretics are commonly used in the initial management of
PE. An empiric systemic antibiotic coverage should be initiated in infections or
potentially septic conditions (parapneumonic PE, empyemas, esophageal perfora-
tion, hemothorax, IA abscesses). Generally, broad-spectrum antibiotics should be
used initially for both aerobic and anaerobic microorganisms. Most commonly, two
antimicrobial agents are necessary to ensure adequate coverage [9799]. Various
effective combinations exist: one is a third-generation cephalosporin (cefotaxime or
ceftriaxone) and clindamycin. If the patient is a nursing home resident, cephalo-
sporin with enhanced antipseudomonas activity is recommended. In children,
monotherapy usually is sufficient.
16.10 Wounds Infections
Wound infection (WI) is the most important complication of surgical procedures and
continues to be a disconcerting source of mortality in surgical patients [10, 102
104]. Postoperative infections, also, prolong hospitalization and are important
causes of postoperative morbidity. Despite antisepsis, surgical site infections (SSI)
are the third most frequently reported nosocomial infection, accounting for 1416%
Table 16.9 Classification of operative wounds
Clean wounds Infection risk of about 15%. Prophylactic antibiotics are not indicated in
a clean operation if the patient has no host-risk factors. Factors suggesting
the need for prophylaxis are remote infection, diabetes, at least three
concomitant medical diagnoses. Additive risk factors are abdominal
operations and operations expected to last [2 h. Prostheses implants are
clean procedures, some of which requires antibiotic prophylaxis. Inguinal
hernia repair with biomaterials does not benefit from antibiotic
prophylaxis
Clean-contaminated B10% risk of infection. Clean-contaminated surgery usually requires
wounds prophylaxis
Contaminated A 1020% risk of infection. This type of surgery needs prophylaxis.
wounds Biliary, hepatobiliary, and pancreatic operations usually meet criteria of
clean-contaminated wounds definitions. In biliary-tract procedures,
prophylaxis is required only for cases at high risk of contamination: bile
obstruction, jaundice, stones in common duct, reoperation, and
cholecystitis. Prophylaxis is always required in hepatobiliary and
pancreatic surgery because these operations are long. In gastroduodenal
operations, the risk is low if gastric acidity is normal and bleeding, cancer,
gastric ulcer and obstruction are absent. Colorectal procedures are usually
contaminated cases. The unique goal of prophylaxis includes preoperative
reduction of bacterial concentration in feces. In association with
mechanical bowel preparation, prophylaxis is used in major elective
abdominal procedure (i.e., vascular graft) to prevent bacterial
translocation from the gut
Infected wounds Risk of infection of about 2050% meeting the following criteria: acute
bacterial inflammation, without pus; transection of clean tissue for the
purpose of surgical access to a collection of pus; traumatic wound with
retained ischemic tissues, foreign bodies, fecal contamination, or delayed
treatment. Appropriate treatment is to administer antibiotic therapy, not as
prophylaxis, because infection is already present
of all nosocomial infections among hospitalized patients [103]. Among surgical

patients, they are the most common nosocomial infection, accounting for 38% of all
such infections. Of these, two-thirds were confined to the incision, and one-third
involved organs or spaces accessed during the operation. When surgical patients
with nosocomial SSI die, 77% of deaths are reported to be related to the infection;
the majority (93%) were serious infections involving organs or spaces accessed
during the operation [103, 104]. A simple method of evaluating the probability that
WI will develop consists of classifying the wound according to the scheme illus-
trated in Table 16.9.
16.10.1 Risk Factors
Surgical WI occurs whenever the combination of microbial numbers and vir-

ulence is sufficiently large to overcome the local host defence mechanisms and
establish progressive growth [102109]. It is immediately evident that different

types of surgical procedures, involving a greater or lesser degree of contami-
nation, are then associated with a different probability of developing WI.
Several patient factors, besides the type of procedure and operator skill, are
important in affecting the probability of incurring a postoperative WI. One is
patient age: the higher the age, the higher the incidence. Another factor is the
period of hospital stay before surgery: patients hospitalized for [12 days are
more susceptible to WI, showing that a relationship is likely to exist between
the incidence of infection and the hospital environment and bacterial flora
[106, 107].
16.10.2 Factors Involved in WI Development
It is clear that the mechanism whereby a patient develops WI is linked to three

critical elements: the closed space; the infectious agent, which must be present in
sufficient number and with sufficient virulence; and the susceptible host. Microbial
contamination of the surgical site is a necessary precursor of SSI. Quantitatively, it
has been shown that if a surgical site is contaminated with [105 microorganisms
per gram of tissue, SSI risk is markedly increased. Injury produces enclosed
environments due to pockets of extravasated blood, necrotic tissue, infarcted areas,
foreign bodies, and prostheses. The environment in these enclosed spaces soon
becomes hypoxic, hypercarbic, and acidic, favoring bacterial growth. In abdominal
surgery, GIT represents a huge reservoir of pathogenetic bacteria, and recently it
was hypothesized that it could act as an undrained abscess, causing infection and
MOF [102, 104, 108]. Any infectious agent can contaminate a closed space, but
relatively few cause infection. Streptococcus spp. invade even minor breaks in the
skin and spread through connective tissue planes and lymphatics. Staphylococcus
spp. are less invasive but more pathogenic. Pseudomonas and Serratia spp. are
often seen as opportunistic invaders. Many fungi and parasites may cause
abscesses or sinus and are typical of the immunocompromised patient. Anaerobes,
such as Bacteroides spp., and Peptostreptococcus spp. are more frequently isolated
because of improvements in culture techniques. Postsurgical WI is a multimi-
crobial disease in which many bacteria act in synergism [108]. Finally, there are
host-dependent predisposing factors to the development of postsurgical infections.
Among these are diabetes, severe trauma, burns, malnutrition, cancer, hemato-
logical disorders, transplantation, and immunosuppressive drugs. In many patients,
these factors are believed to be primarily responsible for decreased reactivity to
delayed hypersensitivity antigens, creating an anergic state associated with an
increased incidence of infectious complications. Considering these three elements,
it is easier to understand the cycle: immunosuppression and anergy are highly
common in a surgical patient; they lead to infection, and the infection itself can
deteriorate the immune system.
16.10.3 Treatment
As infectious complications in surgical patients are responsible for prolonged

wound healing, disability, and even death; and as the patients quality of life can
be affected or even permanently altered by themwith huge socioeconomic
costsit is important to prevent them as far as possible. The importance of the
nutritional status is emphasized by the higher incidence of infectious and other
surgical complications in malnourished patients. Among several nutritional status
indices, the prognostic nutritional index (PNI), which explicitly includes serum
albumin value, correlates with the probability of infection after surgery: to values
of PNI [50%, there corresponds a high risk of postoperative complications; to
values between 40 and 49%, an intermediate risk; to values \40% a low risk.
Furthermore, perioperative administration of supplemental oxygen is a practical
method of reducing the incidence of surgical WIs [110].
16.10.4 Antibiotic Prophylaxis
The first important point about antibiotic prophylaxis is its timing of administra-
tion [104, 106]. Administration of antibiotics just before, during, and up to 3 h
after surgery effectively prevents WIs. Many studies demonstrate that prophylactic
antibiotics are most useful if given to patients before contamination occurs: the
most relevant protective effect was observed when antibiotic was given so that
good tissue levels were present at the time of the procedure and for the first 34 h
after surgical incision. A practical approach would then be to contemplate
administering a single preoperative dose, followed by an intraoperative dose if the
procedure lasts [3 h or twice the half-life of the antibiotic, and massive hemor-
rhage occurs during surgery [102]. Principles of the proper prophylaxis of WIs
include selecting bactericidal antibiotics effective against likely pathogens. Single-
agent prophylaxis is almost always effective in the majority of clinical situations,
provided that the half-life of the antibiotic is long enough to maintain adequate
tissue levels throughout the operation and that the given dose is equal to a full
therapeutic i.v. dose. Recommended antibiotics for prophylaxis of WI caused by
Gram-positive and Gram-negative aerobic bacteria are cefazolin (1 gm i.v./i.m.) or
vancomycin (1 g i.v.) in patients allergic to cephalosporins. First-generation
cephalosporins, such as cefazolin, are good choices because they are not expen-
sive, incur a low rate of allergic responses, and have a broad-spectrum of activity
against likely aerobic pathogens. Prophylaxis against both Gram-negative aerobes
and anaerobes includes clindamycin or metronidazole plus tobramycin, or a single
broad-spectrum agent such as cefoxitin or cefotetan, or sulbactam/ampicillin.
Gram-negative anaerobes (Bacteroides spp.) are of GIT origin and are synergistic
with Gram-negative aerobes in causing infections after GIT surgical procedures.
Even though the problem is still debated, combination of two antibiotics is in
general more powerful than a single broad-spectrum agent active against both
bacterial components [102, 106]. A second point concerns the cases for which
antibiotic prophylaxis as described is indicated. In general, an approach such as

that outlined above is indicated for GIT and anorectal surgery, biliary-tract sur-
gery, vaginal hysterectomy, insertion of artificial devices, or prolonged ([3 h)
clean surgery. In contaminated or dirty surgery, appropriate therapy should be
started as soon as possible: the right treatment is to administer antibiotic therapy,
not as prophylaxis, because infection is already present [102]. In relation to
operative contamination and increasing risk of WI, classifying the operative
wound includes four categories mentioned above. This is the most widely applied
classification of surgical procedures in terms of contamination and probability that
a WI will develop; this risk is about 5, 10, 15, and 30% for the four reported
classes, respectively [102].
16.11 Urinary Tract Infections
Urinary tract infection (UTI) is defined as significant bacteriuria in the presence of

symptoms [111, 112]. This common clinical entity affects an estimated 20% of
women at some time during their lifetimes. Successful management includes
proper specimen collection, use of immediately available laboratory testing for
presumptive diagnosis, appreciation of epidemiological and host factors that may
identify patients with clinically unapparent upper UTI, and selection of appropriate
antimicrobial therapy with recommendations for follow-up care [113115].
16.11.1 Pathophysiology
The urinary tract is normally sterile. Uncomplicated UTI involves the urinary
bladder in a host without underlying renal or neurological disease. The clinical
entity is termed cystitis and represents bladder mucosal invasion, most often by
enteric coliform bacteria (E. coli), which inhabit the periurethral vaginal introitus
and ascend into the bladder via the urethra. Sexual intercourse may promote this
migration, and cystitis is common in otherwise healthy young women [114]. Urine
is generally a good culture medium; factors unfavorable to bacterial growth
include a low pH (B5.5), a high concentration of urea, and the presence of organic
acids derived from a diet that includes fruits and protein. Frequent and complete
voiding has been associated with a reduced incidence of UTI [115119]. Nor-
mally, a thin film of urine remains in the bladder after emptying, and any bacteria
present are removed by the mucosal cell production of organic acids. If lower
urinary tract mechanisms fail, upper urinary tract or kidney involvement occurs,
which is termed pyelonephritis. Host defences at this level include local leukocyte
phagocytosis and renal production of antibodies that kill bacteria in the presence of
complement. Complicated UTI occurs in the setting of underlying structural,
medical, or neurological disease [115]. Patients with a neurogenic bladder or
bladder diverticulum, and postmenopausal women with bladder or uterine pro-
lapse, have an increased frequency of UTI due to incomplete bladder emptying.
This eventually allows residual bacteria to overwhelm local bladder mucosal

defences. High urine glucose content and defective host immune factors in patients
with diabetes mellitus also predispose to infection.
16.11.2 Frequency and Morbidity/Mortality Rates
An estimated 11% of women in the USA report at least one physician-diagnosed

UTI per year, and the lifetime probability that a woman will have a UTI is 60%
[120]. Although simple cystitis may resolve spontaneously, effective treatment
lessens symptom duration and reduces the incidence of progression to upper UTI.
Pyelonephritis is associated with substantial morbidity rate [117119], including
systemic effects such as fever, vomiting, dehydration, and loss of vasomotor tone
resulting in hypotension. Complications include acute papillary necrosis with
possible development of ureteral obstruction, septic shock, and perinephric
abscess. Chronic pyelonephritis may lead to renal scarring with diminished
function. Younger patients have the lowest rates of morbidity and mortality.
Despite appropriate intervention, 13% of patients with acute pyelonephritis die.
Factors associated with unfavorable prognosis are general debility and old age,
renal calculi or obstruction, recent hospitalization or instrumentation, diabetes
mellitus, sickle cell anemia, underlying carcinoma, intercurrent chemotherapy, or
chronic nephropathy. The largest group of patients with UTI is adult women. The
incidence increases with age and sexual activity. Rates of infection are high in
postmenopausal women because of bladder or uterine prolapse, causing incom-
plete bladder emptying; loss of estrogens, with attendant changes in vaginal flora;
loss of lactobacilli, which allows periurethral colonization with Gram-negative
aerobes such as E. coli; and higher likelihood of concomitant medical illness, such
as diabetes. UTI is unusual in men \50 years, and symptoms of dysuria and
frequency are usually due to urethral or prostatic infection. In older men, however,
the incidence of UTI rises because of prostatic obstruction or subsequent
instrumentation.
16.11.3 Symptoms and Physical Findings
Classic symptoms of UTI in the adult are dysuria with accompanying urinary
urgency and frequency. A sensation of bladder fullness or lower abdominal dis-
comfort is often present. Bloody urine (hemorrhagic cystitis) is reported in as
many as 10% of cases of UTI in otherwise healthy women. Fevers, chills, and
malaise may be noted, though these are associated more frequently with pyelo-
nephritis. Most adult women with simple lower UTI have suprapubic tenderness
with no evidence of vaginitis, cervicitis, or pelvic tenderness. The patient with
pyelonephritis usually appears ill and, in addition to fever, sweating, and pros-
tration, is found to have flank tenderness in the majority of cases.
16.11.4 Diagnosis
If UTI is suspected, the initial test of choice is urinalysis. The midstream-voided

technique is as accurate as catheterization if the proper technique is followed.
Pyuria, as indicated by a positive result of the leukocyte esterase dip test, is found
in the vast majority of patients with UTI. However, low-level pyuria [620 WBCs
per high power field (HPF) microscopy on a centrifuged specimen] may be
associated with an unacceptable level of false-negative results with the leukocyte
esterase dip test. In the female patient with appropriate symptoms and examination
findings suggestive of UTI, urine microscopy may be indicated despite a negative
leukocyte esterase dip test result. Current emphasis in UTI diagnosis rests with
pyuria detection. As noted, a positive leukocyte esterase dip test suffices in most
instances. According to Stamm and Hooton [119], pyuria levels as low as 25
WBCs per HPF in a centrifuged specimen are important in the female patient with
appropriate symptoms. The presence of bacteriuria is as significant. A positive
result on the nitrate test is highly specific for UTI, typically because of urease-
splitting organisms, such as Proteus spp. and, occasionally, E. coli; however, it is
highly insensitive as a screening tool, as only 25% of patients with UTI have a
positive nitrite test result. Low-level or, occasionally, frank hematuria may be
noted in otherwise typical UTI; its PPV is poor, however. Historically, the defi-
nition of UTI was based on the finding of a single organism at culture of 100,000
colonies per millimeter. If the patient has had a UTI within the last month, the
same organism probably causes relapse, which represents treatment failure [111].
Reinfection occurs in 16 months and is usually due to a different organism (or
serotype of the same organism). In the vast majority of patients with UTI, no
imaging studies are indicated. If findings are suggestive of nephrolithiasis com-
plicating the presentation, an intravenous pyelogram (IVP) or renal US should be
obtained to exclude the possibility of obstruction or hydronephrosis [116118].
Recent studies with dynamic helical CT scan are proving that this study provides
information similar to that yielded by IVP without the need for dye injection.
Dynamic CT scans can also serve as a convenient screen for abdominal aortic
aneurysm masquerading as UTI or renal colic. Additional testing may be indicated
if the diagnosis is in doubt. For example, a pelvic US may be indicated in a young
woman with pelvic tenderness, cervical discharge, and unilateral adnexal tender-
ness; a CT scan may be indicated in the elderly patient whose presentation is not
typical for UTI but who has abdominal pain, lower abdominal tenderness, and
pyuria. Catheterization is indicated if the patient cannot void spontaneously, is too
debilitated or immobilized, or if obesity prevents the patient from obtaining a
suitable specimen. Postvoiding residual urine volume, measurable by catheteri-
zation, may reveal urinary retention in a host with a defective bladder emptying
mechanism.
16.11.5 Treatment
Orally administered therapy with an antibiotic effective against Gram-negative

aerobic coliform bacteria, such as E. coli, is the principal treatment intervention in
patients with UTI. The patient with an uncomplicated, presumed, lower UTI or
simple cystitis that has symptom duration \48 h may be treated with one of the
following agents for a total of 3 days: (1) cotrimoxazole; (2) ciprofloxacin or
similar fluoroquinolone; (3) nitrofurantoin macrocrystals; (4) amoxicillin/clavul-
anate [115]. Clinical management of UTI is complicated by the increasing inci-
dence of infections caused by strains of E. coli that are resistant to commonly used
antimicrobial agents [121]. In recent studies, the rate of resistance to trimetho-
primsulfamethoxazole among E. coli isolates from women with UTI ranged from
15 to 22% [121]. Pregnant and otherwise healthy women with no evidence of
upper UTI may be treated with a 2-week course of a cephalosporin [117, 118].
Pregnant patients should be treated for all episodes of pyuria or bacteriuria,
regardless of whether they have symptoms. Ambulatory younger women who
present with signs and symptoms of pyelonephritis may be candidates for outpa-
tient therapy. They must be otherwise healthy and must not be pregnant. They
must be treated initially with vigorous fluids administered orally or i.v., antipyretic
pain medication, and a dose of parenteral antibiotics. Studies show that outpatient
therapy for selected patients is as safe as, and much less expensive than, inpatient
therapy for a comparable group of patients. The decision regarding admission of a
patient with acute pyelonephritis is dependent on age; host factors, such as im-
munocompromising chemotherapy or chronic diseases; known urinary tract
structural abnormalities; renal calculi; recent hospitalization; or urinary tract
instrumentation. Initial treatment should include i.v. antibiotic therapy (cotrim-
oxazole, directed at coliform Gram-negative bacteria; third-generation cephalo-
sporin; or an aminoglycoside), adequate fluid resuscitation to restore effective
circulating volume and generous urinary volumes, and antipyretic pain medica-
tions [115]. In the patient with a complicated UTI, coverage for unusual or mul-
tiple antibiotic-resistant organisms, such as P. aeruginosa, must be considered.
16.11.6 Catheter-Associated Urinary Tract Infections
The urinary tract is the most common site of healthcare-associated infection,

accounting for [30% of infections reported by acute-care hospitals. Virtually all
healthcare- associated UTIs are caused by urinary tract instrumentation. Colo-
nization of urinary catheters is inevitable and expected. Once microorganisms
colonize the urine, they rapidly progress, within 72 h, to concentrations [105
CFU/ml [122]. The longer the catheter is in place, the more likely colonization
will occur. It is believed that extraluminal colonization is the most likely route of
entry for microorganisms, particularly via the shorter female urethra. Urinary
catheter colonization is perpetuated by microorganism-produced biofilm. Fol-
lowing insertion of a standard urinary catheter, a conditioning film comprising
proteins, electrolytes, and other components of urine is deposited on the surface

of the catheter [123]. Initially composed of a single species, the biofilm on a
long-term indwelling urinary catheter can also contain multiple species. Under
unfavorable conditions, organisms can detach from the biofilm and become free-
floating. Free-floating organisms in the urine can lead to symptomatic infection.
Most microorganisms causing catheter-associated UTIs (CAUTI) derive from the
patients own colonic and perineal flora (such as E. coli) or from the hands of
healthcare personnel during catheter insertion or manipulation of the collection
system. Duration of catheter use is the most important risk factor for infection
[124], which increases by an estimated 510% each day the catheter remains in
place. Patients requiring long-term catheter insertion are almost assured of
developing CAUTI. Urinary stasis and bladder overdistention are additional risk
factors for bacteriuria and potential infection. Other risk factors are female sex,
older age, and failure to maintain a closed drainage system. Catheters coated
with nonantibiotic antimicrobial noble metals are now available to help reduce
CAUTI, although the main strategy remains avoiding long-term catheter use and
unnecessary repositioning.
References
1. Bosscha K, van Vroonhoven JMV, van der Werken C (1999) Surgical management of
severe secondary peritonitis. Br J Surg 86:13711377
2. Sganga G, Brisinda G, Castagneto M (2001) Peritonitis: priorities and management
strategies. In: van Saene HKF, Sganga G, Silvestri L (eds) Infection in the critically ill: an
ongoing challenge. Springer, Berlin, pp 2333
3. Sganga G (2000) Sepsi addominali chirurgiche e insufficienza multiorgano (MOFS).
Edizioni Systems Comunicazioni, Milan
4. Bone RG, Balk RA, Cerra FB et al (1992) 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 101:16441655
5. Pieracci FM, Barie PS (2007) Intra-abdominal infections. Curr Opin Crit Care 13:440449
6. Solomkin JS, Mazuski JE, Baron EJ et al (2003) Guidelines for the selection of anti-
infective agents for complicated intra-abdominal infections. Clin Infect Dis 37:9971005
7. Brook I (2008) Microbiology and Management of Abdominal Infections. Dig Dis Sci
53:25852591
8. Krepel CJ, Gohr CM, Edmiston CE, Condon RE (1995) Surgical sepsis: constancy of
antibiotic susceptibility of causative organisms. Surgery 117:505509
9. Sganga G, Brisinda G, Castagneto M (2000) Nosocomial fungal infections in surgical
patients: risk factors and treatment. Minerva Anestesiol 66(Suppl 1):7177
10. Wheeler AP, Bernard GR (1999) Treating patients with severe sepsis. N Engl J Med
340:207214
11. Mannick JA, Rodrick ML, Lederer JA (2001) The immunologic response to injury. J Am
Coll Surg 193:237244
12. Knoferl MW, Angele MK, Diodato MD et al (2002) Female sex hormones regulate
macrophage function after trauma-hemorrhage and prevent increased death rate from
subsequent sepsis. Ann Surg 235:105112
13. Sganga G, van Saene HKF, Brisinda G, Castagneto M (2001) Bacterial translocation. In:
van Saene HKF, Sganga G, Silvestri L (eds) Infection in the critically ill: an ongoing
challenge. Springer, Berlin, pp 3545
14. Gregor P, Prodger JD (1988) Mead Johnson Critical Care Symposium for the Practic
Surgeon. 4. Abdominal crisis in the intensive care unit. Can J Surg 31:331332
15. Gajic O, Errutia LE, Sewan H et al (2002) Acute abdomen in the medical intensive care
unit. Crit Care Med 30:11871190
16. Kollef MH, Allen BT (1994) Determinants of outcome for patients in the medical intensive
care unit requiring abdominal surgery: a prospective, single-center study. Chest
106:18221828
17. Velmahos GC, Kamel E, Berne TV et al (1999) Abdominal computed tomography for the
diagnosis of intra-abdominal sepsis in critically injured patients: fishing in murky waters.
Arch Surg 134:831836
18. Mokart D, Merlin M, Sannini A et al (2005) Procalcitonin, interleukin 6 and systemic
inflammatory response syndrome (SIRS): early markers of postoperative sepsis after major
surgery. Br J Anaesthesiol 94:767773
19. Wickel DJ, Cheadle WG, Mercer-Jones MA, Garrison RN (1997) Poor outcome from
peritonitis is caused by disease acuity and organ failure, not recurrent peritoneal infection.
Ann Surg 225:744756
20. Anbidge AE, Lynch D, Wison SR (2003) US of the peritoneum. Radiographics 23:663684
21. Go HL, Baarslag HJ, Vermeulen H et al (2005) A comparative study to validate the use of
ultrasonography and computed tomography in patients with post-operative intra-abdominal
sepsis. Eur J Radiol 54:383387
22. Whitehouse JS, Weigelt JA (2009) Diagnostic peritoneal lavage: a review of indications,
technique, and interpretation. Scandinavian J Trauma Resuscitation Emerg Med 17:13
23. Lin WY, Chao TH, Wang SJ (2002) Clinical features and gallium scan in the detection of
post-surgical infection in the elderly. Eur J Nucl Med 29:371375
24. Shih-Chuan T, Te-Hsin C, Lin WY, Shyh-Jen W (2001) Abdominal abscesses in patients
having surgery an application of Ga-67 scintigraphic and computed tomographic scanning.
Clin Nucl Med 26:761776
25. Marshall J, Maier RV, Jimenez M, Dellinger EP (2004) Source control in the management
of severe sepsis and septic shock: an evidence-based review. Crit Care Med 32(Suppl
11):S513S526
26. Soares-Weiser K, Paul M, Brezis M, Leibovici L (2002) Antibiotic treatment for
spontaneous bacterial peritonitis. BMJ 324:100102
27. Solomkin JS, Wilson SE, Christou N et al (2001) Results of a clinical trial of clinafloxacin
versus imipenem/cilastatin for intraabdominal infections. Ann Surg 233:7987
28. Mazuski JE (2007) Antimicrobial treatment for intra-abdominal infections. Expert Opin
Pharmacother 8(17):29332945
29. Laterre PF (2008) Progress in medical management of intra-abdominal infection. Curr Opin
Infect Dis 21(4):393398
30. Cohen J (2000) Combination antibiotic therapy for severe peritonitis. Lancet
356:15391540
31. Holzheimer RG, Dralle H (2001) Paradigm change in 30 years peritonitis treatment. A
review on source control. Eur J Med Res 6:161168
32. Schein M, Marshall J (2004) Source control for surgical infections. World J Surg
28(7):638645
33. Rivers E, Nguyen B, Havstad S et al (2001) Early goal-directed therapy in the treatment of
severe sepsis and septic shock. N Engl J Med 345:13681377
34. Platell C, Papadimitriou JM, Hall JC (2000) The influence of lavage on peritonitis. J Am
Coll Surg 191:672680
35. Sganga G, Brisinda G, Castagneto M (2002) Trauma operative procedures: timing of
surgery and priorities. In: Gullo A (ed) Critical care medicine. Springer, Berlin, pp 447467
36. Levison MA, Zeigler D (1991) Correlation of APACHE II score, drainage technique and
outcome in postoperative intra-abdominal abscess. Surg Gynecol Obstet 172:8994
37. Baril NB, Ralls PW, Wren SM et al (2000) Does an infected peripancreatic fluid collection
or abscess mandate operation? Ann Surg 231:361367
38. Brisinda G, Maria G, Ferrante A, Civello IM (1999) Evaluation of prognostic factors in
patients with acute pancreatitis. Hepatogastroenterology 46:19901997
39. Baron TH, Morgan DE (1999) Acute necrotizing pancreatitis. N Engl J Med 340:14121417
40. Beger HG, Rau B, Mayer J, Pralle U (1997) Natural course of acute pancreatitis. World J
Surg 21:130135
41. Bradley EL III (1993) A clinically based classification system for acute pancreatitis.
Summary of the international symposium on acute pancreatitis, Atlanta, GA, Sept. 1113,
1992. Arch Surg 128:586590
42. Marotta F, Geng TC, Wu CC, Barbi G (1996) Bacterial translocation in the course of acute
pancreatitis: beneficial role of nonabsorbable antibiotics and Lactinol enemas. Digestion
57:446452
43. Foitzik T, Fernandez-del Castillo C, Ferraro MJ et al (1995) Pathogenesis and prevention of
early pancreatic infection in experimental acute necrotizing pancreatitis. Ann Surg
222:179185
44. Isenmann R, Schwarz M, Rau B et al (2002) Characteristics of infection with Candida
species in patients with necrotizing pancreatitis. World J Surg 26(3):372376
45. Bradley EL III (1994) Surgical indications and techniques in necrotizing pancreatitis. In:
Bradley EL III (ed) Acute pancreatitis: diagnosis and therapy. Raven Press, New York,
pp 105117
46. Paye F, Rotman N, Radier C et al (1998) Percutaneous aspiration for bacteriological studies
in patients with necrotizing pancreatitis. Br J Surg 85:755759
47. Rau B, Pralle U, Mayer JM, Beger HG (1998) Role of ultrasonographically guided fine-
needle aspiration cytology in diagnosis of infected pancreatic necrosis. Br J Surg
85:179184
48. Mithofer K, Fernandez-del Castillo C, Ferraro MJ et al (1996) Antibiotic treatment improves
survival in experimental acute necrotizing pancreatitis. Gastroenterology 110:232240
49. Sainio V, Kemppainen E, Puolakkainen P et al (1995) Early antibiotic treatment in acute
necrotising pancreatitis. Lancet 346:663667
50. Luiten EJ, Hop WC, Lange JF, Bruining HA (1995) Controlled clinical trial of selective
51. Luiten EJ, Hop WC, Lange JF, Bruining HA (1997) Differential prognosis of gram-negative
versus gram-positive infected and sterile pancreatic necrosis: results of a randomized trial in
patients with severe acute pancreatitis treated with adjuvant selective decontamination. Clin
52. Pederzoli P, Bassi C, Vesentini S, Campedelli A (1993) A randomized multicenter clinical
trial of antibiotic prophylaxis of septic complications in acute necrotizing pancreatitis with
imipenem. Surg Gynecol Obstet 176:480483
53. Ho HS, Frey CF (1997) The role of antibiotic prophylaxis in severe acute pancreatitis. Arch
Surg 132:487493
54. Bassi C, Falconi M, Talamini G et al (1998) Controlled clinical trial of pefloxacin versus
imipenem in severe acute pancreatitis. Gastroenterology 115:15131517
55. McClave SA, Snider H, Owens N, Sexton LK (1997) Clinical nutrition in pancreatitis. Dig
Dis Sci 42:20352044
56. Kalfarentzos F, Kehagias J, Mead N et al (1997) Enteral nutrition is superior to parenteral
nutrition in severe acute pancreatitis: results of a randomized prospective trial. Br J Surg
84:16651669
57. Windsor AC, Kanwar S, Li AG et al (1998) Compared with parenteral nutrition, enteral
feeding attenuates the acute phase response and improves disease severity in acute
pancreatitis. Gut 42:431435
58. Bozzetti F, Braga M, Gianotti L et al (2001) Postoperative enteral versus parenteral nutrition
in malnourished patients with gastrointestinal cancer: a randomised multicentre trial. Lancet
358:14871492
59. Warshaw AL (2000) Pancreatic necrosis. To debride or not to debridethat is the question.
Ann Surg 232:627629
60. Buchler MW, Gloor B, Muller CA et al (2000) Acute necrotizing pancreatitis: treatment
strategy according to the status of infection. Ann Surg 232:619626
61. Farkas G, Marton J, Mandi Y, Szederkenyi E (1996) Surgical strategy and management of
infected pancreatic necrosis. Br J Surg 83:930933
62. Ashley SW, Perez A, Pierce EA et al (2001) Necrotizing pancreatitis. Contemporary
analysis of 99 consecutive cases. Ann Surg 234:572580
63. Gloor B, Muller CA, Worni M et al (2001) Pancreatic infection in severe pancreatitis. The
role of fungus and multiresistant organisms. Arch Surg 136:592596
64. Doglietto GB, Gui D, Pacelli F et al (1994) Open vs closed treatment of secondary
pancreatic infection. A review of 42 cases. Arch Surg 129:689693
65. Kriwanek S, Gschwantler M, Beckerhinn P et al (1999) Complications after surgery for
necrotising pancreatitis: risk factors and prognosis. Eur J Surg 165:952957
66. Dervenis C, Bassi C (2000) Evidence-based assessment of severity and management of
acute pancreatitis. Br J Surg 87:257258
67. Tsiotos GG, Luque-de Leon E, Sarr MG (1998) Long-term outcome of necrotizing
pancreatitis treated by necrosectomy. Br J Surg 85:16501653
68. del Castillo CF, Rattner DW, Makary MA et al (1998) Debridement and closed packing for
treatment of necrotizing pancreatitis. Ann Surg 228:676684
69. Farkas G (2000) Pancreatic head mass: how can we treat it? Acute pancreatitis: surgical
treatment. JOP J Pancreas 1(Suppl 3):138142
70. Mithofer K, Mueller PR, Warshaw AL (1997) Interventional and surgical treatment of
pancreatic abscess. World J Surg 21:162168
71. Widdison AL, Karanjia ND (1993) Pancreatic infection complicating acute pancreatitis. Br
J Surg 80:148154
72. Lumsden A, Bradley EL III (1990) Secondary pancreatic infections. Surg Gynecol Obstet
170:459467
73. Aranha GV, Prinz RA, Greenlee HB (1982) Pancreatic abscess: an unresolved surgical
problem. Am J Surg 144:534538
74. Warshaw AL, Jin G (1985) Improved survival in 45 patients with pancreatic abscess. Ann
Surg 202:408417
75. Brisinda G, Mazzari A, Crocco A et al (2011) Open pancreatic necrosectomy in the
multidisciplinary management of postinflammatory necrosis. Ann Surg 253(5):10491051
76. Ross Carter C, McKay CJ, Imrie CW (2000) Percutaneous necrosectomy and sinus tract
endoscopy in the management of infected pancreatic necrosis: an initial experience. Ann
Surg 232:175180
77. Kjossev KT, Losanoff JE (2001) Laparoscopic treatment of severe acute pancreatitis. Surg
Endosc 15:12391240
78. Freeny PC, Hauptmann E, Althaus SJ et al (1998) Percutaneous CT-guided catheter
drainage of infected acute necrotizing pancreatitis: techniques and results. AJR Am J
Roentgenol 170:969975
79. Baron TH, Morgan DE (1997) Organized pancreatic necrosis: definition, diagnosis, and
management. Gastroenterol Int 10:167178
80. Wada K, Takada T, Hirata K et al (2010) Treatment strategy for acute pancreatitis.
J Hepatobiliary Pancreat Sci 17:7986
81. Zhu JF, Fan XH, Zhang XH (2001) Laparoscopic treatment of severe acute pancreatitis.
Surg Endosc 15:146148
82. Clancy CJ, Nguyen MH, Morris AJ (1997) Candidal mediastinitis: an emerging clinical
entity. Clin Infect Dis 25:608613
83. Gamlin F, Caldicott LD, Shah MV (1994) Mediastinitis and sepsis syndrome following
intubation. Anaesthesia 49:883885
84. Isaacs L, Kotton B, Peralta MM Jr et al (1993) Fatal mediastinal abscess from upper
respiratory infection. Ear Nose Throat J 72:620622
85. Becker M, Zbaren P, Hermans R et al (1997) Necrotizing fasciitis of the head and neck: role
of CT in diagnosis and management. Radiology 202:471476
86. Brunelli A, Sabbatini A, Catalini G, Fianchini A (1996) Descending necrotizing
mediastinitis. Surgical drainage and tracheostomy. Arch Otolaryngol Head Neck Surg
122:13261329
87. Corsten MJ, Shamji FM, Odell PF et al (1997) Optimal treatment of descending necrotising
mediastinitis. Thorax 52:702708
88. Baldwin RT, Radovancevic B, Sweeney MS (1992) Bacterial mediastinitis after heart
transplantation. J Heart Lung Transpl 11:545549
89. El Oakley RM, Wright JE (1996) Postoperative mediastinitis: classification and
management. Ann Thorac Surg 61:10301036
90. Milano CA, Kesler K, Archibald N (1995) Mediastinitis after coronary artery bypass graft
surgery. Risk factors and long-term survival. Circulation 92:22452251
91. Brook I, Frazier EH (1996) Microbiology of mediastinitis. Arch Intern Med 156:333336
92. Shaffer HA Jr, Valenzuela G, Mittal RK (1992) Esophageal perforation. A reassessment of
the criteria for choosing medical or surgical therapy. Arch Intern Med 152:757761
93. Loop FD, Lytle BW, Cosgrove DM (1990) J Maxwell Chamberlain memorial paper. Sternal
wound complications after isolated coronary artery bypass grafting: early and late mortality,
morbidity, and cost of care. Ann Thorac Surg 49:179187
94. Gadek JE, DeMichele SJ, Karlstad MD (1999) Effect of enteral feeding with
eicosapentaenoic acid, gamma-linolenic acid, and antioxidants in patients with acute
respiratory distress syndrome. Enteral nutrition in ARDS study group. Crit Care Med
27:14091420
95. Weinzweig N, Yetman R (1995) Transposition of the greater omentum for recalcitrant
median sternotomy wound infections. Ann Plast Surg 34:471477
96. Andrews CO, Gora ML (1994) Pleural effusions: pathophysiology and management. Ann
Pharmacother 28:894903
97. Bartter T, Santarelli R, Akers SM (1994) The evaluation of pleural effusion. Chest
106:12091214
98. Fenton KN, Richardson JD (1995) Diagnosis and management of malignant pleural
effusions. Am J Surg 170:6974
99. Kennedy L, Sahn SA (1994) Noninvasive evaluation of the patient with a pleural effusion.
Chest Surg Clin N Am 4:451465
100. Light RW (1995) Pleural diseases, 3rd edn. Williams & Wilkins, New York
101. Sahn SA (1988) State of the art. The pleura. Am Rev Respir Dis 138:184234
102. Sganga G, Brisinda G, Castagneto M (2001) Practical aspects of antibiotic prophylaxis in
high-risk surgical patients. In: van Saene HKF, Sganga G, Silvestri L (eds) Infection in the
critically ill: an ongoing challenge. Springer, Berlin, pp 4758
103. Mangram AJ, Horan TC, Pearson ML et al (1996) The hospital infection control practices
advisory committee. Guideline for the prevention of surgical site infection. Infect Control
104. Sganga G, Cozza V (2009) Intra-abdominal infections: diagnostic and surgical strategies.
In: Gullo A, Besso J, Lumb PD, Williams GF (eds) Intensive and critical care medicine.
(WFSICCM) World federation of societies of intensive and critical care medicine. Springer,
Milan, pp 315324
105. Kurz A, Sessler DI, Lenhardt R (1996) Perioperative normothermia to reduce the incidende
of surgical-wound infection and shorten hospitalization. N Engl J Med 334:12091215
106. Classen DC, Evans RS, Pestotnik A (1992) The timing of prophylactic administration of
antibiotics and the risk of surgical-wound infection. N Engl J Med 326:281287
107. Nathens AB, Marshall JC (1999) Selective decontamination of the digestive tract in surgical
patients. A systematic review of the evidence. Arch Surg 134:170176
108. Silvestri L, Mannucci F, van Saene HKF (2000) Selective decontamination of the digestive
tract: a life-saver. J Hosp Infect 45:185190
109. van Saene HKF, Silvestri L, de la Cal M (2000) Prevention of nosocomial infections in the
intensive care unit. Curr Opin Crit Care 6:323329
110. Greif R, Akca O, Horn EP et al (2000) Supplemental perioperative oxygen to reduce the
incidence of surgical-wound infection. N Engl J Med 342:161167
111. Jancel T, Dudas V (2002) Management of uncomplicated urinary tract infections. West J
Med 176:5155
112. Larcombe J (1999) Urinary tract infection in children. BMJ 319:11731175
113. Ellis AK, Verma S (2000) Quality of life in women with urinary tract infections: is benign
disease a misnomer? J Am Board Fam Pract 13:392397
114. Hooten TM, Scholes D, Stapleton AE (2000) A prospective study of asymptomatic
bacteriuria in sexually active young women. N Engl J Med 343:992997
115. Howes DS, Young WF (2000) Urinary tract infections. In: Tintinalli A (ed) Emergency
medicine. A comprehensive study guide. McGraw-Hill, New York, pp 625631
116. Leibovici L, Greenshtain S, Cohen O, Wysenbeek AJ (1992) Toward improved empiric
management of moderate to severe urinary tract infections. Arch Intern Med
152:24812486
117. Millar LK, Wing DA, Paul RH, Grimes DA (1995) Outpatient treatment of pyelonephritis in
pregnancy: a randomized controlled trial. Obstet Gynecol 86:560564
118. Safrin S, Siegel D, Black D (1988) Pyelonephritis in adult women: inpatient versus
outpatient therapy. Am J Med 85:793798
119. Stamm WE, Hooton TM (1993) Management of urinary tract infections in adults. N Engl J
Med 329:13281334
120. Foxman B, Barlow R, DArcy H et al (2000) Urinary tract infection: self-reported incidence
and associated costs. Ann Epidemiol 10:509515
121. Manges AR, Johnson JR, Foxman B et al (2001) Widespread distribution of urinary tract
infections caused by a multidrug-resistant Escherichia coli clonal group. N Engl J Med
345:10071013
122. Stark RP, Maki DG (1984) Bacteriuria in the catheterized patient. N Engl J Med
311:560564
123. Trautner BW, Darouiche RO (2004) Catheter-associated infections. Arch Intern Med
164:842850
124. Lo E, Lindsay N, Classen D et al (2008) Strategies to prevent catheter-associated urinary
tract infections in acute care hospitals. Infect Control Hosp Epidemiol 29:S41S50
Infection in the NICU and PICU
17
A. J. Petros, V. Damjanovic, A. Pigna and J. Farias
17.1 Introduction
In this chapter, we summarise current concepts of infections in neonatal (NICU)

and paediatric (PICU) intensive care units and describe the only two proven
antibiotic manoeuvres to prevent such infections: surgical prophylaxis, and
decontamination of the digestive tract (SDD).
17.2 Current Concepts of Infection
17.2.1 Neonatal Intensive Care Unit
Infections in neonates requiring intensive care are unique in each essential element
of the pathogenesis of infection, i.e. the potential pathogen and its source, the mode
of transmission and the susceptible host. The pathogen, e.g. hepatitis B virus, or
potential pathogen, e.g. Escherichia coli, are closely related to source and mode of
transmission. Many microorganisms are present in the maternal birth canal (the
source). They are most commonly Streptococcus agalactiae, E. coli, Herpes simplex
virus, Listeria monocytogenes and Candida albicans. One or more of these micro-
organisms can be vertically transmitted from the mother to the neonate. When this
type of infection occurs, it will always be present in the first week of the neonates
life (early onset). On the other hand, different microorganisms are acquired in
the NICU: in general, these are coagulase-negative staphylococci (CNS),
A. J. Petros (&)
PICU, Great Ormond Street Hospital, London, UK
290 A. J. Petros et al.
aerobic Gram-negative bacilli (AGNB) (mainly Klebsiella spp. and Pseudomonas

aeruginosa) S. aureus and Candida spp. The sources of these microorganisms
acquired in the NICU are mainly other neonates who carry those microorganisms
and/or who are infected with them. NICU staff, mothers, contaminated materials and
equipment (environment) are uncommon sources. Although these microorganisms
can be transmitted from one neonate to another via equipment, the hands of
healthcare workers are the main mode of transmission [1]. Infections due to
microorganisms acquired in the unit are usually of late onset following episode
period of carriage.
The incidence of infection is higher in the neonatal period than at any time
in life. Neonates, particularly preterm, are extremely susceptible to infection,
and low birth weight is the single most important risk factor [2]. This increased
risk is due primarily to immune system immaturity, poor surface defences, lack
of colonisation resistance [3, 4], invasive medical devices and use of broad-
spectrum antibiotics. Increased susceptibility to carriage and infection in pre-
term neonates is the primary contributor to transmission of potential pathogens
and subsequent outbreaks of infection on the NICU [1]. Moreover, preterm
neonates can be susceptible to new and potential pathogens, usually unknown
in an ICU, such as Hansenula anomala, a saprophytic yeast known as a
contaminant in the brewing industry. This newly recognised potential pathogen
caused an outbreak that lasted for 13 months in the Mersey (UK) regional
NICU [5].
17.2.2 Paediatric Intensive Care Unit
The three elementspotential pathogen, source and mode of transmission and

susceptible host applies to PICU patients. Recent epidemiological studies in
children requiring prolonged PICU stay demonstrated that two-thirds of all
infections diagnosed were due to microorganisms present in patients admission
flora [6]. These infections practically all manifested within in a week of PICU
admission. Infections due to microorganisms acquired in the unit and subsequently
carried invariably manifested after 1 week. The three main microorganisms
causing infections within the first week are coagulase negative staphylococci,
S. aureus and C. albicans; after one week, the two main microorganisms are
S. aureus and P. aeruginosa. Unlike in the neonate, maternal flora is not the source
in a PICU; it is invariably the other patients.
Length of stay in the NICU is substantially longer that in the PICU (median 13
vs. 6.5 days) [4, 7, 8]. An extensive literature search showed that outbreaks are
more common in the NICU (Chap. 13). Finally, higher overall mortality rates of
10% versus 5% [9] support the observation that children in the NICU are more
susceptible than children in the PICU [4].
17 Infection in the NICU and PICU 291
17.3.1 Neonatal Intensive Care
The overall infection rates in neonates on intensive care vary between 15 and 20%.
This is equal to rates reported for adult medical and surgical units and higher than
most paediatric units [10]. The main site of infection is the bloodstream, followed
by the lower airways. In a multicenter study of NIUCs in Oakland, New Haven,
CT, USA in 1994, Beck-Sague et al. [11] reported that nosocomial bloodstream
infection occurred at a rate of 5% when surveillance cultures were performed and
was actually half that reported in studies reporting the rate of all infections.
Bloodstream infections can account for 50% of all NICU infections. Lower airway
infections occur in approximately 3% of neonates during NICU stay [12]. The
main organisms are viruses, S. aureus and AGNB. The survival benefit of NICU
neonates has significantly increased over the last 25 years. In a 2-year study from
New York [19771978], the mortality rates for early-onset sepsis in neonates
\1,000 g was 53.4% and for late-onset sepsis 20.3% [13]. A 5-year study from
Oxford, UK (19821986) reports mortality rates of 28% and 4% for early- and
late-onset sepsis, respectively, in neonates [2]. Data from a 1-year study in a Dutch
NICU show a mortality rate \10% (1997) in 436 neonates of about 2,000 g [4].
17.3.2 Paediatric Intensive Care
Nosocomial infection in the PICU is an important cause of morbidity and mortality

in ventilated children. Blood stream and lower airways are most common and are
almost always due to prolonged use of devices. The incidence of blood stream
infection is reported as 10.646.9/1,000 catheter days [14] and of lower airways
infection 6.520.2/1,000 ventilation days.
17.3.2.1 Bloodstream Infections

In a report from a mixed PICU in Birmingham, UK [15], where all children admitted
were included, the incidence of blood stream infection was 10.6/1,000 patient days.
Consequently, the group as a whole was less ill and stayed for a median of 3 days.
The larger denominator of[12,000 patient days also dilutes the real infection rate:
62% of microorganisms causing positive blood cultures were G ? bacteria, mainly
CNS, S. aureus and enterococci; 32% were due to AGNB; the remaining were due to
yeasts. In a study from Liverpool, UK, 1,000 children requiring a median of 8 days
ventilation, the overall infection rate was 42.2% [9]. Viral infection accounted for
13.6% and bacterial/yeast infections for 34.6%. The incidence of bloodstream
infection was 21/1,000 patient days. The infection rate due to microorganisms
acquired in the PICU was 15.8%; 3.7% of admitted children developed infections due
to resistant microorganisms. Causative microorganisms were CNS, enterococci,
Pseudomonas spp., S. aureus and yeasts. A study from London, UK, reported an
incidence of bloodstream infection of 46.9/1,000 patient days in a subset of 103
children with a line in situ for a median of 6 days [16]. The causative organisms were
CNS, S. aureus, C. albicans and Klebsiella spp.
17.3.2.2 Lower Airway Infections

In a paediatric trauma unit, the rate of lower airway infections was 5.5% [17]. The
most common organisms were S. aureus, Haemophilus influenzae, Enterobacter
and Pseudomonas spp. [17]. In the Liverpool study, the overall airway infection
rate was 11%, with a rate of 9.3 episodes/1,000 patient days [9]. The three main
organisms were S. aureus, P. aeruginosa and H. influenzae.
17.4 Assessing the Magnitude of the Problem
17.4.1 Carrier State: Endogenous Versus Exogenous
The magnitude of the problem can be assessed in different ways based upon carrier
state (Chap. 5). Endogenous must be distinguished from exogenous infection.
Endogenous infection is caused by potential pathogens previously carried by the
patient; if the potential pathogen was present on that patients admission, then the
infection due to this potential pathogen is called primary endogenous. This type of
infection tends to occur early, within the first week. If the infection is due to a
potential pathogen acquired in the unit, after the patient goes through the carriage
phase, then the infection is termed secondary endogenous. Infections caused by
microorganisms not carried by the patient at all are termed exogenous. Obviously,
surveillance cultures are indispensable for this classification [6, 9].
17.4.2 Pathogenicity Index
Some microorganisms cause more serious clinical disease than others. This dif-
ferential pathogenic effect can be used to develop a pathogenicity index for an
individual microorganism, in a specific organ system and in a particular homo-
geneous population for which surveillance cultures are useful [18]. The ratio
between the number of ICU patients infected by a particular microorganisms and
the number of patients simply carrying that organism in their throat and/or gut is
defined as the intrinsic pathogenicity index for a particular microorganism.
Indigenous flora, including anaerobes, will rarely cause infections in the lower
airways of patients requiring ventilation for more than 3 days despite being carried
in high concentrations. This is because they have intrinsic pathogenicity index
values of between 0.01 and 0.03. Low-level pathogens, such as viridans strepto-
cocci, enterococci and CNS are also carried in high concentrations in the
oropharynx by a substantial percentage of ICU patients and are unable to cause
lower airway infections. High-level pathogens such as S. pyogenes and Salmonella
spp. have an intrinsic pathogenicity index approaching 1 and diseases manifest in
virtually all oropharyngeal and gut carriers. The concept of carriage recognises
Oropharynx Wound
N. mening
Line
Lung
S. aureus
S. pneum Skin
S. epi
S. epi
Candida
B Strep
E. coli
E. coli Colon
Bladder
Rectum
Fig. 17.1 Schematic representation of the digestive tract, illustrating that the throat and gut are
the major internal sources of potential pathogens causing endogenous infections of blood, lower
airways, bladder and wounds
about 15 potentially pathogenic microorganisms, with intrinsic pathogenicity

indices between 0.1 and 0.3. these consist of the six normal microorganisms
S. pneumoniae, H. influenzae, Moraxella catarrhalis, E. coli, S. aureus and
C. albicans present in previously healthy individuals, and nine abnormal bacteria
carried by patients with an underlying chronic or acute condition, namely:
Klebsiella, Proteus, Morganella, Enterobacter, Citrobacter, Serratia, Pseudomonas
and Acinetobacter spp. and methicillin-resistant S. aureus (MRSA). The overall
mortality in our PICU is in the range of 5%, but the mortality rate rises to 10% in the
subset of children who require prolonged mechanical ventilation [9].
17.5 Pathogenesis
Figure 17.1 describes the pathogenesis of infection in neonates and children

requiring intensive care. Practically all infections in these two groups are
endogenouspatients become infected with microorganisms they carry. A recent
study in 400 PICU children requiring ventilation demonstrated that 90% of all
lower airway infections were endogenous; 80% were primary endogenous, 10%
secondary endogenous and the remaining 10% exogenous [6].
Bloodstream infections occur due to translocation. Microorganisms in over-
growth of the terminal ileum ([105 microorganisms/ml) migrate into the blood-
stream [19]. This mechanism applies to S. agalactiae, S. aureus and C. spp.
Neonates and children staying longer that 1 week in the NICU or PICU, CNS and
AGNB cause septicaemia due to translocation [20].
Lower airway infections are caused by microorganisms carried in the oro-
pharynx, which then migrate into the lower airways. In a previous healthy child
S. pneumoniae, H. influenzae and S. aureus cause bacterial lower airway infec-
tions. AGNB and MRSA are causative organisms in children who require intensive
care [1 week.
Bladder infections are, in general, endogenous due to migrating faecal bacteria.
Wound infections of the head, neck and thorax are, in general, caused by oral
bacteria, whereas wound infections between the waist and knee are primarily
caused by gut bacteria.
Exogenous infections vary between 5% and 25% and are a particular problem in
patients with tracheostomies [21]. Children with wounds, particularly burns, are at
high risk of exogenous colonisation and infection [9]. Up to 16% of bloodstream
infections are of exogenous pathogenesis following contamination of an indwelling
intravascular device [19]. Gastrostomies can also be considered as a wound, and
recurrent exogenous colonisation/infection is not uncommon in children with such
devices [9]. To identify an exogenous infection, surveillance samples of throat and
rectum are indispensable. Blood cultures or lower airway secretions are positive for
a potential pathogen that is not present in throat and or rectal cultures.
17.5.1 Risk Factors
Risk factor analysis in the pathogenesis of NICU and PICU infections invariably
includes low birth weight; administration of total parenteral nutrition; presence of
invasive and indwelling devices, including endotracheal tube and mechanical
ventilation; length of stay; and prior use of antibiotics [2, 22, 23]. All these factors
are reflected in illness severity and are difficult to modify to control infection. Risk
factor analysis cannot easily contribute to infection control.
17.6 Diagnosis
17.6.1 Infection
Infection is a microbiologically proven, clinical diagnosis of local and/or general

inflammation. The signs of generalised infections in NICU patients, e.g. septi-
caemia, are often nonspecific and may be clinically indistinguishable from those of
noninfectious conditions [2]. For instance, the clinical picture of respiratory dis-
tress in early onset sepsis may be identical to hyaline membrane disease.
Furthermore, the clinical diagnosis of local infection, such as meningitis, may not
differ from that of systemic sepsis without meningeal involvement. On the other
hand, infections in the PICU patient are more specific, and the following
description of local and general infection is related to paediatric patients.
17.6.2 Pneumonia
17.6.2.1 Microbiologically Proven Pneumonia

1. presence of new or progressive pulmonary infiltrates on a chest X-ray for
C48 h, and
2. purulent tracheal aspirate, and
3. fever C38.5C, and
4. leucocytosis [white blood cells (WBC) [ 12,000/ml] or leucopenia
(WBC \ 4,000/ml), and
5. tracheal aspirate C105 colony forming units (CFU)/ml of potentially patho-
genic microorganism or bronchoalveolar lavage (BAL) yielding C104 CFU/ml.
17.6.2.2 Clinical Diagnosis

Criteria 14 above and sterile BAL or tracheal aspirate.
17.6.3 Tracheitis/Bronchitis
1. purulent tracheal aspirate, and

2. fever C38.5C, and
3. leucocytosis (WBC [ 12,000/ml) or leucopenia (WBC \ 4,000/ml), and
4. tracheal aspirate yielding C105 CFU/ml, and, most importantly
5. normal chest X-ray.
17.6.4 Systemic Inflammatory Response Syndrome
Clinical signs of generalised inflammation caused by microorganisms and/or their

products, including at least three of the following: fever, temperature instability,
lethargy, poor perfusion, hypotension.
17.6.5 Blood Stream Infections
Criteria as for Systemic Inflammatory Response Syndrome (SIRS), with a positive

blood culture from either a peripheral vein or an intravascular device.
Intra-abdominal infection is defined as affecting an abdominal organ and the

peritoneal cavity (peritonitis), with local signs such as abdominal tenderness and
generalised symptoms including fever and leucocytosis. Peritonitis can be a
localised or generalised infection of the peritoneal cavity. Following ultrasound
and/or computed tomography scan and/or laparotomy, the diagnosis is confirmed
by the isolation of microorganisms of C3 ? or C105 CFU/ml and [++leucocytes
in the diagnostic sample [24].
Infection of the urinary tract most often involves the bladder. The common fea-
tures of dysuria, suprapubic pain and urinary frequency and urgency are often not
assessable in PICU patients. Therefore, the diagnosis of cystitis is based upon
freshly obtained catheter urine containing C105 CFU/ml and C5 WBC/ml high
power light microscopy field.
Purulent discharge from wounds, a culture yielding C 3 ? or C 105 CFU/ml of

pus and signs of local inflammation. Isolation of skin flora is considered to be
contamination.
17.7 Prevention
Besides the five infection-control interventions (Chap. 10), there is evidence for
the effectiveness of only two antibiotic manoeuvres that prevent infection in NICU
and PICU patients: surgical prophylaxis [25-27] and selective decontamination of
the digestive tract (SDD) (Table 17.1) [2833].
17.7.1 Cardiac Surgical Prophylaxis
The aim of prophylactic antibiotics in cardiac surgery is to prevent infections of the

mediastinal incision and the heart. The main microorganisms causing endocarditis
are CNS, viridans streptococci and enterococci; less common are AGNB and yeasts.
S. aureus, both sensitive and resistant to methicillin, are the main cause of medias-
tinal wound infections. A commonly used combination of antimicrobials prior to
cardiac surgery is a glycopeptide and an aminoglycoside. A glycopeptide such as
teicoplanin covers all streptococci and staphylococci, whereas an aminoglycoside
such as netilmicin is active against AGNB that may translocate following gut
Table 17.1 Cardiac and general surgical prophylaxis and prevention protocol for selective
decontamination of the digestive tract (SDD)
Surgical prophylaxis Total daily dose (mg/kg)
\7 days [7 days 1 month12 years [12 years

Cardiac
Teicoplanin 16 then 8 20, then 10, 400 mg then
then 6 200 mg
Netilmicin 3 \ 2 kg 6 6 200 mg
6 [ 2 kg
General
Cefotaxime 100 150 100200 612 g
Metronidazole 22.5 22.5 22.5 1.5 g
Gentamicin 3 \ 2 kg 6 \ 2 kg 7.5 35
6 [ 2 kg 7.5 [ 2 kg
SDD Total daily dose (mg/day)

Oropharynx
AGNB: polymyxin E 2 g of 2% paste/gel 2 g of 2% paste/gel 2 g of 2% paste/gel
with tobramycin
Yeasts: amphotericin B 2 g of 2% paste/gel 2 g of 2% paste/gel 2 g of 2% paste/gel
MRSA: vancomycin 2 g of 2% paste/gel 2 g of 2% paste/gel 2 g of 2% paste/gel
Gut
AGNB: polymyxin E 100 200 400
with tobramycin 80 160 320
Yeasts: amphotericin B 400 1,000 2,000
MRSA: vancomycin 2040 mg/kg 2040 mg/kg 5002,000
Therapy Total daily dose (mg/kg)
\7 days [7 days 1 month12 years [12 years

Neonatal ICU
Ampicillin: active against 50 100
L. monocytogenes and
S. agalactiae
Gentamicin: AGNB
Paediatric ICU
Cefotaxime: community 100 150 100200 612 g
? hospital microbes
excluding
P. aeruginosa
Ceftazidime: 60 90 100150 69 g
P. aeruginosa
Gentamicin: AGNB 3 \ 2 kg 6 \ 2 kg 7.5 35
6 [ 2 kg 7.5 [ 2 kg
Cephradine: S.aureus 50 50 100 4g
Vancomycin: MRSA 15 then 20 15 then 30 45 2g
Amphotericin B: yeasts, 13 13 13 13
fungi (lipophilic)
ICU intensive care unit; AGNB aerobic Gram-negative bacilli; MRSA methicillin-resistant Staphylo-
coccus aureus; L. monocytogenes, Listeria monocytogenes; S. agalactiae, Streptococcus agalactiae;
P. aeruginosa, Pseudomonas aeruginosa
ischaemia. Short-term prophylaxis of three doses is normally administered: one

immediately prior to surgery to achieve high tissues concentrations, and two further
doses 8-h apart. Under certain circumstances, such as a chest splinted open because
of cardiac oedema, these antibiotics may be continued for 5 days, though there is no
evidence to support the added efficacy of this practice.
17.7.2 General Surgical Prophylaxis
The type of antimicrobial prescribed depends on the proposed surgery and the
associated contamination risk. Clean, sterile procedures do not need antibiotic
cover, whereas clean procedures with the likelihood of contamination need cover
with one antimicrobial, such as cefotaxime. If there is likely to be faecal con-
tamination, then an aminoglycoside such as gentamicin is also necessary to cover
AGNB and enterococci. Finally, if the surgical procedure is likely to be associated
with ischaemia and possible necrotic tissue, then metronidazole should be added to
the prophylactic regimen. Again, three doses as above will suffice.
SDD is a prophylactic intervention designed to prevent early and late infection and
is recommended in the critically ill child requiring [1 week of intensive care
(Chap. 14). Four prospective randomised controlled trials [2832] demonstrate a
significant reduction in infectious morbidity using SDD. As the overall mortality
rate in this population is approximately 10%, a reduction in that rate is harder to
demonstrate than in adults; a huge sample size would be necessary. However, in
adults, where the overall mortality rate is approximately 30%, this method dem-
onstrates a significant reduction of 40% [3335].
There is particular indication for SDD in the NICU, namely, for controlling an
infection outbreak. A decade ago, SDD with nystatin was used to control a
C. parapsilosis outbreak in the Mersey, UK, regional NICU: 76 of 106 neonates who
carried the outbreak strain received nystatin in the throat and gut during the
12-month open trial; six neonates developed fungaemia. Once the carriage rate fell
from 50 to 5%, no new cases of systemic Candida infection were observed. This was
the first report of SDD intervention to control an infection outbreak in an NICU [36].
17.8 Meta-Analysis of Randomised Control Trials in Children

with Severe Lower Airway and Bloodstream Infections
Using SDD
Data was extracted from four randomized controlled trials (RCTs) of selective
digestive decontamination in the paediatric population (Table 17.2). The four
RCTs enrolled 335 patients. Pneumonia occurred in five of 170 (2.9%) of patients
Table 17.2 Data extracted from four randomised controlled trials of selective decontamination
of the digestive tract in the paediatric population
Author Patients Patients with infection Patients with pneumonia Mortality
SDD C SDD C SDD C SDD C

Zobel [28] 25 25 2 10 1 6 3 2
a
Smith [29] 18 18 3 11 0 2 2 3
Ruza [30] 116 110 NA NA 3 8 6 5
Barret [31] 11 12 5 3 1 0 2 1
SDD selective decontamination of the digestive tract; C control; NA not available
a
Patients with Gram-negative infections
Table 17.3 Meta-analysis of the impact of selective decontamination of the digestive tract on
secondary endpoints
Outcome RCTs No. of No. of patients with ORa (95 CI) p value I2 (%)
patients outcome
SDD C SDD C
Infection 3 54 55 10 24 0.34 (0.052.18) 0.25 4.7
Mortality 4 170 163 13 11 1.18 (0.502.76) 0.70 0
RCTs randomised controlled trials; OR odds ratio; CI confidence interval; SDD selective
decontamination of the digestive tract; C control
a
OR less than the unit favours treatment; OR more than the unit favours controls
who received SDD and in 16 of 163 (9.8%) in the control group. This was a
significant reduction in the incidence of pneumonia with SDD [odds ratio (OR)
0.31; 95% confidence interval (CI), 0.110.87; P = 0.027]. A meta-analysis was
performed on the impact of SDD on the secondary endpoints of pneumonia and
overall infection rates (Table 17.3). In the three eligible RCTs of a total of 109
children, infections of various origins were confirmed in ten of 54 (13%) on SDD
and in 24 of 55 (15.9%) in the control group. SDD had no impact on general
infection rates, with no significant difference between groups (OR 0.34; 95% CI
0.052.18; P = 0.25).
Subgroup analyses of type of SDD regimen, randomisation and blinding
revealed a significant impact on pneumonia and infection rates when the full
protocol of parenteral and enteral antimicrobials was used rather than solely
enterally administered antimicrobials. A significant impact on pneumonia and
overall infection was demonstrated when randomisation was adequate and in
unblinded studies. The subgroup analyses for mortality were consistent with
previous pooled results whether the intervention was parenteral/enteral or enteral,
whether the design was blinded or not and whether the randomisation process was
adequate or not. SDD made no significant impact on mortality rates.
Day one: absence of knowledge of causative microorganism

Empirical treatment: Cefotaxime combined with gentamicin if seriously ill
Day two: presumptive identification of causative microorganism

Tailored treatment:
Normal potential patmhogen Abnormal potential pathogen
S. pneumoniae S. aureus Candida spp. Aerobic Pseudomonas spp.
Gram-
negative
bacilli
Stop Stop Stop Continue Replace cefotaxime
gentamicin cefotaxime/ cefotaxime/ cefotaxime/ with ceftazidime
gentamicin gentamicin gentamicin
Monotherapy: Monotherapy: Monotherapy:

cefotaxime cephradine amphotericin
B
Day three: clinical improvement
Day five: stop or change antimicrobial treatment

Stop Change
Improved after careful clinical, radiological Not improved after careful clinical,
and microbiological evaluation radiological and microbiological evaluation
Fig. 17.2 Flow diagram for treating an infection
17.9 Treatment
17.9.1 Neonatal Intensive Care Unit
Early-onset infections in the NICU may be due to L. monocytogenes and/or

S. agalactiae. Ampicillin is the most active antibiotic against these two microor-
ganisms, which are acquired from the mother, and is combined with gentamicin to
cover AGNB and S. aureus. Late-onset infections are treated in the manner
described in Fig. 17.2.
17.9.2 Paediatric Intensive Care Unit
When a child is admitted to the PICU with a severe infection, a decision must be
made as to which antimicrobial will be used. Antimicrobials are used in combi-
nation depending on the severity of the illness; our experience over 20 years led us
to choose cefotaxime and gentamicin. This choice is empirical due to the absence
of any knowledge regarding the causative microorganism, though reasonable
assumptions can be made depending of the childs presentation. For example, a
child with meningococcal disease requires cefotaxime only; metronidazole can be
added in case of presumed anaerobic involvement. When a presumptive identifi-
cation of the microorganism can be made, the physician can then tailor therapy.
Cefotaxime/gentamicin can be replaced by cephradine for an infection due to
S. pneumoniae, S. pyogenes and S. aureus. When P. aeruginosa is isolated, ceft-
azidime should replace cefotaxime and the gentamicin continued. Yeast infections
require liposomal amphotericin B in place of cefotaxime/gentamicin (Fig. 17.2).
The efficacy of the antimicrobial treatment can be monitored using C-reactive
protein (CRP) levels in addition to the clinical, radiographic and microbiological
variables. Providing the antimicrobials used are correct, the child will improve
within 3 days; in our experience, a short, 5-day course of intravenously admin-
istered antibiotics is as effective as a course of 2 weeks or more (Chap. 12). After
5 days, the child is monitored for signs of infection; when there are no signs of
infection, antibiotics are discontinued. Should there be no improvement after
5 days, a change in antibiotic regimen is necessary.
Metronidazole is given for 3 days only. The antifungal agent, liposomal
amphotericin B, is given for 3 weeks and may be discontinued once the CRP level
is normal. Systemic antimicrobials are combined with enterally administered SDD
agents to guarantee prevention of potential pathogens becoming resistant to the
systemic agents.
References
1. Damjanovic V, van Saene HKF (1998) Outbreaks of infection in a neonatal intensive care
unit. In: van Saene HKF, Silvestri L, de la Cal MA (eds) Infection control in the intensive
care unit. Springer, Milan, pp 237248
2. Isaacs D, Moxon ER (1991) Neonatal infections. Butterworth-Heinemann, Oxford
3. van Saene HKF, Leonard EM, Shears P (1989) Ecological impact of antibiotics in neonatal
units. Lancet II:509510
4. de Man P, Verhoeven BAN, Verburgh HA et al (2000) An antibiotic policy to prevent
5. Murphy N, Damjanovic V, Hart CA et al (1986) Infection and colonisation of neonates by
Hansenula anomala. Lancet I:291293
6. Silvestri L, Sargison RE, Hughes J et al (2002) Most nosocomial pneumonias are not due to
nosocomial bacteria in ventilated patients: evaluation of the accuracy of the 48 h time cut-off
using carriage as the gold standard. Anaesth Intensive Care 30:275282
7. Petros AJ, OConnell M, Roberts C et al (2001) Systemic antibiotics fail to clear multidrug-
resistant Klebsiella from a paediatric ICU. Chest 119:862866
8. Goldmann DA, Leclair J, Macone A (1978) Bacterial colonization of neonates admitted to an
intensive care environment. J Pediatr 2:288293
a prospective three year study based on knowledge of the carrier state. Crit Care Med
32(3):839847
10. Baltimore RS (1998) Neonatal nosocomial infections. Semin Perinatol 22:2532
11. Beck-Sague CM, Azimi P, Fonseca SN et al (1994) Bloodstream infections in neonatal
intensive care unit patients: results of a multicenter study. Pediatr Infect Dis J 13:11101116
12. Gaynes RP, Martone WJ, Culver DH et al (1991) Comparison of rates of nosocomial infections
in neonatal intensive care units in the United States. Am J Med 91(suppl 3B):192196
13. La Gamma EF, Drusin LM, Mackles AW et al (1983) Neonatal infections: an important
determinant of late NICU mortality in infants less than 1,000 g at birth. Am J Dis Child
137:838841
14. Richards MJ, Edwards JR, Culver DH, Gaynes RP, The National Nosocomial Infections
Surveillance System (1999) Nosocomial infection in pediatric intensive care units in the
United States. Pediatrics 103(4):e39
15. Gray J, Gossain S, Morris K (2001) Three-year survey of bacteraemia and fungemia in a
pediatric intensive care unit. Pediatr Infect Dis J 20:416421
16. Pierce CM, Wade A, Mok Q (2000) Heparin-bonded central venous lines reduce thrombotic
and infective complications in critically ill children. Intensive Care Med 26:967972
17. Patel JC, Mollitt DL, Pieper P, Tepas JJ III (2000) Nosocomial pneumonia in the pediatric
trauma patient: a single centers experience. Critical Care Med 28:35303533
18. Leonard EM, van Saene HKF, Stoutenbeek CP et al (1990) An intrinsic pathogenicity index
for micro-organisms causing infections in a neonatal surgical unit. Microb Ecol Health Dis
2:151157
19. van Saene HKF, Taylor N, Donnell SC et al (2003) Gut overgrowth with abnormal flora: the
missing link in parenteral nutrition-related sepsis in surgical neonates. Eur J Clin Nutrit
57:548553
20. Donnell SC, Taylor N, van Saene HKF et al (2002) Infection rates in surgical neonates and
infants receiving parenteral nutrition: a five year prospective study. J Hosp Infect 52:273280
21. Morar P, Singh V, Makura Z et al (2002) Differing pathways of lower airway colonization
and infection according to mode of ventilation (endotracheal vs. tracheostomy). Arch
Otolaryngol Head Neck Surg 128:10611066
22. Singh-Naz N, Sprague BM, Patel K et al (2000) Risk assessment and standardized
nosocomial infection rate in critically ill children. Crit Care Med 28:20692075
23. Mahieu LM, De Muynck AO, De Dooy JJ et al (2000) Prediction of nosocomial sepsis in
neonates by means of a computer-weighted bed side scoring system (NOSEP score). Crit
Care Med 28:20262033
24. ACourt CH, Garrard CS, Crook D et al (1993) Microbiological lung surveillance in
culture. QJM 86:635638
25. Petros AJ, Marshall JC, van Saene HK (1995) Should morbidity replace mortality as an
endpoint for critical trials in intensive care? Lancet 345:369371
26. Kaiser AB (1986) Antimicrobial prophylaxis in surgery. New Engl J Med 315:11291138
27. Infection in Neurosurgery Working Party of the British Society for Antimicrobial
Chemotherapy (1994) Antimicrobial prophylaxis in neurosurgery and after head injury.
Lancet 344:15471551
28. Zobel G, Kuttnig GHM et al (1991) Reduction of colonization and infection rate during
pediatric intensive care by selective decontamination of the digestive tract. Crit Care Med
19:12421246
transplants. Transplantation 55:13061309
30. Ruza F, Alvarado F, Herruzo R et al (1998) Prevention of nosocomial infection in a pediatric
intensive care unit (PICU) through the use of selective decontamination. Eur J Epidemiol
14:719727
in severely burned pediatric patients. Burns 27:439445
32. Alder Hey Therapeutic Guideline Index.http://www.nppg.scot.nhs.uk/protocols/therapeutic_
guidelines_index.htm
33. DAmico R, Pifferi S, Leonetti C et al (1998) Effectiveness of antibiotic prophylaxis in
critically ill adult patients: systematic review of randomized controlled trials. BMJ 316:
12751285
patients: a systematic review of the evidence. Arch Surg 134:7076
35. De Jonge E, Schultz MJ, Spanjaard L et al (2002) Effects of selective decontamination of the
digestive tract on mortality and antibiotic resistance. Intensive Care Med 28(Suppl 1):S12
nystatin for the control of a Candida outbreak in the neonatal intensive care unit. J Hosp
Infect 24:245259
Early Adequate Antibiotic Therapy
18
R. Reina and M. A. de la Cal
18.1 Introduction
The most widely accepted definition of inappropriate antibiotic treatment in the

intensive care unit (ICU) encompasses the use of an agent to which the pathogen
is resistant [1] or delay in prescribing the antibiotic to which the pathogen is
sensitive. Some attention has been given to the fact that failure of anti-infective
therapy in the ICU might also occur due to inappropriate dosing, potentially
leading to suboptimal exposure to the broad-spectrum antimicrobial agent at the
infection site, even if it is administered in a timely manner [2].
The pattern of in vitro antibiotic sensitivity/resistance has been universally
accepted as the main component of antibiotic therapy appropriateness. The delay
between initial symptoms and antimicrobial administration is still a matter of
discussion and might vary according to the study and symptom severity of the
included patients. Kollef et al. [3] defined inadequate antimicrobial treatment as
microbiological documentation of infection that was not effectively treated at the
time the causative microorganism and its antibiotic susceptibility were known,
without paying attention to the time delay between the first symptoms of sepsis and
the microbiological results. Harbarth et al. [4], in a population of septic patients,
considered their treatment as inappropriate because the antimicrobial agent was
not administered within 24 h of primary microbial isolation from blood or a
remote infection site. Kumar et al. [5], in septic shock patients, considered therapy
to be effective when antimicrobials with in vitro activity appropriate for the iso-
lated pathogen or pathogens was received after the onset of recurrent or persistent
M. A. de la Cal (&)
Hospital Universitario de Getafe, Getafe, Spain
306 R. Reina and M. A. de la Cal
hypotension or initiated within 6 h of administration of the first new antimicrobial

agent.
18.2 Impact on Morbidity and Mortality
Appropriate antibiotic treatment was consistently associated with reduced

mortality rates in infected ICU patients, and this benefit was higher in patients with
severe sepsis and septic shock [3, 4]. Garnacho et al. [6] conducted a retrospective
case-control study of 87 ICU patients with sepsis adjusted by origin of sepsis,
inflammatory response at admission, surgical or medical status, hospital- or
community-acquired sepsis, Acute Physiology and Chronic Health Evaluation
(APACHE) II score (+2 points) and age (+10 years). Therapy was considered
inadequate when no effective drug against the isolated pathogen(s) was included in
the empirical antibiotic treatment within the first 24 h of admission to the ICU or
the doses and pattern of administration were not in accordance with current
medical standards. ICU mortality rates in patients with inadequate treatment were
67.8% [95% confidence interval (CI) 58.077.6] versus 28.7% (95% CI 19.238.2)
in patients receiving appropriate treatment. ICU length of stay in surviving patients
was 11 (719) and 7 (619) days, respectively. Kumar et al. [5], in a retrospective
study of 2,731 patients with septic shock, found a strong relationship between
delay in effective antimicrobial initiation and in-hospital mortality rates [adjusted
odds ratio (OR) 1.12 per hour delay, 95% CI 1.1031.136]. This increased
mortality rate per hour of delay was not associated with the type of infection. A
clinical conclusion of the study was that in patients with septic shock, the adequate
treatment should be initiated as soon as possible after taking samples for micro-
biological cultures when indicated. Barochia et al. [7] performed a systematic
review of eight unblinded trials, one randomized trial, and seven with historical
controls to assess the efficacy of all the recommended practices in treating patients
with sepsis [8]. They concluded that for all studies reporting the impact in mor-
tality rates, the only intervention consistently associated to a reduced rate is the
appropriateness of antibiotics. Five trials [913] provided data of the association
between actual mean time from presentation and antibiotic administration. All but
one [9] showed a survival benefit of early antibiotic administration.
18.3 Empirical Treatment: Monotherapy

Versus Combined Therapy
The rationale for appropriate antibiotic therapy is to increase the likelihood that the
infective pathogen will be susceptible in vivo to the prescribed antimicrobial. It
implies the in vitro susceptibility to the antibiotic, delay between symptoms and
antibiotic administration, dose, and pharmacodynamics (PD) and pharmacokinet-
ics (PK) of antimicrobials. According to these premises, prescription of the ade-
quate antibiotic must taken into account:
18 Early Adequate Antibiotic Therapy 307
type of infection: community-acquired pneumonia, ventilator-associated

pneumonia, abdominal sepsis, catheter-related bloodstream infection, soft-
tissue infection, etc;
antibiotic sensitivity/resistance of ICU flora;
patient comorbidity: asplenia, renal failure, neutropenia, immunosuppression,
allergy, obesity,etc.
Adequate empirical treatment is crucial to reduce mortality risk in severely ill
patients. Theoretically, this objective might be better achieved with combined therapy
instead of with monotherapy, but this subject is still under discussion [14, 15].
The effect of monotherapy (meropenem) versus combined therapy (meropenem
plus ciprofloxacin) has been assessed in a randomized clinical trial of 740
mechanically ventilated patients who developed suspected ventilator-associated
pneumonia (VAP) [16] Patients who were known to be previously infected or
colonized with Pseudomonas spp. or methicillin-resistant Staphylococcus aureus
(MRSA) were excluded from the trial. The median time from suspicion of VAP to
initiation of study antibiotics was 4 h. The proportion of patients who received
adequate initial antibiotics was significantly greater in the combination group than
in the monotherapy group (93.1 vs. 85.1%, p = 0.01). Reasons for inadequate
initial therapy were related to the presence of Pseudomonas or Acinetobacter spp.,
Stenotrophomonas maltophilia, other multi-drug-resistant Gram-negative bacteria,
and MRSA. There were no differences between groups in the main outcomes of
mortality, time from randomization to discontinuation of mechanical ventilation
alive, and discharge from ICU alive. These results support the use of monotherapy
as empirical treatment in patients with suspected pneumonia if local resistance
patterns or individual patient risk factors do not suggest the possibility of multi-
drug-resistant organisms or other difficult-to-treat organisms.
Garnacho et al. [17] retrospectively estimated the association of monotherapy
versus combined therapy with mortality in 183 episodes of VAP caused by
Pseudomonas aeruginosa. When P. aeruginosa was susceptible to the antimicro-
bial used in monotherapy, there was no difference in mortality rates compared with
the group treated with combined therapy. Monotherapy versus combined therapy
has been also compared in patients with sepsis and septic shock. In a systematic
review, Safdar et al. [18] estimated the effect of both types of therapy in Gram-
negative bacteremia. Seventeen studies were included (five prospective cohort
studies, two prospective randomized trials, and ten retrospective cohort studies).
There was no difference in mortality rates between monotherapy and combined
therapy (OR 0.96; 95% CI 0.701.32). The same conclusion was reached in a
subgroup analysis [19] of four studies (three retrospective cohort studies, one
prospective cohort study) of patients with P. aeruginosa bacteremia (OR 1.31; 95%
CI 0.622.79), but the low design quality of the studies and the high heterogeneity
of their results (42%) preclude any strong conclusion. The addition of aminogly-
coside to beta-lactams increases the risk of nephrotoxicity [20].
Kumar et al. [21], in a subgroup analysis of observational studies and
randomized clinical trials of bacterial infections potentially associated with sepsis
or septic shock, found that when the monotherapy groups have a mortality rate
[25%, there is a potential benefit of combined therapy (OR 0.49; 95% CI

0.350.70). The authors did not provide data of the main confounding factor:
appropriateness of the initial antimicrobial treatment.
In conclusion, the best evidence supports that in nonneutropenic critically ill
patients, combined therapy is only justified as empirical treatment when it is
suspected to be caused by highly resistant microorganisms, such as resistant
Pseudomonas or Acinetobacter spp., other multi-drug-resistant Gram-negative
bacteria, and MRSA. Most severe patients, i.e., septic shock patients, may benefit
from broad coverage with combined therapy to provide a higher likelihood of
adequate treatment.
18.4 De-escalation
Initial empiric therapy can be appropriate in 8090% patients if it is selected

following the recommendations provided by guidelines supported by most
authoritative scientific societies. Following initial empiric therapy, de-escalation
means using microbiologic and clinical data to change from an initial broad-
spectrum or multidrug empiric therapy regimen to a regimen with fewer antibiotics
and agents of narrower spectrum, reduced therapy duration, and stopping treatment
as dictated by results of microbiological cultures and clinical response [22]. The
rationale of de-escalation is to reduce the use of antimicrobials to reduce costs,
adverse effects, and antibiotic resistance. It is included as a component of the
stewardship to optimize antibiotic use in hospitals [23].
De-escalation strategies have been mostly evaluated in VAP. Alvarez-Lerma
et al. [24] performed one prospective multicenter study to evaluate the results of
implementing a protocol of nosocomial pneumonia therapy that recommended
de-escalation. They enrolled 258 patients; appropriate treatment was prescribed
in 89% of cases. In 113 of them, microbiological cultures of respiratory samples
were negative, and the empirical treatment was continued unmodified. In 108
patients with adequate treatment, at least one pathogen was identified. In only 56
of those patients (24%) was therapy de-escalated. The results emphasize two
conclusions: (1) different options to treat patients with suspected nosocomial
pneumonia with negative microbiological cultures should be tested in random-
ized clinical trials; (2) implementation of de-escalation in ICU may be increased.
Similar de-escalation percentages have been reported in another observational
study [25].
Singh et al. [26] performed a randomized clinical trial in 81 critically ill patients
with suspected nosocomial pneumonia and low clinical pulmonary infection score
(B6) to evaluate the outcomes of early discontinuation of empirical treatment
(ciprofloxacin) when the score remained B6 versus standard therapy. Results of a
comparison between a short course of ciprofloxacin and standard therapy showed
mortality rate 0 versus 7% (not significant); length of ICU stay 9 versus 14 days
(p = 0.04); antimicrobial resistance 14 versus 38% (p B 0.017) and total costs of
the episode UD $6,482 versus $16,004, respectively.
The usefulness of biomarkers to guide treatment has been estimated in clinical

trials. Christ-Crain et al. [27] performed a cluster-randomized clinical trial in 243
patients (87 with pneumonia) with suspected lower respiratory tract infections
admitted to the emergency department. Patients were randomly assigned either to
standard antimicrobial therapy or procalcitonin-guided antimicrobial treatment.
On the basis of serum procalcitonin concentrations, use of antibiotics was more
or less discouraged (\0.1 lg/L or \0.25 lg/L) or encouraged (C0.5 lg/L or
C0.25 lg/L), respectively. Duration of antibiotic treatment was shorter in the
procalcitonin group (11 vs. 13 days; p = 0.03) as well as antibiotic costs of
US $96 versus US $202 per patient (p \ 0.0001).These results were reproduced by
the same group in patients with suspected community-acquired pneumonia [28].
Stolz et al. [29], in a multicenter randomized clinical trial, assigned 101 patients
with suspected VAP to a therapy according to the American Thoracic Society
guidelines or a protocol guided by serum procalcitonin levels during the first
2 days of treatment. Procalcitonin level at day 2:
\0.25 lg/L discontinuation of antibiotics was strongly encouraged;
between 0.25 and 0.5 lg/L or a decrease by C80% compared with day 0
discontinuation of antibiotics was encouraged;
C0.5 lg/L or decrease by \80% compared with day 0 discontinuation of
antibiotics was discouraged;
[1 lg/L antibiotic discontinuation was strongly discouraged.
Overall duration of antibiotic therapy was reduced from 15 to 10 days (p = 0.04).
The number of mechanical ventilation-free days alive, ICU-free days alive, length of
hospital stay, and mortality rate on day 28 for the two groups were similar.
Nobre et al. [30] tested the hypothesis that an algorithm based on serial mea-
surements of procalcitonin allows reduction in antibiotic therapy duration in
patients with severe sepsis or septic shock. Patients in the procalcitonin-guided
group were re-evaluated following a protocol in which: (1) patients with baseline
levels C1 lg/L were re-evaluated at day 5; if procalcitonin dropped [90% from
baseline or \0.25 lg/L was reached, investigators encourage to discontinue
antibiotics; (2) patients with baseline levels \1 lg/L were re-evaluated at day 3;
treating physicians were encouraged to discontinue antibiotics when the procalc-
itonin level was \0.1 lg/L. Duration was 3.5 days shorter in the procalcitonin
group (6 vs. 9.5 days; p = 0.15). In summary guiding treatment using biomarkers
is safe but only provides marginal benefits.
18.5 Antibiotics Dosage in Critically Ill Patients

with Severe Sepsis and Septic Shock:
Pharmacokinetics/Pharmacodynamics Parameters
Prescribing antibiotics, including empirical antibiotics, for critically ill patients is a

complex process that requires ongoing patient health evaluation to account for the
dynamic sepsis disease process. Pathophysiological changes, such as organ dys-
function, fluid shifts, and altered immune status, are common and are able to
Table 18.1 Time- and concentration-dependent antibiotics
Time-dependent killing Concentration-dependent killing

Penicillins
Cephalosporins Aminoglycosides
Macrolides Fluoroquinolones
Carbapenems Amphotericin B
Clindamycin Metronidazole
Linezolid Colistine
Vancomycin
Tigecycline
reduce the efficacy of anti-infective treatment. In recent years, more attention is

being paid to how critical illness can influence the PK/PD parameters of antimi-
crobials by altering their volume of distribution, rate of excretion and elimination,
and impairing penetration into tissues [2, 31, 32], and how these changes may
modify the efficacy of prescribed antimicrobials.
PK describes absorption, distribution, metabolism, and elimination of drugs in
serum and at infection sites. PD describes the relationship that exists between the
drug concentration to which the bacteria is exposed at various sites of infection and
bacterial killing activity. For antibiotics, PD activity is integrated with PK param-
eters and pharmacologic effect with the minimum inhibitory concentration (MIC)
for a particular pathogen in order to evaluate the ability of the antibiotic to kill the
infective organism or inhibit its growth. According to these PK/PD parameters,
antimicrobial agents can be categorized based on their mode of bacterial killing and
the presence of a postantibiotic effect. Thus, the pattern of bacterial killing can be
concentration- or time dependent [33] (Tables 18.1 and 18.2). Antibiotics are
considered concentration dependent if higher concentrations result in more exten-
sive elimination of the pathogen. The kill rate for these antibiotics is closely related
to peak or maximum concentration above breakpoint MIC (Cmax/MIC). An antibi-
otic is considered time dependent if its bacterial killing effectiveness depends upon
the duration of pathogen exposure to the agent. The kill rate is better related to how
long concentrations are sustained above breakpoint MIC.
Kasiakou et al. [34] performed a systematic review of randomized clinical trials
to evaluate the comparative PK/PD properties of continuous versus intermittent
intravenous administration of various antimicrobials. They concluded that from a
PD point of view, data suggest that continuous intravenous infusion with time-
dependent bacteria-killing activity (b-lactams or vancomycin) seems to be superior
to the intermittent intravenous administration, without differences in mortality or
nephrotoxicity rates.
Few studies have assessed the relationship between PK/PD data and
clinical outcome in humans. Roberts et al. [35] performed a systematic review of
randomized clinical trials comparing the efficacy of continuous versus bolus
Table 18.2 Antibiotics, their killing activities, and pharmacokinetic and pharmacodynamic
(PK/PD) parameters
Antibiotic Killing activity PK/PD parameters

Aminoglycosides Concentration dependent Cmax/MIC
Metronidazole Concentration dependent AUC024/MIC; Cmax/MIC
Fluoroquinolones Concentration dependent AUC024/MIC; Cmax/MIC
Cholistine Concentration dependent AUC024/MIC; Cmax/MIC
Penicillins Time dependent T [ MIC
Cephalosporins Time dependent T [ MIC
Carbapenems Time dependent T [ MIC
Monobactams Time dependent T [ MIC
Clindamycin Time dependent T [ MIC
Vancomycin Time dependent AUC024/MIC; Cmax/MIC
Macrolides Time dependent AUC024/MIC; Cmax/MIC
Linezolid Time dependent AUC024/MIC; Cmax/MIC
Tigecycline Time dependent AUC024/MIC; Cmax/MIC
AUC area under curve, MIC minimum inhibitory concentration, Cmax maximal concentration, T time
administration of b-lactams in 846 patients (14 randomized clinical trials),

including a large proportion of critically ill patients. Continuous infusion of a
b-lactam antibiotic was consistently not associated with an improved clinical cure
(OR 1.04; 95% CI 0.741.46) or mortality (OR 1.00, 95% CI 0.482.06) rates.
The rationale for continuous infusion of b-lactams has been documented from a
PD point of view, but new studies should be performeddespite the limitation of
requiring a large number of patients to achieve significant resultsbefore making
a recommendation about using this PK/PD approach.
18.6 Conclusions
Appropriate early antibiotic treatment for ICU patients with infection is associated
with strong reduction in mortality and morbidity rates. At present, the challenge is
to achieve a level of adequate empirical therapy C90%. Combined therapy is only
justified as empirical treatment when the infection is suspected to be caused by
highly resistant microorganisms or in the most severely ill patients to provide a
higher likelihood of adequate treatment. De-escalation of the initial antibiotic
therapy must be encouraged because is saves money and reduces costs. The role
of biomarkers in helping to discontinue antibiotic therapy is usually marginal.
The PK/PD approach to prescribing the correct dose and route of administration is
a potential field of outcome research in humans and should be included in the
differential diagnosis of treatment failure.
References
1. Davey PG, Marwick C (2008) Appropriate vs. inappropriate antimicrobial therapy. Clin
Microbiol Infect 14(Suppl 3):1521
2. Pea F, Viale P (2009) Bench-to-bedside review: appropriate antibiotic therapy in severe
sepsis and septic shockdoes the dose matter? Crit Care 13:214
3. Kollef MH, Sherman G, Ward S, Fraser VJ (1999) Inadequate antimicrobial treatment of
infections: a risk factor for hospital mortality among critically ill patients. Chest 115:462474
4. Harbarth S, Garbino J, Pugin J et al (2003) Inappropriate initial antimicrobial therapy and its
effect on survival in a clinical trial of immunomodulating therapy for severe sepsis. Am J
Med 115:529535
5. Kumar A, Roberts D, Wood KE et al (2006) Duration of hypotension before initiation of
effective antimicrobial therapy is the critical determinant of survival in human septic shock.
6. Garnacho-Montero J, Ortiz-Leyba C, Herrera-Melero I et al (2008) Mortality and morbidity
attributable to inadequate empirical antimicrobial therapy in patients admitted to the ICU
with sepsis: a matched cohort study. J Antimicrob Chemother 61:436441
7. Barochia AV, Cui X, Vitberg D et al (2010) Bundled care for septic shock: an analysis of
clinical trials. Crit Care Med 38:668678
8. Dellinger RP, Levy MM, Carlet JM et al for the International Surviving Sepsis Campaign
Guidelines Committee; American Association of Critical-Care Nurses; American College of
Chest Physicians; American College of Emergency Physicians; Canadian Critical Care Society;
European Society of Clinical Microbiology and Infectious Diseases; European Society of
Intensive Care Medicine; European Respiratory Society; International Sepsis Forum; Japanese
Association for Acute Medicine; Japanese Society of Intensive Care Medicine; Society of
Critical Care Medicine; Society of Hospital Medicine; Surgical Infection Society; World
Federation of Societies of Intensive and Critical Care Medicine (2008) Surviving sepsis
campaign: international guidelines for management of severe sepsis and septic shock: 2008. Crit
Care Med 36(1):296327. Erratum in: Crit Care Med 36:13941396
9. Rivers E, Nguyen B, Havstad S, Early Goal-Directed Therapy Collaborative Group et al
(2001) Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl
J Med 345:13681377
10. Shapiro NI, Howell MD, Talmor D et al (2006) Implementation and outcomes of the Multiple
Urgent Sepsis Therapies (MUST) protocol. Crit Care Med 34:10251032
11. Micek ST, Roubinian N, Heuring T et al (2006) Beforeafter study of a standardized hospital
order set for the management of severe septic shock. Crit Care Med 34:27072713
12. Nguyen HB, Corbett SW, Steele R et al (2007) Implementation of a bundle of quality
indicators for the early management of severe sepsis and septic shock is associated with
decreased mortality. Crit Care Med 35:11051112
13. El Solh AA, Akinnusi ME, Alsawalha LN, Pineda LA (2008) Outcome of septic shock in
older adults after implementation of the sepsis bundle. J Am Geriatr Soc 56:272278
14. Cunha BA (2008) Sepsis and septic shock: selection of empiric antimicrobial therapy. Crit
Care Clin 24:313334
15. Abad CL, Kumar A, Safdar N (2011) Antimicrobial therapy of sepsis and septic shock
when are two drugs better than one? Crit Care Clin 27:e1e27
16. Heyland DK, Dodek P, Muscedere J, Canadian Critical Care Trials Group et al (2008)
Randomized trial of combination versus monotherapy for the empiric treatment of suspected
ventilator-associated pneumonia. Crit Care Med 36:737744
17. Garnacho-Montero J, Sa-Borges M, Sole-Violan J et al (2007) Optimal management therapy
for Pseudomonas aeruginosa ventilator-associated pneumonia: an observational, multicenter
study comparing monotherapy with combination antibiotic therapy. Crit Care Med 35:
18881895
18. Safdar N, Handelsman J, Maki DG (2004) Does combination antimicrobial therapy reduce
mortality in Gram-negative bacteraemia? A meta-analysis. Lancet Infect Dis 4:519527
19. Paul M, Leibovici L (2005) Combination antibiotic therapy for Pseudomonas aeruginosa
bacteraemia. Lancet Infect Dis 5:192193
20. Paul M, Benuri-Silbiger I, Soares-Weiser K, Leibovici L (2004) b-lactam monotherapy
versus b-lactam-aminoglycoside combination therapy for sepsis in immunocompetent
patients: systematic review and meta-analysis of randomised trials. BMJ 328:668. Erratum
in BMJ 328:884
21. Kumar A, Safdar N, Kethireddy S, Chateau D (2010) A survival benefit of combination
antibiotic therapy for serious infections associated with sepsis and septic shock is contingent
only on the risk of death: a meta-analytic/meta-regression study. Crit Care Med 38:
16511664
22. Niederman MS (2006) De-escalation therapy in ventilator-associated pneumonia. Curr Opin
Crit Care 12:452457
23. Dellit TH, Owens RC, McGowan JE et al (2007) Infectious Diseases Society of America and
the Society for Healthcare Epidemiology of America guidelines for developing an
institutional program to enhance antimicrobial stewardship. Clin Infect Dis 44:159177
24. Alvarez-Lerma F, Alvarez B, Luque P et al (2006) ADANN study group. Empiric broad-
spectrum antibiotic therapy of nosocomial pneumonia in the intensive care unit: a prospective
observational study. Crit Care 10:R78
25. Kollef MH, Morrow LE, Niederman MS et al (2006) Clinical characteristics and treatment
patterns among patients with ventilator-associated pneumonia. Chest 129:12101218.
Erratum in Chest 130:138
patients with pulmonary infiltrates in the intensive care unit. A proposed solution for
indiscriminate antibiotic prescription. Am J Respir Crit Care Med 162:505511
27. Christ-Crain M, Jaccard-Stolz D, Bingisser R et al (2004) Effect of procalcitonin-guided
treatment on antibiotic use and outcome in lower respiratory tract infections: cluster-
randomised, single-blinded intervention trial. Lancet 363:600607
28. Christ-Crain M, Stolz D, Bingisser R et al (2006) Procalcitonin guidance of antibiotic therapy in
community-acquired pneumonia: a randomized trial. Am J Respir Crit Care Med 174:8493
29. Stolz D, Smyrnios N, Eggimann P et al (2009) Procalcitonin for reduced antibiotic exposure
in ventilator-associated pneumonia: a randomised study. Eur Respir J 34:13641375
30. Nobre V, Harbarth S, Graf JD et al (2008) Use of procalcitonin to shorten antibiotic treatment
duration in septic patients: a randomized trial. Am J Respir Crit Care Med 177(5):498505
31. Pea F, Viale P, Furlanut M (2005) Antimicrobial therapy in critically ill patients: a review of
pathophysiological conditions responsible for altered disposition and pharmacokinetic
variability. Clin Pharmacokinet 44:10091034
32. Roberts JA, Lipman J (2009) Dose adjustment and pharmacokinetics of antibiotics in severe
sepsis and septic shock. In: Rello J, Kollef M, Diaz E, Rodriguez A (eds) Infectious diseases
in critical care, 2nd edn. Springer-Verlag, Berlin-Heidelberg, pp 122146
33. Petrosillo N, Drapeau CM, Agrafiotis M, Falagas ME (2010) Some current issues in the
pharmacokinetics/pharmacodynamics of antimicrobials in intensive care. Minerva Anestesiol
76(7):509524
34. Kasiakou SK, Lawrence KR, Choulis N, Falagas ME (2005) Continuous versus intermittent
intravenous administration of antibacterials with time-dependent action: a systematic review
of pharmacokinetic and pharmacodynamic parameters. Drugs 65:24992511
35. Roberts JA, Webb S, Paterson D et al (2009) A systematic review on clinical benefits of
continuous administration of beta-lactam antibiotics. Crit Care Med 37:20712078
ICU Patients Following Transplantation
19
A. Martinez-Pellus and I. Cortes Puch
19.1 Introduction
Solid organ and bone marrow transplant recipients are at high risk of infection due to
long-term immunosuppressant therapy that is necessary to help prevent transplant
rejection. Infection prevention and control in this patient population requires a
multifactorial approach using general and pharmacologic measures. In this chapter, we
discuss selective decontamination of the digestive tract (SDD) (polymyxin, tobra-
mycin, and amphotericin B) as a prophylactic measure during patient admission to the
intensive care unit (ICU); the protective effect of fluconazole and topical antifungals
drugs; systematic use of trimethoprim/sulfamethoxazole to manage Pneumocystis
jirovecii; and ganciclovir as a valid strategy to prevent cytomegalovirus (CMV)
infection. Another measure considered is shortening postoperative ICU length
of stay by setting quick weaning protocols, using noninvasive mechanical ventilation,
initiating physiotherapy, and early catheter removal. Long-term maneuvers strongly
depend on patient adherence to a prophylactic regimen over long periods, and
educational programs and involving patients in their prognosis help achieve
this goal.
19.2 Solid Organ Transplantation
Solid organ transplantation is a widespread procedure that has become the therapy
of choice for patients with irreversible and progressive end-stage organ disease.
Potential severe postoperative complications (related to the inflammatory response
A. Martinez-Pellus (&)
Intensive Care Unit, University Hospital Virgen de la Arrixaca,
Murcia, Spain
316 A. Martinez-Pellus and I. Cortes Puch
due to ischemia/reperfusion injury of the graft) and the risk of acute organ
rejection require close postoperative patient monitoring. This surveillance usually
takes place in the intensive care unit (ICU). In the early post-transplant period,
patients will frequently require mechanical ventilation and invasive devices,
putting them at risk for a large variety of infections. Infections at this stage are
frequent [1] and have significant consequences on the prognosis, exceeding the
surgical complications as a cause of mortality [2]. The primary goal in the
transplant recipient is to avoid organ rejection. This requires achieving an adequate
state of immunosuppression, known as net state of immunosuppression [3], which
refers to all factors that contribute to the patients risk of infection. Its major
determinants are immunosuppressive treatment dose, regimen, and duration. This
state can also be affected by the presence of neutropenia, comorbidities, and
concomitant infections with immunomodulating viruses, particularly cytomega-
lovirus (CMV), Epstein-Barr virus (EBV), and hepatitis B and C viruses. When
some of these factors occur in a single patient, excessive immunosuppression takes
place, affecting humoral and cellular mechanisms and exposing the patient to
infection. On the other hand, the environmental exposure of the transplant reci-
pient to a large variety of potentially infectious agents (viruses, fungi, community-
acquired pathogens, endogenous flora, etc.) is continuous and takes place in the
community as well as in the hospital. Other pathogens that scarcely provoke
community-acquired infections in the immunocompetent host can lead to devas-
tating syndromes in the transplant recipient. This is the case with respiratory
viruses, fungi, and mycobacteria. Hence, the application of preventive strategies
for long periods is necessary. If infection takes places despite these preventive
measures, it will represent a serious problem, leading to hospitalization, complex
diagnostic procedures, and empirical broad-spectrum antibiotic treatment. Current
immunosuppressive regimens have been reached by relative consensus. They
always include steroids, together with cyclosporine or tacrolimus, and myco-
phenolate or azathioprine, in different combinations. There is evidence that links
all of these drugs with several opportunistic pathogens, such as P. jirovecii and
Aspergillus (steroids), reactivation and replication of latent viruses (cyclosporine
and tacrolimus), and CMV and bacterial infections (mycophenolate). The risk of
infection after transplantation changes over time, especially with modifications in
immunosuppression. The maximal immunosuppressive effect of these regimens
takes place after several months of treatment. Therefore, we can establish a pre-
dictable timeline of the specific infections after transplantation according to
the level of immunosuppression achieved (Table 19.1). Exceptions to this timeline
are so rare that they suggest either a massive environmental exposure or excessive
immunosuppression. This predictable pattern in which infections occur allows
the establishment of specific prophylactic measures against them. These pre-
ventive strategies must start before surgery and continue during prolonged periods
or even for life.
19 ICU Patients Following Transplantation 317
Table 19.1 Timing of infectious complications and preventive measures after transplantation
Time Infectious risk Screening Preventive measures

Pre-transplant period Previous TBC Tuberculin None
test
CMV, EBV, HVZ Serology Vaccination (HBV; HVZ)
Toxoplasma Serology None
Bacterias None Vaccination
(pneumococcus; HI)
Aspergillusa Cultures Amphotericin B aerosol
Active infection Cultures Eradicate
Early post-transplant Surgical infection Cultures Surgical prophylaxis
period
VAP Cultures/RX SDD (LT); NIV
Multi-drug-resistant Cultures Antibiotic policies
pathogens
Fungal infection Cultures/ SDD (LT); Fluconazole
serology
TBC reactivation RX/Ziehl Isoniazid
Long ICU stay Early discharge
Late post-transplant CMV Serology/PCR Ganciclovir
period
Fungal infection Cultivo/ Fluconazole; itraconazoleb
serology
Pneumocystisb BAL Trimethropim/
sulfamethoxazoleb
pneumococcus/H.I vaccination (12th and 24th
months)
TBC reactivation RX/Ziehl Treatment
Bold indicates measures strongly recommended
TBC tuberculosis, CMV cytomegalovirus, EBV Epstein-Barr virus, HVZ Herpes Varicella zoster,
HI Haemophilus influenzae, VAP ventilator-associated pneumonia, SDD selective decontamina-
tion of the digestive tract, LT liver transplant, NIV noninvasive ventilation, RX chest X-ray, BAL
bronchoalveolar lavage
a
Lung transplant
b
Cardiac transplant
19.2.1 Preoperative Strategies
Candidates for solid organ transplantation must undergo an exhaustive preopera-

tive evaluation in order to define adequate prophylactic measures. Individualized
epidemiologic exams can guide preventive strategies and must include unusual
pathogen exposition, childhood vaccinations, previous surgeries, antibiotic treat-
ments, recurrent infections, etc. A tuberculin skin test is necessary, as is serologic
testing for CMV, EBV, Toxoplasma gondii, and syphilis. If the tuberculin skin test is
positive, the measures that should be taken are controversial due to the potential
hepatotoxicity of the tuberculostatics and their interferences with the pharmacoki-
netics of tacrolimus and cyclosporine. Periodic negative clinical and radiologic
evaluations should rule out the need for prophylaxis. Vaccination against tetanus,
diphtheria, mumps, influenza, pneumococcus, Haemophilus influenza type B, and
hepatitis B virus must be performed during the pretransplant period if they were not
previously completed [4]. Live vaccines are generally contraindicated after trans-
plantation, as they may trigger an infection or facilitate organ rejection. Varicella-
zoster vaccination should also be performed if the patient is seronegative, although
its effectiveness is doubted [5]. The importance of these measures lies in the fact that
all of these infections may be latent in the patient and undergo reactivation during
immunosuppression. The transplant recipient must not have a proven or suspected
infection by the time of surgery. Respiratory carriage of pathogens such as
Aspergillus or Burkholderia cepacia must be ruled out or eradicated before surgery.
In the case of Aspergillus, some studies demonstrate a significant reduction in the rate
of infections in lung or cardiac transplant recipients who received amphotericin B
aerosols as prophylaxis [6]. Nevertheless, the detection of this fungus in a respiratory
sample may only indicate colonization and, therefore, the interpretation of this
microbiological data is difficult. In an attempt to establish the predictive value of
Aspergillus isolates, a study revealed that cultures with more than two colonies or
more than one site of infection were predictive of significant infection and portended
a poor prognosis and development of invasive disease [7].
19.2.2 Early Postoperative Period and Preventive Strategies
Surgical complexity and, particularly, the impact of cold ischemia times and
consequent graft reperfusion, as well as the risk of systemic complications, leads to
a close immediate postoperative management that usually takes place in the ICU.
The probability of ICU hospitalization is highest during the first month following
transplantation, when infectious complications are similar to those that occur in an
immunocompetent host undergoing surgery. They are comparable in site (surgical
wound infection, urinary tract infection, catheter-related bloodstream infections,
etc.) and etiology (typical nosocomial pathogens such as Staphylococcus aureus,
coagulase-negative staphylococcus, and Gram-negative bacilli, to which fungi are
progressively added). Prophylactic measures must be directed to these expected
microorganisms. However, we must be aware that patients with previous pro-
longed hospitalizations while waiting for transplantation may become colonized
with hospital-acquired antimicrobial-resistant organisms (Pseudomonas,
methicillin-resistant S. aureus (MRSA), Aspergillus, etc.).
19.2.2.1 Topical Antibiotic Prophylaxis

Preventive strategies based on selective decontamination of the digestive tract
(SDD), which include pharyngeal and gastric application of polymyxin,
tobramycin, and amphotericin B (PTA), have proved to be useful in preventing

respiratory tract infections in critical care patients [8]. The implementation of
these strategies in solid organ transplantation has yielded conflicting results. The
aim of selective gut decontamination in these patients is to reduce the bacterial
burden of the digestive tract, which is the major reservoir of potentially pathogenic
endogenous flora. The first open study in this field [9], performed in a cohort of
patients waiting for orthotopic liver transplantation, showed a significant decrease
of aerobic Gram-negative bacteria and Candida colonization following 3 days of
oral administration of polymyxin, gentamicin, and nystatin. This prophylactic
measure was continued for 21 days, and there was no documented episode of
infection caused by aerobic Gram-negative bacilli during the first month following
transplantation. Following discontinuation of SDD, recolonization of the gastro-
intestinal tract with aerobic Gram-negative bacteria and Candida occurred in 90
and 35% of the patients, respectively, within 5 days. More recently, four com-
parative and randomized studies analyzed the effect of a standard SDD regimen
(PTA ? systemic antibiotics). Smith et al. [10], in a pediatric population, com-
pared patients receiving standard prophylaxis with systemic antibiotics with
patients who received SDD added to the standard prophylaxis regimen. They
found a reduction of the incidence of infections from 50 to 11% (p \ 0.001). In a
randomized controlled trial, Bion et al. [11] reported a significant reduction in
respiratory tract infections in the SDD group but a similar incidence of endotox-
emia episodes and the development of organ system failures. Nevertheless, in that
study, SDD was started in the immediate postoperative period, and it has been
demonstrated that the eradication of aerobic Gram-negative bacilli (as a source of
endotoxemia) requires a minimum of 3 days of SDD [12]. Emre et al. [13], in a
retrospective cohort study, compared 212 consecutive orthotopic liver transplant
recipients receiving perioperative systemic antibiotics and SDD (polymyxin,
gentamicin, and nystatin administered through a nasogastric tube) for 21 days with
the next 157 consecutive patients, who only received systemic antibiotics. The
results yielded a significant reduction of infections in the SDD group, irrespective
of site, caused by either aerobic Gram-negative bacilli or by Gram-positive cocci.
However, in a cost-effectiveness study comparing SDD with placebo, van
Enckevort et al. [14] found no differences in the global number of infections, with
a higher cost in SDD group. Nevertheless, that study was performed in a small
sample (26/29 patients), and it does not describe the sites of infections or the
pathogens isolated. Furthermore, SDD was applied from the time the patients were
enrolled on the transplant waiting list (mean 133 days) until the 30th postoperative
day, with frequent surveillance samples, justifying the excessive cost found in the
SDD group. In another study with a similar approach, which evaluates the work
load and the additional costs related to the use of SDD, no difference was found in
these items. Actually, a reduction in the number of processed microbiological
samples to rule out infection (blood cultures, bronchial aspirates, exudates, and
serological tests) was found [15]. In a systematic literature review, Nathens et al.
[16] analyzed mortality rates and the most relevant infections (pneumonia, urinary
tract infection, surgical wound infection, bloodstream infections) in surgical
patients. The results showed that SDD significantly reduced the incidence of
infection in transplant organ recipients [odds ratio (OR) 0.44; 95% confidence
interval (CI) 0.230.87]. Mortality rates were not significantly altered (OR 0.29;
95% CI 0.061.47). In a randomized placebo-controlled trial performed in a small
group of liver transplant patients, Zwaveling et al. [17] found a similar incidence
of infections in both groups but with a significant reduction in those caused by
aerobic Gram-negative bacilli (p \ 0.01) and Candida spp (p \ 0.05) in the SDD
group. In a recent meta-analysis [18] that reviewed four randomized trials com-
paring SDD versus placebo or no treatment at all, the overall incidence of infection
was similar despite the use of SDD (OR 0.88; CI 95%: 0.71.1). Nevertheless,
SDD was able to reduce the incidence of infections caused by aerobic Gram-
negative bacilli (OR 0.16; CI 95% 0.070.37) and the incidence of postoperative
pneumonias. SDD regimens should be adjusted, however, to the most prevalent
pathogens found in each ICU, bearing in mind their particular resistance profile.
The use of probiotics during the pretransplant period could represent an alter-
native to SDD. In a randomized trial, Rayes et al. [19] compared the administration
of a fiber-containing formula plus living Lactobacillus plantarum 299 versus SDD.
They observed a reduction in the incidence of bacterial infections from 48 to 13%,
with early enterally provided nutrition with the probiotics. These findings should
be validated in further studies, as this is the first study performed in this field, and
it is not powered high enough to draw solid conclusions.
19.2.2.2 Systemic Antibiotic Prophylaxis

Surgical prophylaxis is well standardized in the absence of previous colonization
or infection with problematic pathogens. A 24-h regimen of second-generation
cephalosporins should be adequate to prevent infection subsequent to the surgical
intervention. The use of vancomycin in surgical prophylaxis should be restricted
due to the risk of selecting vancomycin-resistant enterococcus.
In patients with risk factors for tuberculosis (latent tuberculosis, familiar
exposure, malnutrition, etc.), a 912-month course of isoniazid can be considered.
This course should start early after transplantation and must include close moni-
toring of liver enzymes levels.
19.2.2.3 Other Measures to Prevent Infection
An additional risk in the transplant recipient for postoperative infections is a pro-
longed stay in an area with a high environmental exposure to potential pathogens and
multiple instrumentation, such as the ICU. A study analyzing factors related to short-
term prognosis of liver transplant recipients [20] found that patients who stayed in the
ICU [3 days had a higher mortality rate (19.5 vs. 2%; p \ 0.001) and more infec-
tious episodes (41.5 vs. 11%; p \ 0.001) than those who stayed for\3 days (given a
similar age of recipient and donor, cold ischemia time, intraoperative bleeding, and
previous functional state). Another prospective and randomized trial [21] compared
noninvasive ventilation with standard treatment (supplemental oxygen administra-
tion) in 51 recipients of solid organ transplantation with acute hypoxemic respiratory
failure. The use of noninvasive ventilation was associated with improvement in the
partial pressure of oxygen in arterial blood/fractional inspiratory oxygen (PaO2/

FiO2) ratio in a larger number of patients (60 vs. 25%; p \ 0.03), as well as with a
significant reduction in the rate of endotracheal intubation (20 vs. 70%; p = 0.002),
length of ICU stay of survivors [mean standard deviation (SD) days 5.5 (3) vs. 9
(4); p = 0.03), and ICU mortality rate (20 vs. 50%; p = 0.05). Finally, simple
maneuvers such as early and continuous postural changes, in order to avoid atelec-
tasis and other problems related to prolonged immobilization, have proven to be
effective in decreasing the incidence of lower respiratory tract infections in the early
post-transplant period [22].
19.2.3 Late Postoperative and Preventive Maneuvers
Fungal infection is frequent in the transplant recipient, with variable incidence

according to the type of transplant. Functional disruption of the intestinal barrier and
bacterial translocation are involved in this process [23]. The database of the Spanish
Network of Infection in Transplantation (RESITRA), which includes 3,500 solid
organ transplant recipients during 20032005, estimates the incidence of fungal
infection in this population, showing an incidence of Candida spp. infection of
33.6% and Aspergillus spp. between 0.15% (renal transplant) and 5.9% (lung and
bone marrow transplants) [24]. These data are markedly less than those reported in
the literature two decades ago, with incidences varying between 2 and 40%,
depending on the transplanted organ [1]. This could be a reflection of a better
management of these patients and the standardization of prophylactic measures. The
efficacy of this prophylaxis is amply documented. In a randomized, double-blind,
placebo-controlled trial comparing fluconazole (400 mg/day) versus placebo until
10 weeks after transplantation, performed in 212 liver transplant recipients, Winston
et al. [25] found a significant reduction in fungal infections (43 vs. 9%; p \ 0.001),
both superficial (28 vs. 4%; p \ 0.001) and invasive (23 vs. 6%; p \ 0.001). In
another study performed in liver transplant recipients, Fortun et al. [26] compared the
administration of lipid amphotericin B (AmBisome) formulations with no treatment
at all. They found a significant reduction in the rate of invasive fungal infections
(0 vs. 32%; p = 0.03), with no significant effect in mortality rates. However, in
another randomized study performed in 129 consecutive liver transplant recipients
divided into three intervention groups (liposomal amphotericin B i.v. plus itraco-
nazole po; fluconazole i.v. plus itraconazole po; placebo), no significant differences
were found regarding the incidence of fungal infection, although fungal colonization
was higher in the placebo group [27].
With regard to viruses, CMV infection is one of the most frequent in trans-
plantation recipients [28]. The risk of CMV infection and disease following
transplantation is highest in CMV-seronegative recipients receiving a CMV-
seropositive organ and is associated with an increased predisposition to chronic
allograft rejection due to an immune-mediated vascular lesion [29]. This is why
prophylactic measures against CMV have been amply investigated. In a controlled,
multicenter trial, 155 organ transplant recipients (seropositive donor/seronegative
recipient) received ganciclovir i.v. (5 mg/kg per day) for 510 days and then either
acyclovir po (400 mg t.i.d.) or ganciclovir po (1 g t.i.d.) for an additional
12 weeks. Treatment with ganciclovir po was associated with a significant
decrease in the incidence of CMV symptomatic disease when compared with the
acyclovir po group (32 vs. 50%; p \ 0.05) [30]. In another trial comparing the
different strategies of cytomegalovirus prophylaxis in liver transplant recipients,
ganciclovir po was found to be the most cost-effective strategy [31]. In a recent
systematic revision [32], prophylactic treatment with antivirals was found to be
associated with a significant decrease in CMV infection [relative risk (RR) 0.62;
95% CI 0.530.73; p \ 0.001) and disease (RR 0.51; 95% CI 0.410.64;
p \ 0.001) compared with placebo or no treatment. This revision failed to show
significant differences in the incidence of organ rejection or mortality rates.
Another meta-analysis [33] of 1,980 patients from 17 controlled trials raises
similar conclusions, showing a significant decrease in the incidence of CMV
disease (OR 0.20; 95% CI 0.130.31) and the rate of allograft rejection. CMV
infection is potentially severe but rarely requires ICU hospitalization. Regardless,
this disease should be considered with pneumonias presenting with an atypical
course, particularly in patients immunosuppressed with mycophenolate.
Beyond the sixth month after transplantation, the organ recipient achieves a stable
degree of immunosuppression, and infectious problems at this stage are similar to those
that take place in the immunocompetent host. However, patients with chronic rejec-
tion who need a strong and maintained state of immunosuppression, are vulnerable to
opportunistic infections (P. jirovecii, Listeria monocytogenes, Nocardia asteroides,
Aspergillus, Cryptococcus neoformans, etc.) [3]. This raises the need for prolonged
prophylaxis with cotrimoxazole and antifungals. Some of these prophylactic strategies
are already well defined. The use of daily single-dose cotrimoxazole, for example,
during the 412 months following transplantation, has nearly eradicated the incidence
of P. jirovecii and has decreased the rate of infections caused by L. monocytogenes,
N. asteroides, and T. gondii in high-risk patients such as cardiac transplant recipients.
In this late post-transplant period, the risk of tuberculosis (reactivation of a
latent infection) is 2070 times higher than in the general population. In a large
series of 4,634 solid organ transplant recipients, Garca-Goetz et al. [34] found an
incidence of 0.45%, but it can reach 6% in endemic areas [35]. Mortality rates of
these patients can be as high a 30%. According to this, patients with risk factors for
tuberculosis (latent tuberculosis, familiar exposure, or radiologic suspicion) should
receive prophylaxis with isoniazid at least for 912 months, starting early after
transplantation.
19.2.4 Treatment of Confirmed or Suspected Infections
19.2.4.1 Early Postoperative Infection

The rate of infection in the solid organ transplant recipient is thought to be higher
than in other diagnostic groups of critical care patients [1]. Despite this data, an
analysis of 429 cardiac and liver transplant recipients admitted to our ICU in the
Table 19.2 Infectious episodes during intensive care unit (ICU) stay (transplanted patients vs
overall ICU)
Transplanted ICU overall

Patients 75/429 (17.5%) 1796/6192 (29%)
Episodesa 152 (35.1*) 2131 (33.8*)
Type of infection [n (*)]
VAP 9 (2.1) 153 (2.4)
LRTI 5 (1.1) 209 (3.3)
UTI 19 (4.4) 451 (7)
PB ? CRB 51 (11.7) 690 (10.9)
Peritonitis 8 (1.8) 54 (0.8)
Surgical wound 13 (3) 130 (2)
Other 8 (1.8) 160 (2.5)
SB 13 (3) 158 (2.5)
VAP ventilator associated pneumonia, LRTI lower respiratory tract infection, UTI urinary tract
infection, PB primary bacteremia, CRB catheter-related bacteriemia, SB secondary bacteremia
a
Excludes secondary bacteriemias
*Episodes/1,000 ICU days
last 7 years shows a global incidence of early infection of 17.5%, which is similar
to the incidence in other severely ill surgical patients and lower than in the rest of
the patients admitted with other diagnoses (29%). According to this analysis, there
was no difference in the site and incidence of infections between transplant
recipients and the rest of the patients, with exception of peritonitis, which was
more frequent in liver transplant recipients (1.8 vs. 0.8 per 1,000 days of stay)
(Table 19.2). According to etiology, the analysis showed no difference between
groups regarding prevailing bacterial infections over fungal and viral ones
(Table 19.3). A low incidence of multi-drug-resistant pathogens was found, with
the exception of Acinetobacter baumannii, which was more frequent in the
transplant recipients (9 vs. 4.3%; p \ 0.005). Aerobic Gram-negative bacilli were
more frequent in the group of nontransplant patients (50 vs. 44%; p \ 0.05),
whereas CMV infection rates were higher in transplant recipients (4 vs. 0%;
p \ 0.001). In a univariate analysis, the risk factors of infection were acute liver
failure or code 0 previous to transplantation, Acute Physiology and Chronic
Health Evaluation (APACHE) II score at admission, the need for reintervention for
any cause, early organ rejection, and pretransplant ICU length of stay. The rates of
days with mechanical ventilation and with central venous catheters were also
higher in patients with infection. In a regression analysis, the only risk factors
for infection found were code 0 as the cause of transplantation (OR 7.5, 95% CI
2.819.5), prolonged mechanical ventilation (OR 11.9, 95% CI 6.123.2), and
Table 19.3 Microbiology of infectious episodes in patients admitted to the intensive care unit
(ICU) after transplantation versus patients admitted for other reason (each infectious episode
could involve several microorganisms)
After transplantation Other reason

Microbiology (n) 224 2,700 p value
CNS 56 (25%) 386 (14.3%) \0.001
Enterococcus faecalis 14 (6.3%) 326 (12.1%) \0.004
Enterococcus spp. 6 (2.7%) 40 (1.5%) ns
Staphylococcus aureus 16 (7.2%) 144 (5.3%) ns
(% MR) (45) (45)
Other GPC 10 (4.5%) 136 (5%) ns
Total GPC 46.3% 38.2% ns
Acinetobacter baumannii 20 (9%) 117 (4.3%) \0.005
Escherichia coli 18 (8.2%) 303 (11.2%) ns
Klebsiella pneumoniae 15 (6.8%) 165 (6.1%) ns
Serratia marcescens 3 (1.3%) 106 (3.9%) ns
Stenotrophomonas maltophilia 9 (4%) 10 (0.37%) ns
Pseudomonas aeruginosa 15 (6.8%) 267 (9.8%) ns
Other AGNB 17 (7.6%) 381 (14.1%) \0.005
Total AGNB 44% 50.3% \0.05
Candida albicans 6 (2.7%) 144 (5.3%) \0.02
Candida spp. 10 (4.5%) 134 (4.9%) ns
Total fungi 7.2% 10.3% ns
Anaerobes 11 (0.4%) ns
CMV 9 (4%) 0 \0.001
CNS coagulase-negative staphylococci, MR methicillin-resistant S. aureus, GPC Gram-positive
cocci, AGNB aerobic Gram-negative bacillus, CMV cytomegalovirus
APACHE II score [16 at the admission (OR 2.19, 95% CI 1.14.1). Length of
ICU stay and mortality rate were highly conditioned by infection (Table 19.4).
Managing these infections should not be difficult, always bearing in mind the
characteristic pathogens of each center and their resistance profile.
19.2.4.2 Late Infections

Admission to the ICU of an organ transplant recipient with an infectious disease
and who requires hemodynamic or respiratory support is a challenging situation
for intensivists. The difficulty lies not only in establishing an etiologic diagnosis,
but also in choosing an adequate empiric antibiotic therapy, which should be
Table 19.4 Risk factors for infection in 429 solid-organ-transplanted patients
Infection
Total (%) Yes (%) No (%) p value

Number 429 75 (17.5) 354 (82.5)
Age 52.5 12 51 13 52.7 11.7 ns
Male 308 (71.8) 52 (69) 256 (72.3) ns
Cirrhosis 364 (84) 54 (72) 310 (87) ns
Alcohol abuse 132 (31) 19 (25) 113 (32) ns
Hepatocarcinoma 44 (10) 9 (12) 35 (10) ns
Code 0 31 (7.2) 19 (13) 12 (3.4) \0.001
Unscheduled surgery 38 (8.8) 19 (25.3) 19 (5.4) \0.001
APACHE II (admission) 13.3 4.7 16 5.4 12.7 4 \0.01
Early rejection 77 28 (37.3) 49 (13.8) \0.001
Retransplant 9 9 (12) 0 \0.001
ICU stay 10.1 13.6 29 24 64 \0.001
Ratio MV 0.49 0.76 0.20 \0.001
Ratio CVC 0.86 0.99 0.73 \0.05
ICU mortality 33 (7.7) 24 (32) 9 (2.5) <0.001
ICU intensive care unit, APACHE II acute physiology and chronic health evaluation, Ratio MV/
CVC total days of mechanical ventilation (central venous catheter)/total ICU days
started immediately. The chronology after transplantation can be helpful in the

decision-making process given a predictable timeline (depending on the level of
immunosuppression) that correlates with certain pathogens (most of all viruses).
In a medium-term period following transplantation, immunosuppression
favors the development of opportunistic infections (P. jirovecii, CMV, Listeria,
Toxoplasma and Nocardia) and the reactivation of latent infections (tuberculosis).
However, the spectrum of infections in the transplant recipient who is ill enough to
require admission to the ICU is similar to the general population. When starting an
immediate empirical treatment, local resistance profiles must be kept in mind
(prevalence of MRSA, extended-spectrum betalactamase-producing Escherichia
coli, or penicillin-resistant Streptococcus pneumoniae).
Beyond the sixth month after transplantation; with an adequate organ function
and a minimum degree of immunosuppression, most transplant recipients do rel-
atively well, being at risk from the same infections seen in the general community.
However, patients who have had frequent episodes of rejection, requiring aug-
mented immunosuppressive therapy, remain at risk for opportunistic pathogens
(P. jirovecii, L. monocytogenes, N. asteroides, Aspergillus, C. neoformans) [3].
In the rest of the transplant recipients, the most frequent infections during this late
period after transplantation are bloodstream and respiratory tract infections. The
latter can have an atypical clinical or radiologic pattern, frequently requiring
invasive diagnostic procedures and computed tomography (CT) [36]. Acute epi-
sodes with a radiologic pulmonary consolidation suggest bacterial infections
(including Legionella), whereas diffuse edema-like infiltrates usually correspond
to a viral processes. Subacute and chronic episodes are mainly caused by fungi,
Nocardia, tuberculosis, P. jiroveci, and viruses. According to our experience with
27 solid organ transplant recipients admitted to our ICU with severe sepsis, most
episodes had a bacterial etiology (five E. coli, three pneumococcus, three
Pseudomonas, one Corynebacterium, and one MRSA); only three cases of CMV
disease where diagnosed. Two patients had influenza H1N1 (during the 2009
pandemia), and in three cases of pneumonia, cultures yielded no results. Baseline
characteristics of the patients where similar to those of transplant recipients
admitted for any other cause (Table 19.5). The epidemiology of infections was
compared between patients with ICU-acquired infections in the immediate post-
transplant period and previously transplanted patients who were admitted to the
ICU with other diseases. This comparison showed no significant differences in the
site and number of infectious episodes between groups (Table 19.6).
During this late post-transplant period, the risk of reactivation of latent tuber-
culosis is still present. This situation is challenging as a result of drug interactions
between tuberculostatics and immunosuppressive agents. Rifampin, for example,
decreases tacrolimus and cyclosporine levels. This pharmacokinetic interaction
can lead to organ rejection. In these cases, treatment regimens should be based in a
combination of other tuberculostatics (isoniazid ? ethambutol ? pyrazinamide,
for example) and may include the possibility of using fluoroquinolones and
streptomycin if the course is torpid.
19.3 Bone Marrow Transplants
Unlike solid organ transplant patients, bone marrow transplant recipients are never
admitted to the ICU. Nevertheless, they require pretransplant treatment with high-
dose radiotherapy, chemotherapy, or both, in order to abolish the patients immune
system. This causes a severe neutropenia as well as damaging the integrity
of mucosal barriers, favoring bacterial translocation from the digestive tract.
Consequently, infections will be a frequent problem in these patients [37]. Bone
marrow recipients need at least 24 months to become immunocompetent. This
period may be even longer for allogenic transplants. This fact, as well as the wide
etiological spectrum (bacteria, viruses, and fungi) seen in these patients, hinders
the establishment of preventive strategies. Even though many infections of the
bone marrow recipient have an intestinal origin, prophylactic strategies do not
include SDD or systemic antibiotics in the asymptomatic patient due to the lack of
evidence of their effectiveness. Some studies in which fluoroquinolones were
administered as a prophylactic measure during the period of neutropenia showed a
Table 19.5 Previously transplanted patients with/without infection at ICU admission
With infection Without infection p value

Patients 27 50
Age 54 16.3 56 13.5 ns
Male 23 (82%) 35 (70)
Unscheduled surgery 6 (22.2) 10 (20) ns
APACHE II (admission) 22.4 4.8 21.4 5.9 ns
ICU stay 15.7 13.2 15 18 ns
Ratio MV 0.79 0.79 ns
Ratio CVC 1 1 ns
ICU mortality 10 (37) 21 (42) ns
APACHE II acute physiology and chronic health evaluation, ratio MV/CVC total days of
mechanical ventilation (central venous catheter)/total ICU days
Table 19.6 Infectious episodes during ICU stay (admission post-transplant vs. transplant, with
admission for other reason)
Scheduled transplant Other reason p value

Patients 75/429 (17.5%) 20/78 (25.6%)
a b
Episodes 152 (35.1 ) 44 (37.6b) ns
b
Type of infection n nb
VAP 9 (2.1) 2 (1.7) ns
LRTI 5 (1.1) 2 (1.7) ns
UTI 19 (4.4) 8 (6.8) ns
PB ? CRB 51 (11.7) 15 (12.8) ns
CVC 39 (8.9) 9 (7.7) ns
Peritonitis 8 (1.8) 2 (1.7) ns
Surgical wound 13 (3) 5 (4.2) ns
Other 8 (1.8) 2 (1.7) ns
SB 13 (3) 4 (3.4) ns
VAP ventilator associated pneumonia, LRTI lower respiratory tract infection, UTI urinary tract
infection, CVC central venous catheter, PB primary bacteremia, CRB catheter related bacterie-
mia, SB secondary bacteremia
a
Excludes secondary bacteriemias
b
Episodes/1,000 intensive care unit days
reduction in the incidence of bacteremia, with no effect in mortality rates [38].

However, when considering the establishment of a prophylactic protocol, the anti-
biotics included should be adjusted to the resistance patterns of the most common
pathogens present in each center. Moreover, the prevalence of quinolone-resistant
staphylococcus, pneumococcus and E. coli, as well as the appearance of vancomy-

cin-resistant enterococci, must be borne in mind. Therefore, extensive use of these
antibiotics should be avoided. Neutropenic patients are characteristically at risk of
fungal infections. Prophylaxis with fluconazole has markedly reduced the incidence
of fungal infections (Candida spp.), with a favorable impact on mortality rates
[39]. However, the emergence of infections caused by non-albicans Candida spp.
(C. krusei, C. glabrata) has been related to this prophylactic strategy [40]. Surveil-
lance cultures could be useful to decide when to administer prophylactic fluconazole
in patients with prolonged neutropenia. In a study performed by Guiot et al. [41],
patients with low concentrations of Candida in the oropharynx (\10 CFU) and fecal
samples (\103 CFU/g feces) did not develop systemic candidiasis; 19% of those with
one high concentration sample and 86% of those with both samples at high con-
centration had fungemias. There is no consistent evidence so far concerning the use
of SDD in this field. However, its use seems justified during the early phase of the
transplant regime, considering the expected pathogens at this point (Gram-negative
bacilli and fungi). The lack of evidence can be explained by the low number of bone
marrow recipients admitted to the ICU and by the fact that SDD is practically limited
to the intensive care area. Regarding viruses, the most severe and frequent infectious
process (between 30 and 100 days after transplantation) is interstitial pneumonia,
which behaves as an acute respiratory distress syndrome and is usually caused by
CMV. This has led to the use of prophylactic courses of ganciclovir, aiming to avoid
this risk. In a randomized controlled trial [42], bone marrow recipients who had
asymptomatic pulmonary CMV infection (positive cultures for CMV in broncho-
alveolar lavage while remaining asymptomatic) were randomly assigned to either
prophylactic ganciclovir (5 mg/kg twice daily for an initial period of 12 weeks, and
then five to seven times per week until day 120) or observation alone. The results
showed a significant decrease in the incidence of CMV disease and mortality rate,
with few side effects. In the late phase after transplantation ([100 days), vaccination
against encapsulated bacteria (S. pneumoniae, H. influenzae, and N. meningitidis) is
recommended in allogenic transplant recipients with a chronic form of graft versus
host disease requiring treatment with cyclosporine or tacrolimus. Vaccines should be
given 12 and 24 months after transplantation.
Regarding infections during the neutropenia phase of bone marrow transplants,
the primary site is usually unknown. Therefore, an empirical an immediate antibiotic
treatment is necessary [43]. Bacterial and fungal infections are the most frequent.
19.4 Conclusion
Solid-organ and bone marrow transplant recipients are at high risk for infection in
the short and medium term. This risk depends on the level of immunosuppression
necessary to avoid rejection. Preventive maneuvers aimed at minimizing this
problem have gradually increased, and most of them have proven to be effective.
The first goal in a global infection prevention plan is to minimize length of ICU
stay. This goal can be achieved by implementing quick weaning protocols,
removing central lines and drainages as soon as possible, and establishing inten-
sive physiotherapy programs. The fact that in an early phase the epidemiology of
infections is comparable in transplant recipients and other surgical patients
admitted to the ICU could justify the prophylactic use of SDD. This might be
particularly important while the patient is undergoing mechanical ventilation or
until a minimum level of immunosuppression can be achieved (cyclosporine
monotherapy, for example). The use of SDD could also be supported by the
emergence of fungal infections (mainly Candida spp.), favored by the use of
steroids in immunosuppressive protocols. In the liver transplant recipient, several
studies have demonstrated a protective effect of both fluconazole and topical
enterally administered nonabsorbable antifungals as part of SDD. The risk of
selecting non-albicans Candida spp. resistant to azoles, observed with the use of
fluconazole, is not likely to exist with topical use of amphotericin B. SDD has been
the subject of intense debate given the fact that it has no effect on post-transplant
mortality rates. However, this reason should not be enough to refuse treatment
with the potential benefit of avoiding infection. To resolve this controversial issue,
a study with a complex design and a large sample size is necessary. However, it is
unlikely that this kind of study will ever be performed.
In the medium- and long-term periods, the problems concerning prevention rely
on the patients willingness to maintain a long prophylactic regimen, together with
the risk of toxicity, interactions with immunosuppressive agents, and selection of
resistant pathogens. Reported rates of nonadherence to the medical recommen-
dations after transplantation are unacceptably high in some studies, reaching seven
to 36 cases of nonadherence to immunosuppressive treatment per 100 patients/year
[44]. Educational programs and greater patient participation in the prognosis are
useful tools to achieve this goal.
Regarding the profile of infections in transplant recipients, there are three
general time frames to consider: the first month (early postoperative period), the
second through the sixth month, and the late post-transplant period (beyond
the sixth month or first year). In the early postoperative period, infections are
similar in site and etiology to those occurring in general surgical patients.
The intermediate period (26 months after transplantation), is when the risk
of opportunistic infections (P. jirovecii and, less frequently, Listeria and Nocar-
dia), as well as reactivation of latent infections (tuberculosis, CMV, Toxoplasma)
become manifest. Therefore, these etiologies should be ruled out when a transplant
recipient is admitted with an infectious episode. Beyond the first year after
transplantation, if the level of immunosuppression is minimal, the recipient has the
same risk of infection as the general population, with the exception of the possi-
bility of a reactivation of latent tuberculosis. Although it is unusual that patients
with these infections of the late post-transplant period are admitted to the ICU,
they can raise a diagnostic challenge. A high index of suspicion is necessary to
choose the appropriate diagnostic procedures (serological tests, virus polymerase
chain reaction, biopsies, etc.) and to start directed treatment as soon as possible;
the patients prognosis will markedly depend on this.
References
1. Patel R, Paya CV (1997) Infections in solid-organ transplant recipients. Clin Microb Rev
10:86124
2. Torbenson M, Wang J, Nichols L et al (1998) Causes of death in autopsied liver
transplantation patients. Mod Pathol 11:3746
3. Fishman JA (2007) Infection in solid-organ transplant recipients. N Engl J Med 357:
26012614
4. Avery RK (2004) Prophylactic strategies before solid-organ transplantation. Curr Opin Infect
Dis 17:353356
5. White CJ (1997) Varicela-zoster virus vaccine. Clin Infect Dis 24:753
6. Reichenspurner H, Gamberg P, Nitschke M et al (1997) Significant reduction in the number
of fungal infections after lung, heart-lung, and heart transplantation using aerosolized
amphotericin B prophylaxis. Transpl Proc 29:627628
7. Brown RS, Lake JR, Katzman BA et al (1996) Incidence and significance of Aspergillus
cultures following liver and kidney transplantation. Transplantation 61:666669
8. Liberati A, DAmico R, Pifferi S, Brazzi L (2007) Antibiotic prophylaxis to reduce
respiratory tract infections and mortality in adults receiving intensive care. Cochrane Review.
Cochrane Library, Issue 3
9. Wiesner RH, Hermans PE, Rakela J et al (1988) Selective bowel decontamination to decrease
gram-negative aerobic bacterial and Candida colonization and prevent infection after
orthotopic liver transplantation. Transplantation 45:570574
transplants. A randomized prospective study. Transplantation 55:13061309
reduces gram-negative pulmonary colonization but not systemic endotoxemia in patients
12. van Saene JJM, Stoutenbeek CP, van Saene HKF et al (1996) Reduction of the intestinal
endotoxin pool by three different SDD regimens in human volunteers. J Endotoxin Res
3:337343
13. Emre S, Sebastian A, Chodoff L et al (1999) Selective decontamination of the digestive tract
helps prevent bacterial infections in the early postoperativ period after liver transplant. Mt
Sinai J Med 66:310313
14. van Enckevort PJ, Zwaveling JH, Bottema JT et al (2001) Cost effectiveness of selective
decontamination of the digestive tract in liver transplant patients. Pharmacoeconomics 19:523530
15. Garcia-San Vicente B, Canut A, Labora A et al (2010) Descontaminacin digestiva selectiva:
respercusin en la carga de trabajo y el coste de laboratorio de microbiologa y tendencias en
la resistencia bacteriana. Enferm Infecc Microbiol Clin 28:7581
patients: a systematic review of the evidence. Arch Surg 134:170176
digestive tract to prevent posoperative infection: a randomized placebo-controlled trial in
18. Sadfar N, Said A, Lucey MR (2004) The role of selective digestive decontamination for
reducing infection in patients undergoing liver transplantation: a systematic review and meta-
19. Rayes N, Seehofer D, Hansen S et al (2002) Early enteral supply of lactobacillus and fiber
20. Mor E, Cohen J, Erez E et al (2001) Short intensive care unit stay reduces septic
complications and improves outcome after liver transplantation. Transplant Proc 33:
29392940
21. Antonelli M, Conti G, Bufi M et al (2000) Noninvasive ventilation for treatment of acute
respiratory failure in patients undergoing solid organ transplantation: a randomized trial.
JAMA 283:235241
22. Whiteman K, Nachtmann L, Kramer D et al (1995) Effects of continuous lateral rotation
therapy on pulmonary complications in liver transplant patients. Am J Crit Care 4:133139
23. Cole GT, Halawa AA, Anaissie EJ (1996) The role of the gastrointestinal tract in the
hematogenous candidiasis: from the laboratory to bedside. Clin Infect Dis 22(Suppl 2):
S73S88
24. Garrido RS, Aguado JM, Diaz-Pedroche C et al (2006) A review of critical periods for
opportunistic infection in the new transplantation era. Transplantation 82:14571462
25. Winston DJ, Pakrasi A, Busuttil RW (1999) Prophylactic fluconazole in liver transplant
recipients. A randomized, double-blind, placebo-controlled trial. Ann Int Med 131:729737
26. Fortun J, Martin-Davila P, Moreno S et al (2003) Prevention of invasive fungal infections in
liver transplant recipients: the role of prophylaxis with lipid formulations of amphotericin B
in high risk patients. J Antimicrob Chemother 52:813819
27. Biancofiore G, Bindi ML, Baldasarri R et al (2002) Antifungal prophylaxis in liver transplant
recipients: a randomized placebo-controlled study. Transpl Int 15:341347
28. Fernandez A, Amezquita Y, Fernandez-Tarrago E et al (2009) Prophylaxis and treatment of
cytomegalovirus infection postrenal transplantation in two Madrid units. Transpl Proc
41:24162418
29. Arthur SK, Pedersen RA, Kremers WK et al (2009) Delayed-onset primary cytomegalovirus
disease and the risk of allograph failure and mortality after kidney transplantation. Clin Infect
Dis 46:840846
30. Rubin RH, Kemmerly SA, Conti D et al (2000) Prevention of primary cytomegalovirus
disease in organ transplant recipients with oral ganciclovir or oral acyclovir prophylaxis.
Transpl Infect Dis 2:112117
31. Das A (2000) Cost-effectiveness of different strategies of cytomegalovirus prophylaxis in
orthotopic liver transplant recipients. Hepatology 31:311317
32. Couchoud C (2001) Cytomegalovirus prophylaxis with antiviral agents for solid organ
transplantation. Cochrane Database of Systematic Reviews, Issue 4
33. Kalil AC, Levitsky J, Lyden E et al (2005) Meta-analysis: the efficacy of strategies to prevent
organ disease by cytomegalovirus in solid organ transplant recipients. Ann Int Med 143:
870880
34. Garca-Goetz JF, Linares L, Benito N et al (2009) Tuberculosis in solid organ transplant
recipients at a tertiary hospital in the last 20 years in Barcelona, Spain. Transpl Proc 41:
22682270
35. Benito N, Sued O, Moreno A et al (2002) Diagnosis and treatment of latent tuberculosis
infection in liver transplant recipients in an endemic area. Transplantation 74:13811386
36. Torres A, Ewing S, Insausti J et al (2000) Etiology and microbial patterns of pulmonary
infiltrates in patients with orthotopic liver transplantation. Chest 117:502949
37. Kernan NA, Bartsch G, Ash RC et al (1993) Analysis of 462 transplantations from unrelated
donors facilitated by the National Marrow Donor Program. N Engl J Med 328(9):593
38. Cruciani M, Rampazzo R, Malena M et al (1996) Prophylaxis with fluoroquinolones for
bacterial infection in neutropenic patients: a meta-analysis. Clin Infect Dis 23:795
39. Marr KA, Siedel K, Slavin MA et al (2000) Prolonged fluconazole prophylaxis is associated
with persistent protection against candidiasis-related death in allogenic marrow transplant
recipients: long-term follow-up of a randomized, placebo-controlled trial. Blood 96:2055
40. Abi-Said D, Anaissie E, Uzun O et al (1997) The epidemiology of hematogenous candidiasis
caused by different Candida species. Clin Infect Dis 24:1122
41. Guiot HFL, Fibbe WE, Vant Wout JW (1996) Prevention of invasive candidiasis by
fluconazole in patients with malignant hematological disorders and a high grade of candida
colonisation. (Abstract LM33) 36th Interscience conference on antimicrobial agentes and
chemotherapy, New Orleans
42. Schmidt GM, Horak DA, Niland JC et al (1991) A randomized, controlled trial of
prophylactic ganciclovir for cytomegalovirus pulmonary infections in recipients of allogenic
bone marrow transplants: the City of Hope-Standford-Syntex CMV Study Group. N Engl J
Med 324:10051011
43. Hughes WT, Armstrong D, Bedey GP et al (1997) 1997 Guidelines for the use of
antimicrobials agents in neutropenic patients with unexplained fever. Clin Infect Dis
25(3):551
44. Dew A, Di Martni AF, De Vito A (2007) Rates and risk factors for nonadherence to the
medical regimen after adult solid organ transplantation. Transplantation 83:858873
Clinical Virology in NICU, PICU
and AICU 20
20.1 Introduction
Viruses are significant causes of nosocomial infections, but their importance has
been underappreciated in the past. However, outbreak of severe acute respiratory
syndrome (SARS), avian and pandemic influenza with high morbidity and
mortality rates increased the awareness of intensivists regarding the devastating
effects of nosocomial spread of viral infections in intensive care units (ICU).
Advances in medicine have led to a large number of immunocompromised
patients susceptible to severe viral infections. Many such patients are cared for in
the ICU and in turn become an infectious hazard for other vulnerable patients.
Health care workers can acquire common viral infections from the community and
spread them to susceptible patients in the ICU. Patients in neonatal (NICU) and
paediatric (PICU) ICUs are most vulnerable because of the lack of prior immunity
against many viruses circulating in the community. Recent improvements in
diagnostic methods have enabled the rapid diagnosis and monitoring of many viral
infections. Rapid and accurate typing of viral strains using a molecular technique
can help identify the source of outbreaks. Also, specific postexposure prophylaxis
and treatment are now available for many important nosocomial viral infections.
In this chapter, we discuss some of the important viruses that could be associated
with nosocomial infections in the ICU, according to their usual route of trans-
mission (Table 20.1). Infection control measures recommended for preventing
these viral infections are listed in Table 20.2.
C. Y. W. Tong (&)
Infection, Guys and St Thomas NHS Foundation Trust
and Kings College London School of Medicine, London, UK
334 C. Y. W. Tong and S. Schelenz
Table 20.1 Mode of viral infection transmission in the intensive care unit
Respiratory Faecaloral Blood and body Direct contact/

route route fluid fomites
Influenza viruses +++ ++
Respiratory syncytial virus +++ ++
Parainfluenza viruses +++ ++
Adenovirus +++ ++ +++
SARS coronavirus +++ ++
Varicella zoster virus +++ ++
(chickenpox)
Rotavirus + +++ ++
Norovirus ++ +++ ++
Enterovirus/parechovirus + +++ ++
Hepatitis A virus +++ ++
Hepatitis B virus +++
Hepatitis C virus ++
Human immunodeficiency ++
virus
Haemorrhagic fever viruses ++++ +++
Cytomegalovirus + ++
Herpes simplex virus ++
Varicella-zoster virus ++
(zoster)
Rabies ++
SARS severe respiratory distress syndrome; Unlikely; possible; ? common (the number
of ? is an arbitrary indicator of transmissibility)
20.2 Viruses Transmitted by Droplets and Airborne Route
20.2.1 Influenza Viruses
Influenza viruses (family Orthomyxoviridae) are classified into types A, B and C.

Annual seasonal outbreaks of influenza are caused by minor antigenic changes
(antigenic drift) seen in influenza A and B viruses. Major changes in antigenic
subtypes (antigenic shift) are only found in influenza A virus and typically involve
the emergence of novel hemagglutinin (H) and/or neuraminidase (N) proteins on
the viral envelope. Pandemic influenza occurs when a new influenza A strain
emerges, to which the majority of the worlds population has little or no immunity.
Table 20.2 Infection control measures for preventing viral infections
20
Virus Isolation or cohorting Hand Apron/ Gloves Masks/ Incubation Duration of infectivity
washing gown+ goggles
Influenza viruses 4(negative pressure) IB 4 4 4 4 IB 14 days Prodromal phase and 7 days after onset
Respiratory 4 II 4 IA 4 IB 4 IA 28 days 48 h before symptoms and 7 days from
syncytial viruses onset; longer in immunocompromised (up
(RSV) to 30 days)
Parainfluenza 4 4 4 4 24 days As long as symptoms last
viruses
Adenovirus 4 4 4 4 510 days As long as symptoms last
SARS coronavirus 4 (negative pressure) IA 4 4 4 4 (FFP3 46 days (max. Peak at day 10 of illness, no reported
or N95) reported transmission 10 days beyond resolution of
IB 14 days) fever
Clinical Virology in NICU, PICU and AICU
Varicella-zoster 4(negative pressure) IB 4 4 4 1021 days 2 days before first vesicle until all lesions
virus (chickenpox) are crusted
Rotavirus 4 4 4 4 23 days 2 days before symptoms and up to
47 days after onset of illness
Norovirus 4 IB 4 IA 4 IB 4 IB 1548 h Up to 48 h after becoming symptom free
Enterovirus/ 4 (young infants often 4 4 4 225 days 714 days from onset of illness;
parechovirus require care in SCBU or asymptomatic shedding common
NICU)
Hepatitis A virus 4 4 4 4 26 weeks Infectious 1 week before onset of illness,
infectivity declines rapidly after onset of
illness
Hepatitis B virus 4 4 23 months As long as patient is viremic
(continued)
335
336
Virus Isolation or cohorting Hand Apron/ Gloves Masks/ Incubation Duration of infectivity
washing gown+ goggles
Hepatitis C virus 4 4 23 months As long as patient is viremic
Human 4 4 36 weeks Indefinitely, though viremia can be
immunodeficiency controlled by therapy
virus (HIV)
Viral hemorrhagic 4 (high security 4 4 4 4 321 days High infectivity during illness
fever viruses isolation)
(VHF)
Cytomegalovirus 4 4 4 36 weeks Congenital infectionfrom birth.
(CMV) (congenital (congenital from primary Asymptomatic shedding common
CMV) CMV) infection.
Herpes simplex 4 4 Often due to Until lesions have healed
virus (HSV) reactivation
Varicella-zoster 4 4 4 Due to Until vesicles crusted over
virus (shingles) reactivation
Rabies virus 4 4 4 4 4 28 weeks or Duration of illness
longer
Categorisation of recommendations: IA strongly recommended for all hospitals and strongly supported by well-designed experimental or epidemiological
studies; IB strongly recommended for all hospitals and viewed as effective by experts in the field (these recommendations are based on strong rationale and
suggestive evidence, even though scientific studies may not have been performed). II suggested for implementation in many hospitals (these recommen-
dations may be supported by suggestive clinical or epidemiological studies, a strong theoretical rationale or definitive studies applicable to some but not all
hospitals
FFP3 or N95 high-filtration respirators, SCBU special care baby unit, NICU neonatal intensive care unit, 4 recommended and suggested for implementation
in most settings, SARS severe acute respiratory distress syndrome, HIV human immune deficiency virus, VHF viral haemorrhagic fevers, CMV
cytomegalovirus
20 Clinical Virology in NICU, PICU and AICU 337
There were three influenza pandemics in the last century, of which the pandemic in
1918 due to the H1N1 virus was the most severe. The first pandemic of this
century occurred in 2009 [1] and was due to another H1N1 variant that emerged
through a quadruple reassortment of viral RNAs derived from human, avian,
Eurasian and North American swine influenza sources [2]. The presence of animal
influenza subtypes, particularly avian influenza viruses such as H5N1, is of con-
tinuous concern, as these could be the source of future pandemics. Though with
relatively high case-fatality rate, H5N1 avian influenza virus has so far only caused
a limited number of human infections in restricted geographical locations with
little evidence of human to human spread. However, the 2009 pandemic H1N1
virus proved to be a major burden for ICU staff [3].
Clinically, influenza infection is characterised by abrupt onset of fever, sore
throat, myalgia, cough, headache and malaise. Young children may develop croup,
pneumonia or middle ear infection. With seasonal influenza, complications are
often seen in the elderly, the immunocompromised and those with pre-existing
chronic heart or lung disease or diabetes. During the 2009 H1N1 pandemic,
children and young adults were more susceptible [4]. Overall fatality rate was
\0.5%, but as many as 931% of hospitalised patients needed ICU admission [5].
Severe disease and high mortality rates were seen in pregnant women, patients
with underlying medical pulmonary, cardiac, metabolic, neuromuscular illness and
severe obesity, and those in whom the diagnosis and admission was delayed [68].
Respiratory failures could be caused by viral pneumonia and acute respiratory
distress syndrome (ARDS). In addition, secondary bacterial infection with Strep-
tococcus pneumoniae or Staphylococcus aureus (often methicillin resistant) were
found in 2024% of ICU patients and 2638% of patients who died [3, 5, 9]. Fatal
cases were often complicated by multiorgan failure.
Influenza has a short incubation time of 14 days. The virus is transmitted via
droplets, and patients are infectious during the prodromal phase and up to 7 days
after symptom onset. Rapid antigen detection from respiratory secretions is
available, but this was found to be insensitive for the 2009 H1N1 pandemic virus
[10]. More sensitive and specific real-time polymerase chain reaction (PCR)
methods had to be used [11]. Due to the infection-control hazards of taking
nasopharyngeal aspirates or bronchoalveolar lavage, the use of throat and nasal
swabs were advocated. A complete respiratory diagnostic workup needed to be
performed to exclude other viral, bacterial and noninfectious causes. A single
negative influenza PCR result on an upper respiratory sample did not definitively
exclude the diagnosis [12]. In addition, other concurrent or secondary infections
had to be considered. Protocols needed to be in place to ensure satisfactory triage
of patients according to severity [13]. Early administration of specific neuramin-
idase inhibitors, such as oral doses of oseltamivir or inhalation zanamivir, seemed
to be beneficial [14]. In more refractory cases, the off-license use of intravenously
administered zanamivir or peramivir was tried. Extracorporeal membrane
oxygenation (ECMO) was found to be useful in very severe cases [12].
The risk of nosocomial transmission to other hospitalised patients and staff is
well documented. Infected patients should ideally be cared for in a single room
or cohorted together. Health care workers should be protected through the

proper use of personal protective equipments, including respirators or masks, eye
protection, gowns/aprons and gloves [15, 16]. High-filtration respirator to FFP3
(Europe) or N95/N99 (USA) standard should be used for staff carrying out
aerosol-generating procedures after fit testing and training. Surgical masks
should be adequate for nonaerosol contacts [16]. Environmental contamination is
an important source of transmission. Good hand hygiene can prevent transmis-
sion through this route.
Vaccination is the most specific preventative measure. Annual seasonal influ-
enza vaccination to vulnerable individuals and health care workers has been
advocated. A specific vaccine against the H1N1 pandemic strain was developed
within months of the onset of the outbreak. However, vaccine uptake rates
amongst health care workers are usually poor, and more needs to be done to
educate both patients and staff.
20.2.2 Respiratory Syncytial Virus
Respiratory syncytial virus (RSV) (family Paramyxoviridae) is a major cause of

lower respiratory tract infections in young children and infants. There are two
subtypes, A and B, with varying dominance in different years [17]. The incidence
of RSV is seasonal in temperate climates, and hospital admissions usually peak
during winter months. Prematurity, bronchopulmonary dysplasia and congenital
heart disease are associated with a significant risk for admission to high-depen-
dency units or PICU. In Switzerland, it was estimated that approximately 12% of
each annual birth cohort required such admission. RSV can also cause significant
disease in adults, particularly in immunocompromised individuals such as patients
undergoing therapy for haematological malignancies, the elderly and those with
chronic pulmonary disease [18].
The most rapid diagnosis of RSV is by direct antigen detection methods such as
chromatographic immunoassays. A typical rapid test method is completed within
30 min and can be used as a point of care testing method in emergency rooms
and ICUs. However, these rapid tests lack sensitivity [19]. More recently, many
laboratories have begun using multiplex real-time nucleic acid amplification
techniques (NAAT) to diagnose respiratory tract infections, including RSV [20].
Although NAAT is highly sensitive, it is not a rapid testing method. Hence, it is
desirable to have a mixed strategy of diagnostic approaches, such as an initial rapid
direct antigen test followed by retesting of negative samples by NAAT.
Nosocomial transmission of RSV in the ICU and haemoncology units has
frequently been reported. It is important to identify infected patients and to apply
prompt and effective infection control measures (Table 20.2). It is recognised that
a combination of cohorting patients using dedicated health care staff, contact
isolation of patients, strict adherence to hand hygiene; and screening visitors,
family members and health care staff for upper respiratory tract infection symp-
toms significantly reduce the cross-infection rate of RSV. In haemoncology units,
the practice of enhanced seasonal infection control programs for RSV has been
shown to be effective [21]. The usefulness of wearing masks and goggles is less clear.
There is no safe and effective vaccine to prevent RSV infection. However,
immunoprophylaxis in the form of RSV immunoglobulin (RSV-IG) or humanised
monoclonal antibodies (palivizumab) is available as prophylaxis for some high-
risk patients to prevent serious RSV disease or to limit further nosocomial spread.
Both palivizumab and RSV-IG have been shown to decrease the incidence of RSV
hospitalisation and ICU admission, although there was no significant reduction in
the risk of mechanical ventilation or mortality rate. When given prophylaxis,
infants born\35 weeks gestational age and those with chronic lung and congenital
heart disease all had a significant reduction in the risk of RSV hospitalisation [22].
Treating RSV infection is mainly supportive, including oxygen, ventilation and
bronchodilatative drugs. Aerosolised ribavirin has often been used in severe cases,
with or without gamma globulin i.v. [23]. However, evidence for the clinical
efficacy of ribavirin in RSV infection remains inconclusive [24]. The use of
aerosolised ribavirin needs to be carefully controlled, as there are potential
teratogenic effects on pregnant staff and visitors. Others have tried a combination
of palivizumab i.v. with or without ribavirin [25].
Another paramyxovirus, known as human metapneumovirus (hMPV), shares a
similar spectrum of clinical illness as RSV. It is likely that general infection
control measures against RSV would also be effective against hMPV.
20.2.3 Parainfluenza Viruses
There are four types of human parainfluenza virus (PIV) types: PIV 14 (family
Paramyxoviridae). Infections with PIV1 and 2 are seasonal, with a peak in autumn
affecting mainly children between 6 months and 6 years of age. Clinically,
patients often present with croup or a febrile upper respiratory tract infection.
In contrast, PIV3 is endemic throughout the year and infects mostly young infants
in the first 6 month of life and up to 2 years of age. Clinically, there is no specific
presentation in PIV3, but bronchiolitis and pneumonia are not uncommon.
In immunocompromised adults, such as stem cell transplant recipients, PIV3 is
associated with a high mortality rate. Such patients often present with severe
pneumonia and many require admission to the ICU.
The diagnosis of PIV infection can be confirmed by immunofluorescence
antigen detection or NAAT [26]. Nosocomial transmission is often due to PIV3
and has been documented in neonatal care and adult haematology units [27].
Infection control precautions are the same as for RSV. Despite several uncon-
trolled case series of apparent successful use of intravenously, orally or aerosolised
administration of ribavirin to treat PIV infections, there is no clear evidence that
ribavirin with or without immunoglobulin alters mortality rates from PIV3
pneumonia or decreases the duration of viral shedding from the nasopharynx [28].
Nevertheless, there may be a role for pre-emptive early therapy with ribavirin to
prevent progression of upper airway infection to pneumonia.
20.2.4 Adenovirus
Adenovirus (family Adenoviridae) multiplies in the pharynx, conjunctiva or small

intestine. Clinically, the infection is localised and typically presents with phar-
yngitis, conjunctivitis or gastroenteritis depending on serotype. However, in young
infants and immunocompromised patients such as organ transplant recipients or
AIDS patients, adenovirus can cause severe pneumonia, disseminated infection or
haemorrhagic cystitis. The diagnosis can be confirmed by specific antigen detec-
tion tests on respiratory or stool samples. Viremia and viruria can be confirmed
and quantified using real-time PCR.
In respiratory infections, the virus spreads via droplets or through contaminated
hands or fomites. Nosocomial adenovirus infections have been reported and can be
a particular problem in neonatal units. It is important to adhere to strict infection
control procedures to prevent nosocomial spread (Table 20.2). In vitro, adenovirus
is susceptible to antivirals such as cidofovir and ribavirin [29]. Use of cidofovir in
selected patients may be successful [30].
20.2.5 Severe Acute Respiratory Syndrome: Coronavirus
A respiratory virus that caused a severe acute respiratory syndrome (SARS)

emerged from southern China in 2002. The virus was subsequently identified as a
novel virus from the Coronaviridae family and was named SARS coronavirus
(SARS CoV) [31]. SARS was associated with a high mortality rate, and of the
most concern to the international community was the potential in causing noso-
comial infections. From a single index case in a Hong Kong hotel, a series
of chains of outbreaks occurred in Vietnam, Singapore and Canada [32].
Subsequently, infections were reported in major cities in Asia, Europe and USA,
transmitted through international travel. In total, 8,422 individuals were infected,
with 916 deaths around the world. The emergence of SARS was the first wake-up
call to the medical community regarding the need for comprehensive infection
control policies in hospitals and ICU. This also led to the general provision of
personal protective equipment (PPE) with training and fitting programmes for
health care workers in many countries.
SARS is infectious from the onset of illness and infectiousness correlates with
the degree of viral shedding. Incidences of superspreaders or superspreading
events may have accounted for most of the large-scale transmissions. Older age
and underlying comorbidity are major risk factors for fatality [33]. Viral loads in
various anatomical sites also correlate with the severity of symptoms and
mortality. Shedding of SARS CoV peaks at day 10 after the onset of symptoms.
The disease pathology is characterized by uncontrolled viral replication, with a
major proinflammatory response. The optimal therapy for SARS is still not clear,
as there were no randomized controlled trials conducted. Treatment with interferon
(IFN)-a, steroid, protease inhibitors (such as lopinavir) together with ribavirin,
or convalescent plasma containing neutralising antibody, could all be useful.
Prophylaxis with IFN or hyperimmunoglobulin may also be considered as post-

exposure prophylaxis [34].
SARS CoV is identified as a zoonosis with a natural reservoir in Chinese
horseshoe bats [35]. Its emergence is associated with local culinary practice in
southern China, leading to captured palm civets acting as the amplifying host and
passing on infection to human. As long as the reservoirs and amplifying hosts
coexist, there is a potential for SARS to re-emerge. Intensivists should always be
on the lookout for patients with unexplained severe respiratory infections and
consider SARS as a possible differential diagnosis.
20.2.6 Varicella Zoster Virus: Chickenpox
Primary varicella zoster virus (VZV) (family Herpesviridae) infection causes

chickenpox. This is a common self-limiting childhood infection characterised by a
mild fever and a generalised vesicular rash. Risk factors for severe disease include
immunosuppression, smoking and pregnancy. Complications include bacterial
sepsis, pneumonia, encephalitis, ataxia, toxic shock, necrotising fasciitis and
haemorrhagic chickenpox with disseminated coagulopathy and fatality [36].
Chickenpox is highly infectious and can be transmitted via inhalation of
respiratory secretions or by direct contact. Patients are likely to be infective 48 h
before the appearance of the rash until the last lesion has crusted over. Outbreaks
in the ICU have frequently been reported [37, 38]. Infected patients should be
promptly isolated, preferably in negative-pressure rooms.
A rapid diagnosis of chickenpox can be made by electron microscopy or
immunofluorescence of scrapings from the vesicle base. A person who has had
chickenpox does not develop chickenpox again, but the virus may reactivate as
zoster/shingles. Susceptibility to chickenpox can be determined by testing for the
presence of VZV immunoglobulin (Ig)G. Infected patients need to be isolated
immediately, and exposed patients and staff investigated. Exposed staff who are
susceptible to VZV should be excluded from contact with high-risk patients for
821 days postexposure. Susceptible individuals at risk of severe disease should
receive varicella-zoster immunoglobulin (VZIG) prophylaxis, which could be
given up to 10 days after exposure.
Neonates born to mothers who developed chickenpox 7 days before to 7 days
after delivery are highly susceptible due to a lack of protective maternal anti-
bodies. In such cases, VZIG prophylaxis to the neonate is recommended. The baby
should also be isolated. Intravenously administered acyclovir should be started
promptly at the first sign of illness. Most childhood chickenpox does not require
treatment. However, in severe cases (e.g. pneumonitis, disseminated disease with
visceral involvement and patients requiring hospitalisation), intravenously
administered acyclovir (10 mg/kg 8 hourly) is the treatment of choice. Treatment
of neonates will require a higher dose (20 mg/kg 8 hourly). A live attenuated
vaccine against VZV is available. Susceptible health care workers should be
immunised.
20.3 Viruses Transmitted by the FaecalOral Route
20.3.1 Rotavirus
Rotavirus (family Reoviridae) is highly infectious and a significant cause of

nosocomial gastroenteritis, particularly in children \5 years of age. Patients
present with sudden onset of fever, vomiting, abdominal pain and watery diar-
rhoea. Due to the high viral shedding in the faeces, a diagnosis can be easily
obtained using antigen-detection enzyme-linked immunosorbent assay (ELISA) or
electron microscopy.
In temperate climates the infection is seasonal with peaks in winter, and hos-
pital outbreaks often coincide with outbreaks in the community. In Europe, it was
found that 4963% of paediatric nosocomial gastroenteritis was positive for
rotavirus, with an incidence of 12.3 per 1,000 hospital days, leading to prolonged
hospitalisation between 1.5 and 4.5 days [39]. Very sick infants with gastroen-
teritis may require intensive care and could, in turn, be the source of nosocomial
infection in ICU. Premature and very low birth weight infants (\1,500 g) are
particularly at risk, as severe complications such as necrotising enterocolitis and
intestinal perforation are commonly reported. A Dutch study found that amongst
all nosocomially acquired viral infections in NICUs, 10% were due to rotavirus,
which demonstrates the importance of this infection in the ICU setting [40].
Nosocomial rotavirus infections in adults have also been reported and occa-
sionally cause serious complications in the elderly and immunosuppressed
patients. Nosocomial transmission has been previously associated with ungloved
nasogastric feeding, contaminated toys, shortage of nurses, overcrowding and high
patient turnover. Adherence to effective infection control measures (hand hygiene,
enteric precautions; Table 20.3), as well as adequate staffing and patient cohorting/
isolation can therefore help prevent or manage an outbreak [41]. The recently
developed rotavirus vaccine could substantially reduce the incidence of nosoco-
mial infections [42].
20.3.2 Norovirus
Norovirus (family Caliciviridae) is the most common cause of nosocomial out-

breaks of gastroenteritis. Symptoms typically comprise profuse diarrhoea and
projectile vomiting. The diagnosis can be confirmed by ELISA, RT-PCR or
electron microscopy of stool samples. Noroviruses are highly infectious and are
usually transmitted by direct contact via the faecaloral route or via oropharyngeal
exposure to aerosolised vomit. A number of outbreaks have recently been
described in NICUs involving mainly premature neonates, some of whom devel-
oped necrotising enterocolitis. Neonates and immunocompromised patients can
shed the virus for a prolonged time over months, which emphasises the need
for rigorous adherence to effective infection control measures (Table 20.3).
Additional measures such as increased hand hygiene and wiping of floors and
Table 20.3 General measures to control outbreak of viral gastroenteritis [41]
Hand washing (liquid soap) or decontamination (aqueous antiseptic/alcohol based-hand rub)

(A). However, alcohol-based products are known to be less effective against nonenveloped
viruses, for which hand washing with soap and water is preferred (B)
Wear disposable gloves and aprons when contact with stool or vomitus is likely (B)
Isolate symptomatic individuals (particularly with uncontrolled diarrhoea, incontinence, and
children) (B)
Avoid unnecessary movement of patients to unaffected areas (B)
Staff working in affected areas must not work in unaffected areas within 72 h (B)
Exclude symptomatic staff members from duty until symptom free for 72 h (B)
If a large number of patients is involved and no further isolation facilities are available, close
the unit to new admissions or transfers until 72 h after the last new case (B)
Terminal cleaning of the environment, using freshly prepared hypochlorite (1,000 ppm) on hard
surfaces (B)
Caution visitors and emphasise hand hygiene (B)
Categorisation of recommendations: A strongly recommended for all hospitals and strongly
supported by well-designed experimental or epidemiological studies; B strongly recommended
for all hospitals and viewed as effective by experts in the field
incubators with agents active against caliciviruses have been proven to be

particularly useful in controlling outbreaks in NICU wards [43].
20.3.3 Enteroviruses and Parechoviruses
Both enteroviruses and parechoviruses (family Picornaviridae) have numerous

subtypes. Enteroviruses include polioviruses, coxsackieviruses, echoviruses and
other numbered enteroviruses. There are as many as 14 types of human parech-
oviruses [44]. Parechovirus type 3, in particular, can cause severe infection in
young infants [45].
Both viruses are significant causes of nosocomial infections, particularly in the
NICU. Enterovirus outbreaks involving up to 23 neonates have been reported [46],
and an attack rate of 29% was reported. Enterovirus infections can present as
neonatal sepsis, meningoencephalitis, myocarditis, hepatitis or gastroenteritis.
Necrotising enterocolitis with pneumatosis intestinalis is a known complication in
neonates. Some enteroviruses, such as Enterovirus 71, can cause severe and fatal
illness in older children. Parechoviruses can cause meningoencephalitis [47] and a
sepsis syndrome in young infants [48]. Enteroviruses and parechoviruses are
genetically distinct from each other and require a different RT-PCR for diagnosis.
Sequencing of the gene encoding the VP1 region of the virus has been used to
identify outbreak strains.
With the global polio eradication programme, poliomyelitis is no longer a
common nosocomial infection, although health care workers in the ICU who may
be in contact with live vaccine poliovirus shedding infants should ensure that they
are immunised.
Rigorous hand washing (Table 20.3) is the most important measure during an
outbreak. Cohort nursing, source isolation and screening are other measures fre-
quently used (Table 20.3). Clearance of the virus by the host is antibody-mediated
and many have advocated the use of normal human immunoglobulin (NHIG).
20.3.4 Hepatitis A Virus
Hepatitis A virus (family Picornaviridae) belongs to the same family as entero-

viruses and is usually transmitted via the faecaloral route. Nosocomial trans-
mission of hepatitis A virus is well documented. An outbreak in an adult ICU
(AICU) occurred as a result of inadequate precautions taken while handling bile of
a patient not suspected of incubating hepatitis A [49]. Most other outbreaks
occurred in PICUs or NICUs, with attack rates varying between 15 and 25%. Risk
factors for outbreaks have been attributed to handling soiled bed pads, nappies or
gowns of an index patient, failure to wash hands, and eating in the ICU. In the
NICU, vertical transmission and blood transfusion have been implicated as the
cause of infection in the index case. The effect of nosocomial hepatitis A infection
varies from asymptomatic to classic presentation with acute hepatitis. Diagnosis is
by serological detection of hepatitis-A-specific IgM. The use of molecular tech-
niques such as RT-PCR can help identify early infection or in difficult cases, such
as those with immunodeficiency. Sequencing of PCR products is useful in
establishing epidemiological linkage during outbreaks. NHIG has been success-
fully used for postexposure prophylaxis to control outbreaks. There is now
increasing evidence that hepatitis A vaccine can be used for prophylaxis if the
contact occurs within 14 days from onset of illness in the index case [50].
20.4 Viruses Transmitted by Blood and Body Fluid
The most commonly encountered nosocomial blood-borne viruses are hepatitis B

virus (HBV), hepatitis C virus (HCV) and human immunodeficiency virus (HIV).
The main risks are transmission from patients to health care workers. However,
transmissions between patients and from health care workers to patients have been
reported. The best way to prevent occupational exposure of blood-borne viruses is to
practice universal precautions. Blood and body fluids (Table 20.4) from any patient,
whether or not there are identifiable risk factors, should be considered as a potential
risk. This encourages good and safe practice and helps prevent unnecessary acci-
dents. Physical isolation of patients with blood-borne virus infection is generally not
necessary unless there is profuse uncontrolled bleeding. Infection-control teams and
occupational health departments should adopt a proactive approach to educate and
prevent sharps injury (Table 20.5). There should also be specific instructions on how
to deal with blood and body fluid exposure (Table 20.6).
Table 20.4 Body fluids that may pose a risk for hepatitis B and C virus (HBV, HCV) and human
immunodeficiency virus (HIV) after significant exposure
Amniotic fluid
Breast milk
Cerebrospinal fluid
Exudate from burns or skin lesions
Pericardial fluid
Peritoneal fluid
Pleural fluid
Saliva after dental treatment
Synovial fluid
Unfixed tissues or organs
Any other fluid if visibly blood stained
Saliva, urine, vomitus or stool that are not blood stained are not considered as high risk for blood
borne viruses
20.4.1 Hepatitis B Virus
HBV is the most infectious of the three common blood-borne viruses. The risk of
transmission depends on the viral load of the source patient. An HBV-infected
individual with hepatitis B e antigen (HBeAg) tends to have a high viral load
and is therefore more infectious than carriers without HBeAg. Estimate of infec-
tivity ranges from 2% (HBeAg absent) to 40% (HBeAg present). All health care
workers should be immunised against HBV. Exposed health care workers who are
susceptible (not immunised or vaccine nonresponders) should receive hepatitis B
immunoglobulin for postexposure prophylaxis. A booster dose of vaccine should
be given to those exposed individual who had previously been successfully
immunised.
20.4.2 Hepatitis C Virus
HCV is probably the commonest blood-borne virus encountered in Western

countries. In the UK over a 3-year period, 462 incidences of occupational exposure
to HCV were reported in comparison with 293 of HIV and 151 of HBV [51].
Follow-up studies of health care workers who sustained a percutaneous exposure
to blood from a patient known to have HCV infection have reported an average
incidence of seroconversion of 1.8% (range 07%). No vaccine or postexposure
prophylaxis was available to prevent HCV transmission. Early diagnosis is
essential, as early interferon treatment after seroconversion has a high success rate
for eradication [52]. Exposed health care workers should be followed up at 6 and
Table 20.5 Example of points for sharps safety education (adapted from the Infection Control
Team of Guys and St. Thomas NHS Foundation Trust, London)
Never resheath needles

Dispose all sharps in an approved container
Dispose sharps at the point of use
If carried, sharps must be placed in a tray
Do not overfill sharps binsdispose of sharps bins when three-fourths full
Table 20.6 Example of actions to be taken immediately after a blood/body fluid exposure
(adapted from the Infection Control Team of Guys and St. Thomas NHS Foundation Trust,
London)
Rinse area thoroughly under running water

If a wound, wash site with soap, encourage the wound to bleed, and apply a waterproof dressing
(if appropriate)
Report it immediately to:
Your supervisor (or line manager)
Occupational health or accident and emergency
Document incident by completing an incident form with source name and details
12 weeks for HCV RNA testing and promptly referred for treatment if found
infected.
20.4.3 Human Immunodeficiency Virus
The average risk of HIV transmission after percutaneous exposure to HIV-infected

blood is about 0.3%. After mucocutaneous exposure, the risk is estimated to be
\0.1%. A case-control study [53] identified four factors with increased risk of
transmission:
deep injury;
visible blood on the device that caused the injury;
injury with a needle that has been placed in a source patients artery or vein;
terminal HIV-related illness in the source patient.
This study also showed that the use of zidovudine prophylaxis reduce the risk of
transmission by 80%. Postexposure prophylaxis (PEP) should therefore be offered
to all health care workers who have significant exposure to blood or body fluid
from a patient known to be at high risk of or to have HIV infection. Various PEP
options are available depending on national recommendations. This should be
started as soon as possible after exposure and continued for 4 weeks.
Table 20.7 Viruses responsible for viral haemorrhagic fevers with nosocomial concern in the
intensive care unit
Virus Geographic distribution Reservoir Vector

Marburg and Sub-Sahara Africa Bats None
Ebola
Rift Valley fever Mainland Africa Sheep, cattle Mosquito
Lassa West Africa Rodents None
Crimean Congo East, West and South Africa, North and Central Cows, hares Ticks
haemorrhagic Asia, Middle East, India and Pakistan, Balkans, birds,
fever West China hedgehogs
20.4.4 Viral Haemorrhagic Fevers
Viral haemorrhagic fevers (VHFs) are severe and life-threatening diseases caused
by a range of viruses. They are either zoonotic or arthropod-borne infections and
are often endemic in certain parts of the world. They are often highly infectious
through close contact with infected blood and body fluid and therefore pose a
significant risk of hospital-acquired infection. As many patients with VHF
present with shock and require vigorous supportive treatment, it is a potential
problem in the ICU. The major viruses of nosocomial concern in this setting are
Marburg, Ebola, Rift Valley fever, Lassa and Crimean Congo haemorrhagic
fever (Table 20.7). The incubation period for these VHFs ranges from
321 days. Initial symptoms are often nonspecific but may eventually lead to
haemorrhage and shock. Any febrile patient who has returned from an endemic
area of one of the VHF agents or has a history of contact with cases suspected to
have VHF within 3 weeks should be considered as at risk. However, malaria
should always be excluded. A risk assessment needs to be performed, and any
patient known or strongly suspected to be suffering from VHF should be
admitted to a high-security infectious disease unit that is designed to manage
these patients. While awaiting transfer to a secure unit, such patients should be
placed in a negative-pressure room with strict source isolation. Specimens for
patient management should be processed in a high-security laboratory designated
for category 4 pathogens, and the aetiological agent established using PCR,
serology and virus culture. All areas and materials in contact with infected
patients should be autoclaved, incinerated or treated with hypochlorite
(10,000 ppm of available chlorine). If the patient dies, the body should be placed
in a sealable body bag sprayed or wiped with hypochlorite. Individuals who have
been in contact with a case of VHF should be put under surveillance for
3 weeks. The successful i.v. use of ribavirin has been reported in some cases of
VHFs (Lassa, Crimean Congo haemorrhagic fever and Hantaan). Apart from
yellow fever, no vaccines are available.
20.5 Viruses Transmitted by Direct Contact
20.5.1 Varicella Zoster Virus; Shingles
Shingles or zoster is the result of the reactivation of latent VZV (family Herpes-
viridae) in the dorsal root or cranial nerve ganglia. The clinical presentation is a
painful vesicular eruption covering the affected dermatome. The clinical diagnosis
can be confirmed rapidly by immunofluorescence, electron microscopy or PCR of
the cellular material obtained from a vesicular scraping. The infection is usually
self-limiting but can be more severe in immunocompromised patients, in whom it
may present over multiple dermatomes or as a disseminated infection. The latter
cases should be managed as if they were chickenpox, and respiratory precautions
for infection control have to be enforced.
Patients or health care staff members with classic shingles are contagious from
the day the rash appears until the lesions are crusted over. There is some risk of
nosocomial transmission if the lesions are on exposed areas of the body or in
immunocompromised infected patients. Nonimmune (VZV-IgG negative) patients
or health care staff members with no history of chickenpox are susceptible if
they have close contact with shingles and should be managed as described for
chickenpox contact.
20.5.2 Herpes Simplex Virus
The herpes simplex virus (HSV) (family Herpesviridae) consist of two types:
HSV-1 and HSV-2. Clinically, they most commonly manifest with oral (mainly
HSV-1) or genital (mainly HSV-2) ulcerations/vesicles, and reactivation is com-
mon, particularly in the ICU. Other presentations include keratitis, encephalitis,
meningitis, herpetic whitlow or neonatal infection.
The diagnosis can be confirmed rapidly by immunofluorescence, electron
microscopy or PCR of vesicle/ulcer scrapings. In the immunocompromised
patient, HSV can cause life-threatening disseminated infection and, early treatment
with acyclovir i.v. is recommended. It has also been suggested that occult herpes
virus reactivation may increase the mortality risk of ICU patients [54].
As the infected lesions contain virus, there is an increased risk of nosocomial
transmission until the lesions have crusted over. Standard isolation precautions
should be in place to reduce transmission (Table 20.2). Patients with active lesions
should be nursed away from high-risk patients (i.e. immunocompromised, severe
eczema, burns, or neonates). As patients can be asymptomatic secretors, health
care workers should wear gloves when dealing with mucosal secretions (i.e. saliva)
to avoid infections such as herpetic whitlow. Infected staff should cover lesions if
possible and should not attend those at risk.
Neonatal herpes is usually transmitted from mother to the child at the time of
delivery and may not be noticed until the infant develops the disease. Universal
precautions, in particularly, hand washing, should always be in place to reduce
transmission of infection. To contain or prevent an outbreak, infected cases should

be cohorted and nursed by dedicated staff who will not attend noninfected infants.
20.5.3 Rabies Virus
Rabies virus (family Rhabdoviridae) is usually transmitted to humans following

exposure to saliva of a rabid animal (e.g. dog, fox, bat) via a bite or scratch, but
only 40% of exposed people develop disease. The virus spreads from the wound to
the central nervous system causing fatal encephalitis, and the virus may be present
in the patients saliva, skin, eye, and brain tissue. The diagnosis can be confirmed
by demonstrating the virus directly in brain tissue or saliva by RT-PCR or by
immunofluorescence detection of antigen in skin biopsies from the nape of the
neck. Due to the severe and paralysing effect, patients may be admitted to the ICU.
To date, no case of nosocomial transmission has been reported apart from two
patients who received corneal transplants from infected donors. Suspected or
proven cases should be placed in standard isolation and appropriate precautions
taken when dealing with potential infectious secretions (e.g. wearing of mask if
dealing with oral secretions). Any health care worker with a significant exposure
(e.g. splash of secretion onto mucosa or broken skin) should receive rabies vaccine
and specific immunoglobulin.
20.6 Summary
Viral infection can cause significant morbidity and mortality and has the potential
to result in cross infection, involving patients as well as health care workers. Good
infection-control practice is essential to prevent nosocomial infection. Intensivists
should be on the alert for important viruses causing infections according to age
group of patients and mode of transmission and should never be complacent. Good
liaison with the laboratory is essential for determining correct diagnostic tests and
timely report of results to help in patient management.
References
1. Fitzgerald DA (2009) Human swine influenza A. Paediatr Respir Rev 10:154158
2. Trifonov V, Khiabanian H, Rabadan R (2009) Geographic dependence, surveillance, and
origins of the 2009 influenza A (H1N1) virus. N Engl J Med 361:115119
3. Webb SA, Pettila V, Seppelt I et al (2009) Critical care services and 2009 H1N1 influenza in
Australia and New Zealand. N Engl J Med 361:19251934
4. Itoh Y, Shinya K, Kiso M et al (2009) In vitro and in vivo characterization of new swine-
origin H1N1 influenza viruses. Nature 460:10211025
5. Writing committee of the WHO consultation on clinical aspects of pandemic (H1N1) 2009
Influenza (2010) Clinical aspects of pandemic 2009 Influenza A (H1N1) virus infection.
N Engl J Med 362:17081719
6. Siston AM, Rasmussen SA, Honein MA et al (2010) Pandemic 2009 influenza A(H1N1)
virus illness among pregnant women in the United States. JAMA 303:15171525
7. Campbell A, Rodin R, Kropp R et al (2010) Risk of severe outcomes among patients

admitted to hospital with pandemic (H1N1) influenza. CMAJ 182:349355
8. Hanslik T, Boelle PY, Flahault A (2010) Preliminary estimation of risk factors for admission
to intensive care units and for death in patients infected with A(H1N1)2009 influenza virus,
France, 20092010. PLoS Curr. 2010 March 9; 2:RRN1150
9. Anonymous (2009) Bacterial coinfections in lung tissue specimens from fatal cases of 2009
pandemic influenza A (H1N1)United States, MayAugust 2009. MMWR Morb Mortal
Wkly Rep 58:10711074
10. Anonymous (2009) Evaluation of rapid influenza diagnostic tests for detection of novel
influenza A (H1N1) virusUnited States, 2009. MMWR Morb Mortal Wkly Rep 58:826829
11. Ellis J, Iturriza M, Allen R et al (2009) Evaluation of four real-time PCR assays for detection
of influenza A (H1N1) viruses. Euro Surveill 14(22): pii: 19230
12. Flagg A, Danziger-Isakov L, Foster C et al (2010) Novel 2009 H1N1 influenza virus infection
requiring extracorporeal membrane oxygenation in a pediatric heart transplant recipient.
J Heart Lung Transpl 29:582584
13. Christian MD, Joynt GM, Hick JL et al (2010) Critical care triage. Recommendations and
standard operating procedures for intensive care unit and hospital preparations for an
influenza epidemic or mass disaster. Intensive Care Med 36(Suppl 1):S55S64
14. Jain S, Kamimoto L, Bramley AM et al (2009) Hospitalized patients with 2009 H1N1
influenza in the United States, AprilJune 2009. N Engl J Med 361:19351944
15. Taylor BL, Montgomery HE, Rhodes A, Sprung CL (2010) Protection of patients and staff
during a pandemic. Recommendations and standard operating procedures for intensive care
unit and hospital preparations for an influenza epidemic or mass disaster. Intensive Care Med
36(Suppl 1):S45S54
16. Cheng VC, Tai JW, Wong LM et al (2010) Prevention of nosocomial transmission of swine-
origin pandemic influenza virus A/H1N1 by infection control bundle. J Hosp Infect 74:271277
17. Reiche J, Schweiger B (2009) Genetic variability of group A human respiratory syncytial
virus strains circulating in Germany from 1998 to 2007. J Clin Microbiol 47:18001810
18. Berger TM, Aebi C, Duppenthaler A, Stocker M (2009) Prospective population-based study
of RSV-related intermediate care and intensive care unit admissions in Switzerland over a
4-year period (20012005). Infection 37:109116
19. Caram LB, Chen J, Taggart EW et al (2009) Respiratory syncytial virus outbreak in a long-
term care facility detected using reverse transcriptase polymerase chain reaction: an argument
for real-time detection methods. J Am Geriatr Soc 57:482485
20. Gadsby NJ, Hardie A, Claas EC, Templeton KE (2010) Comparison of the Luminex RVP
Fast assay with in-house real-time PCR for respiratory viral diagnosis. J Clin Microbiol
48:22132216
21. Lavergne V, Ghannoum M, Weiss K et al (2011) Successful prevention of respiratory
syncytial virus nosocomial transmission following an enhanced seasonal infection control
program. Bone Marrow Transpl 46(1):137142
22. Morris SK, Dzolganovski B, Beyene J, Sung L (2009) A meta-analysis of the effect of antibody
therapy for the prevention of severe respiratory syncytial virus infection. BMC Infect Dis 9:106
23. Falsey AR, Walsh EE (2000) Respiratory syncytial virus infection in adults. Clin Microbiol
Rev 13:371384
24. Boeckh M, Englund J, Li Y et al (2007) Randomized controlled multicenter trial of
aerosolized ribavirin for respiratory syncytial virus upper respiratory tract infection in
hematopoietic cell transplant recipients. Clin Infect Dis 44:245249
25. Chavez-Bueno S, Mejias A, Merryman RA et al (2007) Intravenous palivizumab and
ribavirin combination for respiratory syncytial virus disease in high-risk pediatric patients.
Pediatr Infect Dis J 26:10891093
26. Terlizzi ME, Massimiliano B, Francesca S et al (2009) Quantitative RT real time PCR and
indirect immunofluorescence for the detection of human parainfluenza virus 1, 2, 3. J Virol
Methods 160:172177
27. Maziarz RT, Sridharan P, Slater S et al (2010) Control of an outbreak of human parainfluenza
virus 3 in hematopoietic stem cell transplant recipients. Biol Blood Marrow Transpl
16:192198
28. Nichols WG, Corey L, Gooley T et al (2001) Parainfluenza virus infections after
hematopoietic stem cell transplantation: risk factors, response to antiviral therapy, and
effect on transplant outcome. Blood 98:573578
29. Lenaerts L, De Clercq E, Naesens L (2008) Clinical features and treatment of adenovirus
infections. Rev Med Virol 18:357374
30. Williams KM, Agwu AL, Dabb AA et al (2009) A clinical algorithm identifies high risk
pediatric oncology and bone marrow transplant patients likely to benefit from treatment of
adenoviral infection. J Pediatr Hematol Oncol 31:825831
31. Ksiazek TG, Erdman D, Goldsmith CS et al (2003) A novel coronavirus associated with
severe acute respiratory syndrome. N Engl J Med 348:19531966
32. Poutanen SM, Low DE, Henry B et al (2003) Identification of severe acute respiratory
syndrome in Canada. N Engl J Med 348:19952005
33. Lau EH, Hsiung CA, Cowling BJ et al (2010) A comparative epidemiologic analysis of
SARS in Hong Kong, Beijing and Taiwan. BMC Infect Dis 10:50
34. Wong SS, Yuen KY (2008) The management of coronavirus infections with particular
reference to SARS. J Antimicrob Chemother 62:437441
35. Cheng VC, Lau SK, Woo PC, Yuen KY (2007) Severe acute respiratory syndrome coronavirus
as an agent of emerging and reemerging infection. Clin Microbiol Rev 20:660694
36. Cameron JC, Allan G, Johnston F et al (2007) Severe complications of chickenpox in
hospitalised children in the UK and Ireland. Arch Dis Child 92:10621066
37. Aly NY, Al Obaid I, Al-Qulooshi N, Zahed Z (2007) Occupationally related outbreak of
chickenpox in an intensive care unit. Med Princ Pract 16:399401
38. Apisarnthanarak A, Kitphati R, Tawatsupha P et al (2007) Outbreak of varicella-zoster virus
infection among Thai healthcare workers. Infect Control Hosp Epidemiol 28:430434
39. Fruhwirth M, Heininger U, Ehlken B et al (2001) International variation in disease burden of
rotavirus gastroenteritis in children with community- and nosocomially acquired infection.
40. Verboon-Maciolek MA, Krediet TG, Gerards LJ et al (2005) Clinical and epidemiologic
characteristics of viral infections in a neonatal intensive care unit during a 12-year period.
41. Boyce JM, Pittet D (2002) Guideline for hand hygiene in health-care settings.
Recommendations of the healthcare infection control practices Advisory Committee and
the HIPAC/SHEA/APIC/IDSA Hand Hygiene Task Force. Am J Infect Control 30:S1S46
42. Cunliffe NA, Both JA, Lowe SJ et al (2010) Healthcare-associated viral gastroenteritis among
children in a large pediatric hospital, United Kingdom. Emerg Infect Dis 16:5562
43. Armbrust S, Kramer A, Olbertz D et al (2009) Norovirus infections in preterm infants: wide
variety of clinical courses. BMC Res Notes 2:96
44. Nix WA, Maher K, Pallansch MA, Oberste MS (2010) Parechovirus typing in clinical
specimens by nested or semi-nested PCR coupled with sequencing. J Clin Virol 48:202207
45. Harvala H, Robertson I, Chieochansin T et al (2009) Specific association of human
parechovirus type 3 with sepsis and fever in young infants, as identified by direct typing of
cerebrospinal fluid samples. J Infect Dis 199:17531760
46. Takami T, Sonodat S, Houjyo H et al (2000) Diagnosis of horizontal enterovirus infections in
neonates by nested PCR and direct sequence analysis. J Hosp Infect 45:283287
47. Gupta S, Fernandez D, Siddiqui A et al (2010) Extensive white matter abnormalities associated
with neonatal Parechovirus (HPeV) infection. Eur J Paediatr Neurol 14(6):531534
48. Benschop KS, Schinkel J, Minnaar RP et al (2006) Human parechovirus infections in Dutch
children and the association between serotype and disease severity. Clin Infect Dis 42:204210
49. Hanna JN, Loewenthal MR, Negel P, Wenck DJ (1996) An outbreak of hepatitis A in an
intensive care unit. Anaesth Intensive Care 24:440444
50. Victor JC, Monto AS, Surdina TY et al (2007) Hepatitis A vaccine versus immune globulin
for postexposure prophylaxis. N Engl J Med 357:16851694
51. Evans B, Duggan W, Baker J et al (2001) Exposure of healthcare workers in England, Wales,
and Northern Ireland to bloodborne viruses between July 1997 and June 2000: analysis of
surveillance data. BMJ 322:397398
52. Maheshwari A, Thuluvath PJ (2010) Management of acute hepatitis C. Clin Liver Dis
14:169176
53. Cardo DM, Culver DH, Ciesielski CA et al (1997) A case-control study of HIV
seroconversion in health care workers after percutaneous exposure. Centers for Disease
Control and Prevention Needlestick Surveillance Group. N Engl J Med 337:14851490
54. Cook CH, Martin LC, Yenchar JK et al (2003) Occult herpes family viral infections
are endemic in critically ill surgical patients. Crit Care Med 31:19231929
AIDS Patients in the ICU
21
F. E. Arancibia and M. A. Aguayo
21.1 Introduction
In the early 1980s in the United States, the first medical reports described
outbreaks of Kaposis sarcoma and Pneumocystis carinii (now P. jiroveci)
pneumonia in homosexual men [1, 2]. These reports rate respiratory infections by
opportunistic germs and rare tumors affecting healthy young men. High mortality
rates were observed that were most likely caused by acute respiratory failure
(ARF). Additionally, each report had abnormal ratios of lymphocyte subgroups.
Subsequent reports described an escalating frequency of unusual infections and
tumors, suggesting a profound state of immune suppression in many homosexual
men, injection-drug users, sexual partners of infected persons, hemophiliacs who
had received blood transfusions, and children.
In early 1983, virologists at the Pasteur Institute first isolated the human immu-
nodeficiency virus (HIV) [3], a retrovirus that infects cells of the immune system
(subgroup of CD4+ T cells), destroying or impairing their function. The most
advanced stage of HIV infection was called AIDS. In the early years, the ICU
survival rate of patients with AIDS was low [4]. Based on the belief that ICU care of
patients with AIDS was futile, clinical, ethical, and economic issues were raised
regarding the benefits and burdens of the critical care of these patients. However, the
use of P. carinii pneumonia (PCP) prophylaxis, antiretroviral therapy, and cortico-
steroids for PCP has changed outcomes. Several studies have shown improved
survival rates and costs of HIV-infected patients admitted to the ICU [58].
F. E. Arancibia (&)
Unidad de Cuidados Intensivos,
Instituto Nacional Trax, Santiago, Chile
354 F. E. Arancibia and M. A. Aguayo
Up to 3560% of persons with underlying HIV are unaware of their HIV

infection at the time of ICU admission [9]. According to other authors, this is due
to the fact that the HIV epidemic has two distinct populations of patients. First,
there are patients with access to care and to the full armamentarium of HIV-related
drugs. In contrast to this patient population is the sizable number of individuals
who are either unaware of their HIV status or who lack access to care. These
patients continue to appear in emergency rooms and the ICU with the same
opportunistic infections that were seen in the 1980s.
HIV is the worlds leading infectious killer, with an estimated 2 million AIDS
deaths occurring in 2008 alone. The global HIV burden remains enormous. At the
end of 2008, an estimated 33.4 million people worldwide were living with HIV.
Africa, Asia, and America had the highest incidence. That same year, some
2.7 million people became newly infected with the virus [10].
21.2 Causes for ICU Admission
In the United States and Europe, from 4 to 11% of HIV patients hospitalized
require admission to the ICU. All studies show that respiratory failure has
remained the most common indication for an ICU admission [5, 8, 11, 12].
Other frequent indications are sepsis and neurological compromise (Table 21.1).
Occasionally, patients with severe gastrointestinal symptoms, cardiomyopathies
and/or complications such as lactic acidosis and pancreatitis may require intensive
care. Finally, these patients may have clinical or surgical conditions requiring ICU
admission unrelated to their HIV/AIDS. HIV-infected patients admitted to the ICU
following trauma, elective surgery, or gastrointestinal bleeding may have as good a
prognosis as patients without HIV infection [9].
However, opportunistic illness rates declined precipitously after the intro-
duction of highly active antiretroviral therapy (HAART) and stabilized at low
levels in the subsequent years. Buchacz et al. [13] reported that during
19942007, rates of opportunistic infections (per 1,000 person-years) decreased
from 89 to 13.3 and rates of opportunistic malignancies were from 23.4 to 3.0
(Fig. 21.1). Due to the dramatic improvement in the prognosis of HIV-infected
patients, our notion of HIV as a disease has transformed from a rapidly fatal
diagnosis to a treatable chronic condition.
The etiologies of respiratory failure are diverse and include: (1) PCP, initially
responsible for a high burden of disease and low survival rates; later studies
demonstrated improved survival, even among PCP patients; (2) bacterial pneu-
monia is more frequent in patients with HIV; (3) cytomegalovirus (CMV) is a
minor cause. Also, patients with other associated infections, such as tuberculosis
(TB) and fungal infection; or associated malignancy, such as Kaposis sarcoma
and non-Hodgkins lymphoma; or immune reconstitution syndrome (Fig. 21.2),
may develop respiratory failure during disease evolution [9, 14]. The patients with
ARF due to P. jiroveci and other opportunistic agents can be treated with
21 AIDS Patients in the ICU 355
Table 21.1 General characteristics, intensive care unit admission diagnosis, mechanical venti-
lation, and mortality rates of HIV-infected patients
Nickas Casalino Morris Powell

19921995 [11] 19951999 [8] 19961999 [5] 20002004 [12]
(n = 394) (n = 426) (n = 354) (n = 311)
Age, years, 38 (2064) 38.5 41.7 (2367) 44 (2472)
mean (range)
Male, sex 357 (91) 338 (79.3) 285 (80.5) 235 (79.7)
Respiratory 203 (46.8) 134 (31.4) 144 (40.7) 131 (42)
failure
Sepsis 54 (12.2) 58 (13.6) 42 (11.9) 62 (20)
Pneumocystis 264 (67) 52 (12.2) 36 (10.7) 43 (14)
pneumonia
Neurologic 49 (11.1) 108 (25.4) 44 (12.4) 51 (16)
Cardiac 35 (7.9) 23 (5.4) 35 (9.9)
Mechanical 245 (55) 81 (19) 191(54) 205 (67.7)
ventilation
HAART use 205 (52) 230 (54) 89 (25) 101 (33)
Mortality 264 (67) 98 (23) 103 (29) 96 (31)
HAART highly active antiretroviral therapy
Values are given as Number (%), unless otherwise indicated
Fig. 21.1 Incidence of AIDS-defining opportunistic infections and malignancies, the HIV
infected patient study, 19942007 [13]
noninvasive ventilation (NIV). The use of NIV improves gas exchange and avoids
intubation and mechanical ventilation. Patients who failed with the use of NIV or
present severe ARF still require intubation and mechanical ventilation.
Fig. 21.2 CD4+ cell count ranges for selected HIV-related respiratory illnesses. CMV
Cytomegalovirus; MAC Mycobacterium avium complex [18, 21, 23, 28, 31, 37, 38, 43, 47,
4951, 54, 57]
The global mortality ratios in hospitalized patients infected with HIV requiring
ICU admission went from 23 to 67% [48, 11, 12], most frequently due to
respiratory failure and septic shock. The death of many HIV-infected patients has
been linked directly to late diagnosis and initiation of appropriate antiviral therapy.
Several clinical factors are related to the outcome of AIDS patients in the ICU.
Predictors of increased mortality risk include the need for mechanical ventilation,
and disease severity [high Acute Physiology and Chronic Health Evaluation
(APACHE) II score or simplified acute physiological score (SAPS), modified
multisystem organ failure (MSOF) score, CD4 count, pneumothorax, presence of
cardiovascular instability, and low levels of serum albumin] [14, 15]. Mortality
rates of HIV-infected patients and the need for mechanical ventilation early in the
epidemicprior to 1988reached 82% [16]. However, the use of protective
ventilation (lower tidal volume ventilation) is associated with reduced mortality
rates in HIV-infected patients with acute lung injury and respiratory failure [17].
21.3 Pneumocystis carinii Pneumonia
PCP remains the most prevalent opportunistic infection in patients infected with
the HIV and is often the AIDS-defining illness, occurring most frequently when
the T-helper cell count (CD4+) is \200 cells/ll. First identified as a protozoan
nearly 100 years ago and reclassified as a fungus in 1988, named P. carinii but
renamed P. jiroveci, has a unique tropism for the lung, where it exists primarily as
Fig. 21.3 Chest radiograph of a 62-year-old woman who is unaware of her HIV status. Diffuse
bilateral reticular pattern with foci of consolidation in the left lower lobe. Pneumocystis jiroveci
was detected in bronchoalveolar fluid and transbronchial biopsy
an alveolar pathogen without invading the host. In rare cases, pneumocystis dis-
seminates in the setting of severe underlying immunosuppression or overwhelming
infection [14, 18].
At the beginning of the AIDS epidemic, around 75% of patients developed at
least one episode of PCP, with even higher mortality rates of 3585% in patients
requiring admission to an ICU. With the use of prophylaxis, the incidence of PCP
decreased from 47 to 25% [19]. Prophylaxis against P. jiroveci pneumonia can be
safely discontinued in patients with HIV infection who have had a positive
response to HAART (indicated by a CD4 count [200 cells/ll) with minimal risk
of recurrent P. jiroveci pneumonia [20].
Common symptoms of PCP include the subtle onset of progressive dyspnea,
nonproductive cough, and low-grade fever. Physical examination typically reveals
tachypnea, tachycardia, and normal findings on lung auscultation. It is typically
subacute with a clinical course of days or weeks. Acute dyspnea with pleuritic
chest pain may indicate the development of a pneumothorax [14, 18, 21].
Typical radiographic features of PCP are diffuse bilateral interstitial infiltrates
involving the entire lung or the lower lung fields [18, 22] (Fig. 21.3). Less com-
mon findings include localized infiltrates, upper-lobe infiltrates, solitary or mul-
tiple nodules, and pneumatoceles. Pleural effusions and thoracic lymphadenopathy
are rare. Approximately 6% of patients can develop spontaneous pneumothoraces
during the course of their illness. High-resolution computed tomography, which
is more sensitive than chest radiography, may reveal extensive ground-glass
Fig. 21.4 Transbronchial biopsy specimen stained with Grocotts of an adult woman showing
typical pneumocystis cyst forms of 56 lm (courtesy of Manuel Meneses M.D)
attenuation or cystic lesions. This greater organism burden results in a higher

diagnostic yield of induced sputum to confirm PCP in patients with AIDS. If the
initial specimen of induced sputum is negative for pneumocystis, then bronchos-
copy with bronchoalveolar lavage (BAL) should be performed. Transbronchial or
surgical lung biopsy is rarely needed [14, 18, 21] (Fig. 21.4).
PCP may be difficult to diagnose owing to nonspecific symptoms and signs, the
use of prophylactic drugs in HIV-infected patients, and simultaneous infection
with multiple organisms (such as CMV) in an immunocompromised host.
Although an elevated serum lactate dehydrogenase (LDH) level has been noted in
patients with PCP, it is likely to be a reflection of the underlying lung inflammation
and injury rather than a specific marker for the disease [23].
Since the beginning of the AIDS epidemic, respiratory failure caused by
P. jiroveci has been the most common indication for ICU admission among
patients with HIV infection (Table 21.1). However, the proportion of ICU
admissions caused by respiratory failure has declined [16, 18, 24]. Patients with
ARF due to P. jiroveci can be treated with NIV. In one study, NIV failed only in
13% of HIV-positive patients with severe PCP [25]. Confalonieri et al. [26]
compared early NIV in patients with AIDS and PCP with matched controls treated
with conventional mechanical ventilation. The mortality rate was 38 versus 100%,
respectively. Also, NIV reduced the incidence of pneumothorax.
Patients with HIV infection who also have acute respiratory distress syn-
drome (ARDS) requiring mechanical ventilation should receive such therapy
according to the ARDS Network guidelines with the use of low tidal volumes
and plateau pressures [17, 27]. The application of these guidelines is especially
crucial in patients with PCP because of the frequent presence of pneumatoceles
associated with the infection and the resulting increased risk of pneumothorax
during mechanical ventilation. The presence of pneumothorax is an inde-
pendent risk factor for death in patients with HIV infection and PCP (15).
Trimethoprim/sulfamethoxazole remains the preferred treatment. Other options

could be primaquine plus clindamycin, atovaquone, and pentamidine [18, 28].
Early adjunctive corticosteroid therapy to suppress lung inflammation in
patients with severe PCP and hypoxemia (partial pressure of arterial oxygen
while the patient is breathing room air is \70 mm Hg or the alveolararterial
gradient is [35) reduces the risks of death in patients with AIDS [29].
In addition, Morris et al. reported an improvement in survival with HAART in
HIV-infected patients with severe PCP [30].
In the 1980s and early to mid-1990s, PCP was responsible for a high burden of
the disease, and survival rates were low; later studies demonstrated improved
survival rates. Among most patients with AIDS and PCP, the mortality rate is
1020% during the initial infection, but the rate increases substantially with the
need for mechanical ventilation [18, 25, 29, 30].
21.4 Bacterial Respiratory Infections
21.4.1 Bacterial Pneumonia
Early in the HIV epidemic, researchers noted that bacterial pneumonia (BP) was a
common cause of morbidity. BP is an important cause of morbidity and mortality
in patients with HIV infection and is at least five times more frequent in HIV-
infected patients compared with healthy individuals. In the precombination ART
era, the HIV Infection Study reported the incidence of BP ranged was 3.97.3
episodes per 100 person-years. Since the introduction of ART, a reduction in the
risk for BP has been observed [21, 31, 32]. BP is still among the most common
causes of respiratory failure resulting in ICU admission [32] and might be the first
manifestation of underlying HIV infection. BP can occur at any stage during HIV
disease and at any CD4+ T-cell count, but it is substantially more frequent among
those with \200 CD4+ T-cell counts. Other risk factors include drug use intra-
venously, previous bacterial infection or PCP, smoking, a low socioeconomic
status, alcohol abuse, comorbidities (including cardiovascular disease, renal dis-
ease, and hepatic cirrhosis), and malnutrition [3133].
The microbiologic cause of community-acquired BP identified most frequently
in HIV-infected patients, are Streptococcus pneumoniae and Haemophilus species
[21]. S. pneumoniae is the most common causative agent and is frequently asso-
ciated with bacteremic disease [31, 33]. The rate of pneumococcal bacteremia is
higher in patients with than without HIV infection [33]. Patients with HIV
infection are at increased risk for infection with penicillin and cotrimoxazole-
resistant S. pneumoniae, and identifying this microorganism could lead to changes
in patient management [21, 23, 33].
H. influenzae, both the encapsulated and nonencapsulated types, is also com-
mon. Pseudomonas aeruginosa pneumonia in some studies has been reported as a
common pulmonary complication [33], especially in patients with low leukocyte
and CD4+ T-cell counts and ill enough to require ICU admission. Also, there is a
growing number of literature reports about the occurrence of pneumonia due to
Staphylococcus aureus, especially oxacillin-resistant strains, is this population.
Atypical pathogens (Mycoplasma, Chlamydia, and Legionella) seem not to play a
significant role in HIV-infected patients. Rare causes of pneumonia presenting
with cavitation are Rhodococcus equi and Nocardia asteroides [21, 23, 33].
The clinical and radiographic presentation of BP does not differ substantially
for HIV-infected compared with HIV-uninfected patients. Compared with
P. jiroveci pneumonia and other opportunistic infections of lung, the onset of
fever and other symptoms is more abrupt and the patients is more likely to
experience a productive cough and pleuritic chest pain. In contrast, patients with
low CD4 cell count, who are at an increased risk of BP, often present an atypical
clinical picture with milder symptoms and signs, especially when liver cirrhosis
is also present. The white blood cell count is usually elevated in persons with
BP, and a left shift also might be present. Radiographic features typically include
unilateral, focal, segmental, or lobar consolidation. Also, HIV-infected persons
might present with multifocal or multilobar involvement and with parapneu-
monic effusions [3133]. The American Thoracic Society (ATS) severity criteria
developed to assess community-acquired pneumonia (CAP) in patients not
infected with HIV have been found to be valid also for HIV-infected patients
with bacterial CAP [33].
Prompt and accurate diagnosis is essential, because the outcome of HIV-
associated BP appears to be reasonably good with appropriate treatment. Usu-
ally, collection of specimens for microbiologic studies should be performed
before the initiation of antibiotic therapy. An etiologic diagnosis is obtained in
an average of 35% of cases with standard culture methods. In such conditions,
urinary antigen test for S. pneumoniae identification may help in reaching a rapid
and etiologic diagnosis [31]. However, antibiotic therapy should be administered
promptly, without waiting for the results of diagnostic testing. Guidelines for
managing CAP in persons without HIV infection also apply to HIV-infected
persons [34]. Persons with severe pneumonia who require intensive care should
be treated with an IV beta-lactam plus either azithromycin intravenously or an
IV respiratory fluoroquinolone. If risk factors for P. aeruginosa or S. aureus
infection are present, empiric therapy to cover these pathogens should be con-
templated [28].
In all patients presenting with antimicrobial treatment failure, a regular
microbial reinvestigation is mandatory in order to find potentially life-threatening
etiologies. Given the increased incidence of Mycobacterium tuberculosis in HIV-
infected persons, the diagnosis of TB should always be suspected in those with
pneumonia. Also, noninfectious causes with pulmonary dysfunction should be
considered [21, 28, 31].
BP mortality rate is high, and some studies may reach 30%. Factors associated
with increased mortality in HIV patients with BP include the presence of septic
shock, radiologic progression of infiltrates, and CD4 counts \100 cells/ll.
21.4.2 Nosocomial Pneumonia
Nosocomial pneumonia (NP) appears to be more common in patients with

AIDS as a result of the degree of immunosuppression, prior use of antibiotics,
and exposure to invasive procedures. Although underestimated, NP is asso-
ciated with a higher morbidity and mortality rate. NP is most frequently a
complication of mechanical ventilation. Improved ART options in developed
countries resulted in a decreased hospitalization rate of HIV-infected indi-
viduals and the incidence of NP. A study of surveillance of HIV-infected
inpatients showed an NP incidence decreasing from 13.9/10,000 patient hos-
pital days between 1994 and 1996 to 5.6/10,000 patient hospital days between
1997 and 1998 [35].
NP in HIV-infected patients is usually caused by Gram-negative bacilli and
S. aureus, but fungal, viral, and tuberculin infections causes must also be
considered. Among atypical agents, Legionella pneumonia in HIV-infected
patients should be hospital acquired. Clinical and microbiological surveys in
HIV-positive patients have found that P. aeruginosa is a frequent agent,
accounting for 1667% of nosocomial pneumonias. Empiric antibiotics in HIV-
infected patients with suspected hospital-acquired pneumonia should cover
potentially multi-drug-resistant organisms such as P. aeruginosa, Klebsiella
pneumonia, Acinetobacter spp., as well as methicillin-resistant S. aureus
[21, 23, 35, 36].
21.4.3 Bacteremia Infection
The importance of bacterial infections that complicate the clinical course of

patients with HIV infection has been recognized since the beginning of the AIDS
epidemic. HIV-infected patients have an increased risk of bacteremia compared
with the general population. The presence of bloodstream infection is associated
with an increased mortality rate, length of hospital stay, and ICU admission rate.
Bacteremia infections are responsible for the immediate cause of death of up to
30% of patients with HIV infection [37].
Several risk factors predispose for bacteremia infections among HIV-infected
patients. These include the presence of neutropenia, use of central venous cathe-
ters, low CD4+ lymphocyte count, and IV drug use. The common sources of
bloodstream infection in patients with HIV infection include the lungs, skin,
subcutaneous tissue, and intravascular catheters.
Most bacteremias are community acquired. The most common Gram-positive
organism isolated from the bloodstream of patients with HIV is S. pneumoniae,
followed of S. aureus; the most common Gram-negative organisms are Escherichia
coli, Salmonella spp., and Pseudomonas spp. The annual incidence of pneumococcal
bacteremia is estimated to be as high as 940/100,000 patients with AIDS [21].
However, similar to other infections, the incidence of bacteremia has been declining
since the introduction of HAART. Hospital mortality rates of HIV-positive patients

with bloodstream infection range from 9 to 54%, but mortality rates are higher for
patients with bacterial sepsis.
21.5 Viral Pneumonias
CMV is the most frequent viral pneumonia seen in persons with HIV infection.
Although CMV is often detected in BAL fluid, documented CMV pneumonia is
rare and occurs only in severely immunosuppressed patients with CD4 cell counts
\50/ll [38]. Some studies suggest that CMV in BAL fluid reflects bronchopul-
monary replication of the virus. Although the majority of patients with CMV
pneumonia have additional forms of pulmonary pathology, CMV is the only
causative agent frequently identified in patients with severe pulmonary disease.
Due to the high coinfection rate with P. jiroveci, in cases of PCP treatment failure
and severe immunosuppression, the main differential diagnosis must be established
with CMV. Some authors believe that it represents a preterminal phenomenon in
advanced AIDS [21].
Criteria for establishing that CMV is the cause of pneumonitis and pulmonary
dysfunction have been difficult to establish. Clinical features are nonproductive
cough, fever, progressive dyspnea, hypoxemia, and diffuse interstitial infiltrates
[23]. Respiratory symptoms are typically present for 24 weeks. Physical exam-
ination of the chest may be normal or may reveal crackles or evidence of pleural
effusion. The chest radiographic findings of CMV pneumonia vary and include
reticular or ground-glass opacities, alveolar infiltrates, and nodules or nodular
opacities. Pleural effusions may be seen as well. The latter finding may be helpful
in distinguishing CMV pneumonia from P. jiroveci, in which pleural effusions are
rare. Persons suspected of having CMV pneumonia should undergo a careful
dilated retinal examination by an experienced ophthalmologist. Definitive diag-
nosis of CMV pneumonia requires demonstration of cytopathic inclusions and
widespread specific cytopathic changes in the lungs. Confirming the diagnosis is
often not easy due to the typically extremely serious condition, making it difficult
to perform a lung biopsy. Autopsy studies revealed that patients with AIDS and
CMV pneumonia were successfully diagnosed antemortem in only 1324% of
cases. New techniques using in situ DNA hybridization or monoclonal antibodies
to detect the virus may improve the diagnostic yield of less invasive procedures,
such as bronchoalveolar lavage [39].
When suspected CMV pulmonary disease occurs, therapy must be initiated
immediately. Ganciclovir and foscarnet or cidofovir have been used to treat CMV
pneumonia [28], although few data establish that such therapy affects outcome.
Ganciclovir appears to be less effective against pulmonary infections than against
retinitis or gastrointestinal disease, with response rates of 5060%. Despite the
monolithic use of ganciclovir for CMV-related illness, reports of CMV-resistant
strains have been mostly limited to long-term usage in patients with HIV infection.
The combination of ganciclovir plus foscarnet may be useful in the setting of

ganciclovir-resistant CMV disease [28, 39].
Pneumonitis due to herpes simplex, varicellazoster, and respiratory syncytial
viruses has occasionally been reported in AIDS patients. These viruses are of
practical importance due to the availability of effective treatment [9]. Also, data
from several studies suggest that EpsteinBarr virus (EBV) has a role in pneumo-
nitis. EBV DNA and proteins have been detected in pulmonary lesions from children
with HIV and lymphoid interstitial pneumonitis [40]. The role of influenza and
adenoviruses in causing HIV-related pulmonary complications could be important
during outbreaks of these infections. Patients with HIV infection frequently
developed complications or severe illness with 2009 H1N1 virus infection [41].
21.6 Mycobacteriosis
21.6.1 Tuberculosis
Around the world, TB remains an important public health problem, especially in

developing countries. Since the emergence of AIDS, TB and HIV infections have
been intimately connected. HIV probably increases susceptibility to infection with
M. tuberculosis and the risk of progression to TB. Also, TB appears to accelerate the
course of HIV disease [21]. By the end of 2000, approximately 11.5 million HIV-
infected people worldwide were coinfected with M. tuberculosis. Seventy percent of
coinfected people were in sub-Saharan Africa and 20% in Southeast Asia [42]. TB is
the most common opportunistic disease and the most common cause of death in HIV
patients in developing countries [43]. Severe TB requiring ICU care is rare but
commonly known to be of markedly bad prognosis (Fig. 21.5). The most common
reasons for ICU admission are the development of ARDS and severe organ failure,
such as renal failure. ARF caused by TB necessitating mechanical ventilation has
been associated with high mortality rates. Several studies also reported a critical
course in patients with miliary TB and HIV infection.
The clinical presentation of HIV-infected patients without pronounced immu-
nodeficiency (e.g., CD4+ count[350 cells/ll) with TB is usually similar to that in
HIV-negative cases. However, with progressive immunodeficiency, disseminated
disease and extrapulmonary involvement occurs more frequently. Weight loss and
fever are more common in HIV-positive pulmonary TB patients than in those who
are HIV negative. Conversely, cough and hemoptysis are less common in HIV-
positive pulmonary TB patients than in those who are HIV negative. This is
probably because there is less cavitation, inflammation, and endobronchial irrita-
tion in HIV-positive patients. The physical signs in patients with pulmonary TB
are nonspecific. The most commonly reported extrapulmonary sites of disease are
the lymph nodes, pleura, pericardium, meninges, and genitourinary system. Typ-
ical radiographs are seen in only one-third of patients, and they usually have a CD4
count[200 cells/ll [43, 44]. In advanced HIV disease, chest radiographic findings
Fig. 21.5 A 27-year-old man with advanced HIV infection and highly active antiretroviral
treatment. Chest radiographic (a) and high-resolution computed tomography scans (b) showing
peribronchial thickening, nodularity, and septal lines. Two months later (c), progression of lung
involvement to multiple nodular infiltrates, confluent lesions, and acute respiratory failure; video
thoracoscopic biopsy showed Kaposis sarcoma
of pulmonary TB commonly include lower lobe, middle lobe, interstitial, and

miliary infiltrates; cavitation is less common [28].
The first screening test for suspected pulmonary TB is sputum, bronchial aspirate
secretions, or BAL sample for acid-fast bacilli smear and culture. The chances of
finding TB bacilli are greater with three respiratory samples than with fewer samples
[44]. For patients with signs of extrapulmonary TB, needle aspiration or tissue biopsy
of skin lesions, lymph nodes, or pleural or pericardial fluid should be performed.
Mycobacterial blood cultures might be helpful for patients with signs of dissemi-
nated disease or worsening immunodeficiency [28, 43].
HIV-infected persons who adhere to standard regimens of treatment for TB do
not have an increased risk of treatment failure or relapse. Treatment of drug-
susceptible TB should include a 6-month regimen with an initial phase of isoni-
azid, rifampin or rifabutin, pyrazinamide, and ethambutol administered for
2 months, followed by isoniazid and rifampin (or rifabutin) for 4 additional
months. However, studies have found that HIV-seropositive patients are more
likely to develop acquired drug resistance than are seronegative cases. Patients
with TB caused by drug-resistant (especially multi-drug-resistant) organisms
should be treated with specialized regimens containing second-line anti-TB drugs.
At least four drugs to which the organisms are known or presumed to be sus-
ceptible should be used, and treatment should be given for at least 18 months.
Delayed clinical suspicion and treatment of active pulmonary TB with ARF may
contribute to the persistently high mortality rates in ICU patients with these dis-
eases [28, 43, 44]. The optimal timing for initiating ART in patients with HIV and
TB coinfection remains unclear. Despite World Health Organization (WHO)
guidelines supporting concomitant treatment of the two diseases (TB/HIV), the
initiation of antiretroviral therapy is often deferred until completion of TB therapy
because of concern about potential drug interactions between rifampin and some
classes of antiretroviral drugs, immune reconstitution inflammatory syndrome, and
overlapping side effects. Abdool Karim et al., in a prospective and randomized
trial, found that the initiation of ART during TB therapy in patients with confirmed
TB and HIV coinfection reduced mortality rates by 56% [45].
21.6.2 Mycobacterium avium
M. avium complex (MAC) infection is common in patients with AIDS. In contrast

to the experience in nonimmunocompromised hosts, in whom clinical manifes-
tations are primarily pulmonary, MAC causes disseminated infection, often with
documented mycobacteremia in patients with AIDS. In the absence of effective
HAART or chemoprophylaxis in those with AIDS-associated immunosuppression,
the incidence of disseminated MAC disease is 2040% [28]. Since early 1980s,
MAC infection has been detected at autopsy in [50% of patients dying from
AIDS, and in one study, infection due to MAC was detected antemortem in 44%,
with blood culture being the most sensitive diagnostic means [46]. In HIV patients,
MAC typically occurs among persons with CD4 counts \50 cells/ll, suggesting
that specific T-cell products or activities are required for mycobacterial resistance
[42]. Natural history studies of persons with AIDS in the pre-ART era showed that
almost 40% of patients with \50 CD4+ T cells/ll developed disseminated MAC
within 1 year. Other factors associated with increased susceptibility to MAC are
high plasma HIV RNA levels ([100,000 copies/ml), previous opportunistic
infections, previous colonization of the respiratory or gastrointestinal tract with
MAC, and reduced in vitro lymphoproliferative immune responses to M. avium
antigens, possibly reflecting defects in T-cell repertoire.
Disseminated infection is associated with high mortality rates, especially in
those with CD4 counts \100 cells/mm3. It primarily affects the gastrointestinal
tract and lungs and manifests with fever, cough, abdominal pain, diarrhea, and
weight loss. A significant proportion of total body burden is on MAC inside
macrophages, and this distribution has implications for drug treatment and,
therefore, on drug susceptibility testing. A confirmed diagnosis of disseminated
MAC disease is based on compatible clinical signs and symptoms coupled with
MAC isolation from cultures of blood, lymph node, bone marrow, or other nor-
mally sterile tissue or body fluids [28, 42, 46]. Species identification should be
performed using specific DNA probes, high-performance liquid chromatography,
or biochemical tests.
The treatment is always a combination of three active drugs according to
severity. Drugs may include ethambutol, a rifamycin (rifampin or rifabutin), and a
macrolide (mainly clarithromycin); other possibilities include aminoglycosides
(amikacin) and fluoroquinolones [28, 42].
21.7 Mycoses
21.7.1 Cryptococcosis
Fungal pneumoniasother than PCPoccur in patients with HIV infection but

are not common in most geographic areas. The era of effective ART has led to a
marked reduction in opportunistic infections in countries where such therapies are
available. Opportunistic fungal infections are no exception, and the incidence of
such infections is now 2025% of that seen in the mid-1990s [47]. However,
fungal opportunistic infections remain significant causes of morbidity and mor-
tality in persons with HIV in developing countries. After TB and P. jiroveci
pneumonia, cryptococcosis was the third most common opportunistic infection
reported in Thailand [48].
Cryptococcus neoformans is the most common fungal pulmonary infection in
patients with AIDS and usually coexists with cryptococcal meningitis. Also,
cryptococcal pneumonia may be underdiagnosed and not recognized until dis-
semination. The majority of cases are observed in patients who have CD4 counts
\100 cells/ll. When pulmonary infection is present, symptoms and signs include
cough, fever, and dyspnea in association with an abnormal chest radiograph [28].
The radiographic manifestation includes a diffuse reticular or reticulonodular
pattern that resembles PCP, lobar or segmental consolidation, or multiple nodules
that have a propensity to cavitate. Disseminated disease can occur and manifest as
a miliary pattern that may be associated with lymphadenopathy or pleural effusion.
ARF occurring as a complication of cryptococcosis (including pulmonary
infection) was initially thought to be uncommon, with only a handful of case
reports. Visnegarwala et al. [49] documented ARF as occurring in 29 of 210 cases
of AIDS-associated cryptococcosis (13.8%). The clinical presentation was iden-
tical to that of PCP. Independent predictors of ARF were black race, LDH level
C500 IU/L, presence of interstitial infiltrates, and cutaneous lesions. ARF with
cryptococcosis in AIDS patients is associated with disseminated disease and high
mortality rates.
Diagnosis frequently is not defined before death. Serum cryptococcal antigen
testing is a sensitive and rapid screening method in diagnosing cryptococcosis in
HIV-infected patients. Also, routine blood cultures are useful [50]. The recom-
mended initial standard treatment is amphotericin B deoxycholate combined with
flucytosine or fluconazole [28].
21.7.2 Aspergillosis
Aspergillosis is a life-threatening fungal infection in immunocompromised people,

including HIV patients. Invasive pulmonary aspergillosis (IPA) is a relatively
uncommon but devastating infection in patients with advanced AIDS [51] and was
more common before the advent of HAART. The overall incidence was reported
as being 3.5 cases per 1,000 person-years among HIV-infected patients [52]. The
infection is most frequently caused by Aspergillus fumigatus, although certain
cases are caused by A. flavus, A. niger, and A. terreus. A low CD4 count, generally
\100/ll, was present in almost all cases of AIDS-associated aspergillosis.
Coexistent neutropenia or use of corticosteroids occurred in about 50% of patients;
the remaining cases appear to have no other risk factors other than advanced AIDS
[28, 51, 52].
Symptoms of invasive aspergillosis pneumonia are fever, cough, dyspnea, chest
pain, hemoptysis, and hypoxemia; chest radiograph might demonstrate a diffuse,
focal, or cavitary infiltrate. A halo of low attenuation surrounding a pulmonary

nodule or an air crescent on CT scan of the lung is suggestive of disease [53].
Isolation of an Aspergillus spp. from respiratory secretions has poor predictive
value for invasive disease in AIDS patients. Bronchoscopy with BAL is, however,
a safe and useful tool in high-risk patients suspected of having IPA. In addition to
obtaining samples for fungal stain and culture, detecting antigens in the BAL fluid
may be helpful, as well as excluding other infections. Transbronchial biopsies
usually do not add much to the IPA diagnosis and are associated with increased
risk of bleeding, so are seldom performed. Histopathological diagnosis, by
examining lung tissue obtained by thoracoscopic or open-lung biopsy, remains the
gold standard for diagnosing invasive IPA [28, 51, 53]. However, a large pro-
portion of cases of aspergillosis are diagnosed postmortem, suggesting that
underdiagnosis antemortem may contribute to poor survival.
Treatment of aspergillosis in the HIV-infected population has not been exam-
ined systematically. The recommended treatment for invasive aspergillosis in
patients without HIV infection is voriconazole [28]. Amphotericin B deoxycho-
late, lipid-formulation amphotericin B, and caspofungin and posaconazole are
other alternatives; other echinocandins, such as micafungin and anidulafungin, are
reasonable alternatives. Response to therapy tends to be particularly poor in this
patient population [52]. Other fungi infections, such as Histoplasma capsulatum
and Coccidioides immitis, are less frequent but are present in several regions where
the disease is endemic. Their presence in the lung is often indicative of dissemi-
nated disease and is associated with significant mortality rates.
21.8 Neoplastic Disease
Kaposis sarcoma is a well-recognized cause of pulmonary disease in patients

with HIV. This was first described in healthy, young, homosexual men, in whom
it involved lymph nodes, viscera, mucosa, and skin [2]. Human herpes virus 8
(HHV-8) infection was shown to be the causative viral agent [54]. The variant
epidemic, AIDS-associated Kaposis sarcoma, can progress in weeks or months,
and median patient survival is months. However, HAART development has
influenced the clinical course and reduced the incidence of Kaposis sarcoma,
but it remains the most common AIDS-associated cancer in the United States
[55, 56].
When Kaposis sarcoma occurs in the lung, imaging features include interstitial
or nodular parenchymal opacities. Characteristic peribronchovascular nodule
distribution is frequent, and coalescence of nodules is common in late-stage dis-
ease (Fig. 21.5). Pleural effusion and lymphadenopathy may also be present
[55, 56].
Diagnosis is often anticipated by concurrent skin lesions and the presence of
prominent lesions in the tracheobronchial tree, which are easily recognized by
bronchoscopy. Definitive diagnosis is not easy to establish. Transbronchial
Fig. 21.6 Open lung biopsy specimen shows proliferation of capillaries in a background of
spindle-shaped tumor cells in the lung interstitium, which are characteristic histological features
of Kaposis sarcoma (hematoxylin and eosin) (courtesy of Manuel Meneses M.D)
biopsies of the bronchus or lung parenchyma have a high risk of hemorrhage and
reveal crush artifacts difficult to distinguish from Kaposis sarcoma. On cytology,
there is no diagnostic feature. Thus, tissue must be obtained on either open lung
biopsy or video-assisted thoracoscopy (Fig. 21.6), or a presumptive diagnosis must
be made when Kaposis sarcoma in seen in the tracheobronchial tree and bron-
choalveolar lavage reveals no other likely pathogens. Often, there is an associated
bloody pleural effusion when thoracentesis is performed [9, 56].
Pulmonary Kaposis sarcoma can respond well to chemotherapy [57]. HAART
and opportunistic infection prophylaxis has contributed to the success rates of
management strategies.
Lymphoma continues to be a cause of pulmonary disease. Although primary
central nervous system (CNS) lymphomas have greatly diminished in frequency
among patients treated with HAART; primary B-cell lymphomas elsewhere con-
tinue to occur. Patchy pulmonary infiltrates have been well described. Biopsy or
cytology is needed to establish a diagnosis. Combination chemotherapy for HIV-
associated lymphoma has become impressively more successful when HAART is
continued with opportunistic infection prophylaxis. Stem cell transplantation has
also been used successfully.
As patients are now living longer, and experience with large patient populations
has increased, other pulmonary neoplastic processes have been recognized that
clinicians should be aware of. Primary effusion cell lymphoma can present in the
pleural, pericardial, or abdominal cavities as effusions. This HHV-8 and EBV-
associated tumor is diagnosed by cytology in many cases. It is not clear how
effective chemotherapy is for this tumor.
21.9 Immune Reconstitution Inflammatory Syndrome
During the first few months of HAART, immune reconstitution may be compli-
cated by clinical events in which either previously subclinical infections are found
or preexisting partially treated opportunistic infections deteriorate. This condition,
termed immune reconstitution inflammatory syndrome (IRIS), is thought to be
caused by improvement in the hosts immune response to pathogens [58]. The
inflammatory response may be such that the patient develops ARF and requires
ICU. Abdool Karim et al. [45] reported that the incidence of IRIS was 9.5%.
However, the study found the incidence was higher in the integrated-therapy group
(anti-TB and ART) than in the sequential-therapy group: 12.4 versus 3.8%,
respectively. The term IRIS is most commonly used for mycobacterial infections
(TB and disseminated MAC disease) but is also used for other opportunistic
infections, including P. jirovecii pneumonia, toxoplasmosis, hepatitis B and C
viruses, CMV, varicellazoster virus, cryptococcal infection, and histoplasmosis
[28]. The syndrome is manifested as paradoxical worsening of the underlying
respiratory disease and occurs days to months after HAART initiation. However,
IRIS usually develops within the 48 weeks following HAART initiation and is
caused by an exuberant inflammatory response to pneumocystis or mycobacterial
antigens. On the basis of current knowledge, it is tempting to hypothesize that the
immunological basis of IRIS is a HAART-induced rapid clonal expansion and
redistribution of M. tuberculosis-specific memory T cells, which drives a dereg-
ulated immune activation [59] and a cytokine storm [60]. Antigen load could be
responsible for the overvigorous inflammatory response of a recovering immune
system.
Diagnosing IRIS requires excluding other causes of respiratory decompensa-
tion. Clinical presentations are transient worsening or appearance of new symp-
toms and signs, such as fever, increasing chest radiographic infiltrate, peripheral
and mediastinal lymphadenopathy, or changes in radiographic manifestations.
Studies of TB-associated IRIS indicate that this complication is rarely fatal and
that severe episodes can be successfully managed with corticosteroids [45].
Patients with severe cases are able to continue ART.
References
1. Centers for Disease Control and Prevention (1981) Pneumocystis pneumoniaLos Angeles.
MMWR Morb Mortal Wkly Rep 30:250252
2. Centers for Disease Control and Prevention (1981) Kaposis sarcoma and Pneumocystis
pneumonia among homosexual menNew York City and California. MMWR Morb Mortal
Wkly Rep 30:305308
3. Barr-Sinoussi F, Chermann JC, Rey F et al (1983) Isolation of a T-lymphotropic retrovirus
from a patient at risk for acquired immune deficiency syndrome (AIDS). Science 220:
868871
4. Schein RMH, Fischl MA, Pitchenik AE et al (1986) ICU survival of patients with acquired
immunodeficiency syndrome. Crit Care Med 14:10261027
5. Morris A, Creasman J, Turner J et al (2002) Intensive care of human immunodeficiency

virus-infected patients during the era of highly active antiretroviral therapy. Am J Respir Crit
Care Med 166:262267
6. Vincent B, Timsit JF, Auburtin M et al (2004) Characteristics and outcomes of HIV-infected
patients in the ICU: impact of the highly active antiretroviral treatment era. Intensive Care
Med 30:859866
7. Narasimhan M, Posner AJ, DePalo VA et al (2004) Intensive care in patients with HIV
infection in the era of highly active antiretroviral therapy. Chest 125:18001804
8. Casalino E, Wolff M, Ravaud P et al (2004) Impact of HAART advent on admission patterns
and survival in HIV-infected patients admitted to an intensive care unit. AIDS 18:14291433
9. Masur H (2006) Management of patients with HIV in the intensive care unit. Proc Am Thorac
Soc 3:96102
10. Joint United Nations Programme on HIV/AIDS (UNAIDS) and World Health Organization
(2009) 2009 AIDS epidemic update. http://data.unaids.org/pub/Report/2009/JC1700_Epi_
Update_2009_en.pdf
11. Nickas G, Wachter RM (2000) Outcomes of intensive care for patients with human
immunodeficiency virus infection. Arch Intern Med 160:541547
12. Powell K, Davis JL, Morris AM et al (2009) Survival for patients with HIV ADMITTED to
the ICU continues to improve in the current era of combination antiretroviral therapy. Chest
135:1117
13. Buchacz K, Baker RK, Palella FJ Jr et al (2010) AIDS-defining opportunistic illnesses in US
patients, 19942007: a cohort study. AIDS 24:15491559
14. Huang L, Quartin A, Jones D et al (2006) Intensive care of patients with HIV infection.
N Engl J Med 355:173181
15. Chernilo S, Trujillo S, Kahn M et al (2005) Lung disease among HIV infected patients admitted
to the Instituto Nacional del Trax in Santiago, Chile. Rev Med Chile 133:517524
16. Curtis JR, Yarnold PR, Schwanrtz DN et al (2000) Improvements in outcomes of acute
respiratory failure for patients with human immunodeficiency virus related Pneumocystis
carinii pneumonia. Am J Respir Crit Care Med 162:393398
17. Davis JL, Morris A, Kallet RH et al (2008) Low tidal volume ventilation is associated with
reduced mortality in HIV-infected patients with acute lung injury. Thorax 63:988993
18. Thomas CF, Limper AH (2004) Pneumocystis pneumonia. N Engl J Med 350:24872498
19. Muoz A, Schreger LK, Bacellar H et al (1993) Trends in the incidence of outcomes defining
acquired immunodeficiency syndrome (AIDS) in the multicenter AIDS cohort study:
19851991. Am J Epidemiol 137:423427
20. Ledergerber B, Mocroft A, Reiss P et al (2001) Discontinuation of secondary prophylaxis
against pneumocystis carinii pneumonia in patients with HIV infection who have a response
to antiretroviral therapy. N Engl J Med 344:168174
21. Speich R (2006) HIV, AIDS, and the lung. In: Torres A, Ewig S, Mandell L, Woodhead M
(eds) Respiratory infections. Hodder Arnold, London, pp 719738
22. DeLorenzo LJ, Huang CT, Maguire GP, Stone DJ (1987) Roentgenographic patterns of
Pneumocystis carinii pneumonia in 104 patients with AIDS. Chest 91:323327
23. Erbelding EJ, Chaisson RE (2001) Respiratory infections in persons infected with human
immunodeficiency virus. In: Niederman MS, Sarosi GA, Glassroth J (eds) Respiratory
infections. Lippincott Williams & Wilkins, Philadelphia, pp 251264
24. Alves C, Nicols JM, Mir JM et al (2001) Reappraisal of the aetiology and prognostic
factors of severe acute respiratory failure in HIV patients. Eur Respir J 17:8793
25. Monnet X, Vidal-Petiot E, Osman D et al (2008) Critical care management and outcome of
severe Pneumocystis pneumonia in patients with and without HIV infection. Critical Care
12:R28. doi:10.1186/cc6806
26. Confalonieri M, Calderini E, Terraciano S et al (2002) Noninvasive ventilation for treating
acute respiratory failure in AIDS patients with Pneumocystis jiroveci pneumonia. Intensive
Care Med 28:12331238
27. Acute Respiratory Distress Syndrome Network (2000) Ventilation with lower tidal volumes
as compared with traditional tidal volumes for acute lung injury and the acute respiratory
distress syndrome. N Engl J Med 342:13011308
28. Kaplan JE, Benson C, Holmes KH (2009) Guidelines for prevention and treatment of
opportunistic infections in HIV-infected adults and adolescents: recommendations from
CDC, the National Institutes of Health, and the HIV medicine association of the infectious
diseases society of America. MMWR Recomm Rep 58(RR-4):1207
29. Bozzette SA, Sattler FR, Chiu J (1990) A controlled trial of early adjunctive treatment with
corticosteroids for Pneumocystis carinii pneumonia in the acquired immunodeficiency
syndrome. N Engl J Med 323(21):14511457
30. Morris A, Wachter RM, Luce J et al (2003) Improved survival with highly active antiretroviral
therapy in HIV-infected patients with severe Pneumocystis carinii pneumonia. AIDS 17:7380
31. Madeddu G, Fiori ML, Mura MS (2010) Bacterial community-acquired pneumonia in HIV-
infected patients. Curr Opin Pulm Med 16:201207
32. Wolff AJ, ODonnell AE (2001) Pulmonary manifestations of HIV infection in the era of
highly active antiretroviral therapy. Chest 120:18881893
33. Cordero E, Pachn J, Rivero A et al (2000) Community-acquired bacterial pneumonia in
human immunodeficiency virus-infected patients validation of severity criteria. Am J Respir
34. Mandell LA, Wunderink RG, Anzueto A et al (2007) Infectious diseases society of
American/American thoracic society consensus guidelines on the management of
community-acquired pneumonia in adults. Clin Infect Dis 44(Suppl 2):S27S72
35. Tumbarello M, Tacconelli E, de Gaetano Donati K et al (2001) Nosocomial bacterial
pneumonia in human immunodeficiency virus infected subjects: incidence, risk factors and
outcome. Eur Respir J 17:636640
36. Petrosillo N, Nicastri E, Viale P (2005) Nosocomial pulmonary infections in HIV-positive
patients. Curr Opin Pulm Med 11:231235
37. Afessa B, Morales I, Weaver B (2001) Bacteremia in hospitalized patients with human
immunodeficiency virus: a prospective, cohort study. BMC Infect Dis 1:13
38. Millar AB, Patou G, Miller RF et al (1990) Cytomegalovirus in the lungs of patients with
AIDS. Respiratory pathogen or passenger. Am Rev Respir Dis 141(6):14741477
39. Drew WL (2007) Laboratory diagnosis of cytomegalovirus infection and disease in
immunocompromised patients. Curr Opin Infect Dis 20:408411
40. Cohen JI (2000) EpsteinBarr virus infection. N Engl J Med 343:481488
41. Writing Committee of the WHO Consultation on Clinical Aspects of Pandemic (H1N1) 2009
Influenza (2010) Clinical aspects of pandemic 2009 influenza A (H1N1) virus infection.
N Engl J Med 362:17081719
42. Griffith DE, Aksamit T, Brown-Elliott BA et al (2007) An official ATS/IDSA statement:
diagnosis, treatment, and prevention of nontuberculous mycobacterial diseases. Am J Respir
43. Small PM, Fujiwara PI (2001) Management of tuberculosis in the United States. N Engl J
Med 345:189200
44. Harries A, Maher D, Graham S (2004) TB/HIV: a clinical manual, 2nd edn. Publications of
the World Health Organization, Geneva
45. Abdool Karim SS, Naidoo K, Grobler A et al (2010) Timing of initiation of antiretroviral
drugs during tuberculosis therapy. N Engl J Med 362:697706
46. Wallace JM, Hannah JB (1988) Mycobacterium avium complex infection in patients with the
acquired immunodeficiency syndrome. A clinicopathologic study. Chest 93:926932
47. Clark TA, Hajjeh RA (2002) Recent trends in the epidemiology of invasive mycoses. Curr
Opin Infect Dis 15:569574
48. Chariyalertsak S, Supparatpinyo K, Sirisanthana T et al (2002) A controlled trial of
itraconazole as primary prophylaxis for systemic fungal infections in patients with advanced
human immunodeficiency virus infection in Thailand. Clin Infect Dis 34:277284
49. Visnegarwala F, Graviss EA, Lacke CE (1998) Acute respiratory failure associated with
cryptococcosis in patients with AIDS: analysis of predictive factors. Clin Infect Dis 27:
12311237
50. Feldman CH (2003) Cryptococcal pneumonia. Clin Pulm Med 10:6771
51. Segal BH, Walsh TJ (2006) Current approaches to diagnosis and treatment of invasive
aspergillosis. Am J Respir Crit Care Med 173:707717
52. Holding KJ, Dworkin MS, Wan PC et al (2000) Aspergillosis among people infected with
human immunodeficiency virus: incidence and survival. Adult and adolescent spectrum of
HIV disease project. Clin Infect Dis 31:12531257
53. Zmeili OS, Soubani AO (2007) Pulmonary aspergillosis: a clinical update. Q J Med 100:
317334
54. Moore PS, Chang Y (1995) Detection of herpesvirus-like DNA sequences in Kaposis
sarcoma in patients with and those without HIV infection. N Engl J Med 332:11811185
55. Antman K, Chang Y (2000) Kaposis sarcoma. N Engl J Med 342:10271038
56. Aboulaphia DM (2000) The epidemiologic, pathologic, and clinical features of AIDS-
associated pulmonary Kaposis sarcoma. Chest 117:11281145
57. Martin-Carbonero L, Barrios A, Saballs P et al (2004) Pegylated liposomal doxorubicin plus
highly active antiretroviral therapy versus highly active antiretroviral therapy alone in HIV
patients with Kaposis sarcoma. AIDS 18:17371740
58. Hirsch HH, Kaufmann G, Sendi P et al (2004) Immune reconstitution in HIV infected
patients. Clin Infect Dis 38:11591166
59. Lawn SD, Bekker LG, Miller RF (2005) Immune reconstitution disease associated with
mycobacterial infections in HIV-infected individuals receiving antiretrovirals. Lancet Infect
Dis 5:361373
60. Bourgarit A, Carcelain G, Martinez V et al (2006) Explosion of tuberculin-specific Th1-
responses induces immune restoration syndrome in tuberculosis and HIV co-infected
patients. AIDS 20:F1F7
Therapy of Infection in the ICU
22
J. H. Rommes, N. Taylor and L. Silvestri
22.1 Introduction
The 25-year-plus endurance of the selective decontamination of the digestive tract

(SDD) strategy as an evidence-based infection prevention maneuver in intensive
care medicine has led to fundamental changes in treating infections in intensive
care unit (ICU) patients [1]. To appropriately treat such patients, the intensivist
must have a thorough knowledge of basic principles, such as carriage, coloniza-
tion, and infection, normal and abnormal flora, pathogenicity of microorganisms,
and infection pathogenesis (Chaps. 16). There are six general principles under-
lying treatment for these patients (Box 22.1):
1. Obtain surveillance and diagnostic cultures. Whereas empirical antibiotic
treatment needs to be started immediately [2], it is important that diagnostic and
surveillance cultures are obtained to identify the causative pathogen(s) as soon
as possible so that the initial and empirical treatment can be modified if
necessary. Before antibiotic administration, diagnostic blood samples must be
collected. Immediately thereafter, parenteral antibiotics should be administered.
To obtain diagnostic samples of urine, tracheal aspirate and wounds, together
with surveillance samples of the vagina, throat, and rectum takes more time and
can be postponed until the antibiotics are initially administered but then must be
collected as soon as possible.
2. Administer immediate and adequate parenteral antibiotic treatment. Antimi-
crobial therapy should be administered immediately after obtaining blood
samples for culture. Delivery of antibiotics must be reliable, and this may
only be realistic using parenterally administered antimicrobial agents. Orally
J. H. Rommes (&)
Gelre Ziekenhuizen Apeldoorn, Intensive Care, Apeldoorn, The Netherlands
374 J. H. Rommes et al.
Box 22.1 Six basic principles for treating infections in ICU patients
1. Obtain both surveillance and diagnostic cultures

2. Administer immediate and adequate parenteral antibiotic treatment
3. Eradicate both internal and external sources of potential pathogens
4. Apply topical antimicrobials
5. Replace or if possible remove all plastic invasive devices
6. Evaluate the efficacy of the treatment
administered drugs in critically ill patients are unpredictable with regard to

absorption due to dysfunction of the gastrointestinal tract. The importance of
early appropriate antibiotic administration has been shown in numerous studies
regardless of infection type and has been confirmed in patients with sepsis,
pneumonia, meningitis, and urinary tract infections [3]. The choice between
mono- or combination therapy depends on the patients premorbid health status
and infection site. If the patient was previously in reasonably good health, initial
empirical monotherapy with cefotaxime is sufficient, as the patient carries normal
flora in the oropharynx and gut. Normal flora and highly pathogenic microor-
ganisms such as Streptococcus pyogenes or Neisseria meningitidis are invariably
susceptible to this third-generation cephalosporin. If a previously healthy patient
is admitted with a community-acquired pneumonia (CAP), initial empirical
combination therapy, i.e., cefotaxime with a macrolide covering the atypical
microorganisms must be administered [4]. In the case of an intra-abdominal
infection, gentamicin and metronidazole are added to cefotaxime. Yeasts have to
be covered as long as the results of the diagnostic cultures of intra-abdominal
samples are not available. Carriage of yeasts in the gut is present in 4050% of
patients, and yeast infections are an independent risk factor for mortality [5].
Patients admitted with a chronic underlying condition or who are transferred from
another hospital or ward are usually carriers of both normal and abnormal flora
[6]. When managing these patients, it is important to be aware that abnormal flora
may contain resistant microorganisms. To reduce the likelihood of inappropriate
therapy, which is associated with a higher mortality rate, all infections in these
patients should be treated by combination therapy with two antibiotics of different
mechanisms of action, i.e., aminoglycosides in combination with a beta-lactam
antibiotic, e.g., cefotaxime. In patients with septic shock of unknown origin,
combination therapy with gentamicin and cefotaxime should be started [7]. On
the next day, the microbiologist is able to distinguish normal from abnormal
potential pathogens. This information allows the intensivist to adjust the paren-
teral antimicrobials. There is no evidence to support an antibiotic course
exceeding 1 week when treating the majority of infections [8]. Three days
treatment with aminoglycosides is sufficient and should then be discontinued to
prevent toxicity (Box 22.2).
3. Eradicate the source of potential pathogens causing the infection. The source
of potential pathogens causing the infectionwhether internal or external
requires elimination to eradicate the original infection and to prevent relapses
22 Therapy of Infection in the ICU 375
Box 22.2 Initial and empirical therapy of infections in critically ill patients
Previously healthy
Pneumonia Cefotaxime ? erythromycin
Intra-abdominal infection Gentamicin/cefotaxime/metronidazole/amphotericin B
Urosepsis Gentamicin/cefotaxime
Sepsis of unknown origin Gentamicin/cefotaxime
Meningitis Cefotaxime/ampicillin
Cholangitis Gentamicin/ceftriaxone/metronidazole
Underlying disease/transferred from another ward or ICU
Pneumonia Gentamicin/cefotaxime ? erythromycin
Intra-abdominal infection Gentamicin/cefotaxime/metronidazole/amphotericin B
Urosepsis Gentamicin/cefotaxime
Sepsis of unknown origin Gentamicin/cefotaxime
Meningitis Gentamicin/cefotaxime/ampicillin
Cholangitis Gentamicin/ceftriaxone/metronidazole
and superinfections. In endogenous infections, the throat and gut of the criti-
cally ill are the internal sources of potential pathogens, whereas in exogenous
infections, the source of potentially pathogenic microorganisms (PPM) is
external, i.e., outside the patient. All patients expected to require C2 days of
mechanical ventilation, immediately receive SDD (Chapter XX). Oropharyn-
geal, gastrointestinal, and vaginal carriage can be abolished by topical
administration of the nonabsorbable polymyxin E, tobramycin, and amphoter-
icin B (PTA), with the aim of eradicating the internal sources. If a patient has a
tracheostomy, a paste containing PTA is applied around the tracheostoma [9].
The administration of these nonabsorbable topical antimicrobials is continued
until the patient is weaned from the ventilator and extubated. Identifying and
eradicating an external source is often more difficult and requires close coop-
eration between intensivists, nurses, and the infection control team.
4. Deliver topical antimicrobials to achieve high antibiotic concentrations on the
site of infection. Topical application of antimicrobials is safe and contributes to
a more rapid killing of PPMs, resulting in cultures of colonized/infected sites
becoming sterile earlier. For example, to increase the antimicrobial activity in
the lower airway secretions, topical therapy using aerosolized antibiotics should
be considered [10]. Pastes with PTA can be applied topically to tracheostomies,
gastrostomies, and pressure sores. These antimicrobial agents mixed with a
translucent aquaform gel can also be applied in thin layers over fine mesh gauze
to cover, for example, grafted burn wounds [11].
5. Removal or replacement of invasive devices. Endotracheal tubes, intravascular
lines, and urinary catheters are readily contaminated with microorganisms.
Removal or replacement is thought to contribute to recovery from the original

infection by curtailing the supply of potential pathogens [12].
6. Evaluate efficacy of the therapy. Generally, if appropriate antimicrobial therapy
is administered, diagnostic samples of the infected internal organs are sterile
within 3 days [13]. The oropharyngeal and vaginal cavities are free from
potential pathogens within 3 days of SDD application. The same applies to
topical application of antimicrobial agents to tracheostomies, gastrostomies,
and wounds. Decontamination of the gutin particular, the rectal cavity
depends on peristalsis and passage through the gastrointestinal tract. Liberal use
of laxantia and a parasympathomimetic, such as neostigmine, reduce transit
time and thus the time to gut decontamination [14].
22.2 Treating Infection on Admission
The majority of patients are admitted to the ICU due to or with an infection.
Immediate and adequate therapy based on the six SDD principles should be given
to prevent mortality. Details of treatment for the most frequently encountered
infections on admission are now described (Box 22.3).
22.2.1 Lower Airway Infection
There are two types of lower airway infections: tracheobronchitis and pneumonia.
Tracheobronchitis is an infection of the trachea and/or bronchi with localized
(purulent secretion) and generalized clinical signs (fever, leukocytosis, increased
C-reactive protein (CRP)). Chest X-ray does not show infiltrates. The tracheal
aspirate yields [3+ or [105 CFU/ml of a PPM in the presence of [2+ leukocytes.
The clinical setting in which this type infection most frequently is encountered is
in chronic obstructive pulmonary disease (COPD) patients with acute on chronic
respiratory failure or a patient with neuromuscular weakness who develops
respiratory failure due to retention of secretions and atelectasis followed by
infection. Pneumonia is an infection of the pulmonary tissue. The clinical diag-
nosis is based on the presence of fever, leukocytosis, and increased CRP and a new
or progressive pulmonary infiltrate on chest X-ray. Tracheal aspirate is macro-
scopically purulent and contains [3+ or [105 CFU/ml of a PPM and [2+
leukocytes. Early administration of appropriate antibiotic reduces mortality in
patients with a lower respiratory tract infection [15]. Immediately after obtaining
diagnostic blood samples, systemic antibiotics should be started. In a previously
healthy patient, cefotaxime in combination with a macrolide, e.g., erythromycin,
should provide adequate cover of normal and atypical PPMs. Patients with chronic
underlying disease and those transferred from the ward or another ICU carry both
normal and abnormal PPMs and hence require combination therapy with an
aminoglycoside and cefotaxime. Erythromycin is added in case of CAP. Cultures
will identify the next day normal or abnormal PPMs so that if necessary, anti-
microbial therapy can be adjusted. Infections caused by normal and abnormal
Box 22.3 An effective, rational, and safe antimicrobial drug policy
Antimicrobials for systemic treatment: gentamicin/tobramycin, cefotaxime, erythromycin,

metronidazole, amphotericin B; ceftazidime if Pseudomonas spp.; cephradine/cephazolin
if S. aureus; vancomycin if methicillin resistant S. aureus(MRSA); meropenem if extended
beta-lactamase (ESBL) producing PPM
Aminoglycosides; first dose 5 mg/kg, treatment for 3 days under therapeutic drug monitoring
(TDM)
Initial empiric, broad coverage therapy (the first blow is half the battle); after 24 h, tapering
down based on culture results
Therapy duration for infections caused by PPM: 5 days
Peritonitis: cover yeast with amphotericin B by continuous infusion 30 mg/24 h under TDM
Box 22.4 Dosages of aerosolized antimicrobials
Antimicrobial agent Dose (mg/5 ml) Interval (h)

Cefotaxime 500 6
Gentamicin 40 6
Tobramycin 40 6
Ceftazidime 500 6
Cephradine 500 6
Colistin (polymyxin E) 20 6
Amphotericin B 5 6
Vancomycin 250 6
PPMs can be treated with monotherapy, such as cefotaxime for 5 days. If ESBL
producing PPMs, Serratia or Morganella spp are isolated cefotaxim should be
replaced by meropenem [16]. If Pseudomonas spp. are isolated from throat and/or
tracheal aspirate, cefotaxime has to be discontinued and combination therapy with
gentamicin (3 days) and ceftazidime should be prescribed. To increase the anti-
microbial activity in the lower airway secretions, topical therapy using aerosolized
antibiotics should be applied (Box 22.4). Topical application of antimicrobials by
nebulization is safe and contributes to a more rapid killing of PPMs, resulting in
cultures of the tracheal aspirate becoming sterile earlier [10]. The doses of the
different aerosolized antimicrobials are shown in Box 22.4. Tracheal aspirate is
obtained daily until cultures are sterile. To prevent recolonization of the lower
airways, the tracheal tube should be replaced after 3 days. Gentamicin should be
guided by therapeutic drug monitoring (TDM) and generally can be discontinued
after 3 days, to prevent toxicity. Systemic antibiotics and nebulized antimicrobials
are discontinued when cultures are sterile; usually within 5 days.
22.2.2 Sepsis, Septicemia, Severe Sepsis, and Septic Shock
Sepsis is the clinical picture caused by generalized inflammation due to micro-

organisms and/or their toxic products. The clinical diagnosis of sepsis (synony-
mous with the sepsis syndrome) is based on the presence two of more of the
following clinical criteria:
increased heart rate,
tachypnoe or impaired gas exchange,
fever or hypothermia,
leucocytosis or leucopenia plus a documented or suspected infection.
The clinical picture is caused by the release of inflammatory mediators into the
circulation, i.e., cytokinemia. Blood cultures are always sterile. Septicemia is
defined as sepsis combined with a positive blood culture. Severe sepsis is defined
as sepsis plus sepsis-induced organ dysfunction or tissue hypoperfusion. Septic
shock is defined as sepsis in combination with the clinical signs of cardiovascular
collapse. Once the diagnosis of sepsis, septicemia, or septic shock has been made
and blood cultures have been taken, immediate combination therapy should be
started to provide an adequate spectrum of antibacterial activity. This is a com-
bination of an aminoglycoside with cefotaxime. If an intra-abdominal focus is
suspected, metronidazole and amphotericin B is added to this treatment. Initial
empirical therapy is adjusted according to diagnostic culture results. The source of
sepsis should be identified and eliminated as soon as possible. SDD using enteral
PTA should be commenced immediately to eradicate the internal source.
Intra-abdominal infection is an infection of an abdominal organ and the peritoneal

cavity (peritonitis), with local signs such as abdominal tenderness and generalized
symptoms including fever and leukocytosis. Peritonitis can be a localized or
generalized. Drainage of the infection site should be performed as soon as possible
by the surgeon or interventional radiologist. Again the six basic principles should
be applied [17, 18].
1. Cultures taken to identify the organism.
2. Broad-spectrum systemic antimicrobial cover, i.e., an aminoglycoside with
cefotaxime, metronidazole and amphotericin B, to empirically cover the likely
organisms. This empirical therapy is adjusted when culture results become available
after 24 h. Systemic antimicrobials including amphotericin B are given for 5 days.
3. Eradicate the source with SDD. The gut must be considered as an internal abscess
that has to be drained [17]. PPM eradication from the gut of patients with peri-
tonitis is often difficult due to the disturbance of both physiology and anatomy.
However, decontamination of the gut in that subset of patients is possible but
requires commitment and tenacity by the ICU team. Liberal use of laxantia,
prokinetics, parasympathetic mimetics such as neostigmine, and enteral feeding
restores digestive tract function. Blind loops are decontaminated with SDD
suspension containing 50 mg polymyxin E, 40 mg of tobramycin, and 500 mg of
amphotericin B administered via catheters placed in the stoma. In case of rectal
overgrowth, i.e., [105 CFU/ml of PPM, SDD enemas or suppositories are
administered twice daily until surveillance cultures are free from PPM.
4. During laparotomy and repeat laparotomy after obtaining cultures, the abdominal
cavity is extensively rinsed with a disinfecting agent, 2% Taurolin [19].
5. All potentially contaminated devices may act as a source of infection and
should be replaced or, if possible, removed.
6. Treatment is evaluated by ongoing surveillance cultures of throat, stomas and
rectum.
Clinical signs of a wound infection are purulent discharge, redness, swelling, tender-
ness, and local warmth. The clinical diagnosis is confirmed by isolating C3+ or C105
microorganisms and C2+ leukocytes in the purulent discharge [20]. Systemic
antimicrobial therapy is seldom indicated, unless symptoms of sepsis, septicemia,
or septic shock occur. Local treatment, drainage, debridement, and removal of
plastic devices are essential and generally sufficient. Following local treatment, the
wounds are rinsed twice daily with a disinfectant, 2% Taurolin, for 3 days.
Aquaform gel mixed with 2% PTA and/or vancomycin can be applied to colo-
nized/infected wounds [11].
Urinary tract infection is an infection of the pyelum, ureter, bladder, or prostate.

Severe urinary tract infections and urosepsis occur only if obstruction or lesions in
the urinary tract are present. Blood cultures should be taken, followed by anti-
microbial therapy. The combination of an aminoglycoside and cefotaxime pro-
vides adequate cover. If the Gram stain of the urine reveals yeasts, flucytosine
7,500 mg/day via continuous infusion should be given. Surveillance and additional
diagnostic cultures should be taken and an ultrasound of the urinary tract performed
as soon as possible. The cornerstone of treatment is drainage: the bladder by
catheter, the ureter and kidney by either internal drainage or percutaneous
nephrostomy. Following nephrostomy, the patient may deteriorate due to bacteremia
induced by insertion of the nephrostomy catheter via the rich vasculature of the
kidney into the pus of the pyelum.
To prevent new infections, the patient is treated with SDD. Vaginal carriage can
be the source of PPM causing bladder infections. After drainage, aminoglycoside
can be discontinued, whereas flucytosine and cefotaxime should be continued for a
maximum of 5 days.
22.2.6 Acute Bacterial Meningitis
Bacterial meningitis is an infection of the meninges, the membrane lining of the

brain, spinal cord, and subarachnoid space. Meningitis is a medical emergency
requiring immediate and adequate therapy with antimicrobials and corticosteroids.
Clinical diagnosis is often evident, as almost all patients presents with 2 or more of
the 4 symptoms: fever, headache, neck stiffness, and altered mental state. Blood
cultures must be taken immediately, followed by intravenous infusion of antibi-
otics. At the same time, dexamethasone 10 mg must be given. A 4-day regimen of
dexamethasone 0.6 mg/kg/day divided in four daily doses as additional therapy
reduces the risk of mortality and the incidence of cranial nerve impairment, such as
hearing loss and other neurological sequelae. Five microorganisms are responsible
for the vast majority of community-acquired bacterial meningitis: Haemophilus
influenzae, S. pneumoniae, Listeria monocytogenes, and Streptococcus agalactiae
(in neonates). Empirical treatment with cefotaxime and ampicillin covers these
potential causative bacteria. In early-onset meningitis in neonates, an aminogly-
coside should be added to cover E. coli from the mothers gut. Although rare,
patients may develop meningitis during hospital stay. Empirical treatment in this
small subset of patients should be guided by clinical history. Recent insertion of an
epidural catheter, for instance, is a risk factor for S. epidermidis infection requiring
vancomycin, whereas meningitis related to a basilar skull fracture or recent
neurosurgery must be treated initially with ceftazidime and an aminoglycoside to
cover both normal and abnormal PPMs and vancomycin to cover S. aureus and
S. epidermidis.
Only after administration of antimicrobials and dexamethasone does the
neurologist have the opportunity to perform lumbar puncture to obtain liquid for
clinical chemistry and culture. At the same time, diagnostic and surveillance
cultures must be collected. In pneumococcal meningitis, consultation of an
otorhinolaryngologist or cardiologist is indicated, as 60% of these patients have
contiguous or distant foci of infection, such as sinusitis, otitis, and endocarditis.
Antimicrobial treatment should be continued for 5 days.
22.2.7 Biliary Tract Infections
Infections of the gall bladder or biliary-duct system require drainage by the

surgeon, radiologist, or endoscopist. While the patient is waiting for definitive
therapy, blood cultures should be taken, followed by immediate and adequate
antimicrobial treatment. Combination therapy aimed at gastrointestinal flora,
including an aminoglycoside, third-generation cephalosporin that provides high
levels in the gall (i.e., ceftriaxone and metronidazole) covers the most likely
causative microorganisms. Following definitive treatment, aminoglycoside therapy
can be discontinued, anaerobic therapy after 3 days, and ceftriaxone after 5 days.
Candida infections of the gallbladder and biliary tract are rare, and therefore,
empirical treatment with antifungals is not indicated.
22.3 Managing Inflammation in a Successfully

Decontaminated Patient
Appropriate use of the SDD strategy reduces severe infections of the lower air-
ways and bloodstream by 72 and 37%, respectively [21, 22]. However, although
properly decontaminated, some patients may develop signs of inflammation, and
the diagnostic process in such a patient is a challenge for the intensive care team.
A systematic approach is required (Box 22.5).
If a patient develops symptoms of inflammation such as fever, increased CRP,
and leucocytosis, the first step is a thorough clinical reevaluation: Has the tracheal
aspirate changed in quantity or quality? Is there deterioration of the pulmonary
function? Is the urine cloudy, indicating the presence of leucocytes? How long ago
was the central venous catheter inserted and under what conditions? Are there
signs of an intra-abdominal problem? Is there a new cardiac murmur? In surgical
patients: How are the wounds? Are there signs of an intra-abdominal problem?
The next step is to reevaluate the results of the surveillance cultures. If cultures of
throat, vagina, and gut are free of PPM, an endogenous infection is very unlikely.
Systemic antibiotics should not be prescribed unless vital functions deteriorate
progressively and the intensivist is convinced that the patient is suffering from
sepsis. Diagnostic samples of lower airways, blood, urine, and wound fluid should
be obtained. During decontamination, low-level pathogens such as enterococci and
coagulase-negative staphylococci (CNS) may be found in the tracheal aspirate.
The clinical impact of isolation of these low-level pathogens in the tracheal
aspirate is nil; neither enterococci nor CNS cause lower airway infection.
Colonization by Candida spp. of the lower respiratory tract on admission is
frequently observed and may persist despite decontamination. Fortunately, Candida
pneumonia is extremely rare and occurs only following hematogenous spread.
Changes on X-ray of the thorax require bronchoscopy followed by culture (bacteria,
fungi), stains (Gram/Ziehl-Neelsen), and cytology. Cytology of the bronchoalveolar
lavage (BAL) should include stains aimed at viral-inclusion bodies suggesting
viral infection. Viral pneumonia due to reactivation of herpes or cytomegalovirus
(CMV) may occur in critically ill patients. The value of polymerase chain
reaction (PCR) on BAL fluid aimed at a virus is unknown but probably limited.
In abdominal surgery patients, an abdominal CT scan may provide the diagnosis.
A rather frequent cause of inflammation in a properly decontaminated intensive
care patient is an endogenous bloodstream infection with low-level pathogens,
i.e. CNS catheter-related bloodstream infection, particularly if the catheter
is [7 days in situ [23]. Infections with low-level pathogens are associated with
fever and an increase of CRP but do not cause sepsis syndrome or septic shock, as
these microorganisms lack endotoxin. Diagnosing catheter-related bloodstream
infection is based on one of the following [24]: (1) at least 2 positive blood
cultures before catheter removal and persistently negative cultures after removal;
(2) isolation of the same microorganism from 2 of the following 3 sites: blood
drawn via the suspected line, blood drawn from a peripheral vein, or from the
Box 22.5 Differential diagnostic considerations when confronted with inflammation in a

properly decontaminated patient
1. Is the infection present on admission adequately treated?

i. undrained abscesses/wound infection
ii. inadequate (too short, decreased susceptibility of PPM) antimicrobial therapy
iii. source control
2. Is there a new infection?
i. infection with low pathogens
catheter-related bloodstream infection
bladder infection
ii. lower airway
exogenous
viral infection
Aspergillus
3. Has the patient an ongoing inflammatory response?
i. infection excluded but still signs of inflammation
4. Miscellaneous
i. thromboembolism
ii. drug fever
iii. underlying disease such as rheuma, autoimmune disease, malignancy
iv. pressure sores
catheter tip; (3) quantitative blood cultures drawn via the central venous line and a
peripheral vein reveal the same microorganism in a ratio [5:1. However the
clinical relevance of this expensive microbiological exercise is limited, as the only
effective treatment of a catheter-related bloodstream infection is removal of the
contaminated line. If the diagnosis of catheter-related infection is correct, clinical
signs of infection, particularly temperature, will normalize within 24 h. The pre-
dominant microorganisms involved in this type of infection are the low-level
pathogens CNS and enterococci. Systemic treatment with antimicrobials is seldom
indicated, even if PPMs such as aerobic Gram-negative bacilli and yeasts are
involved. However an S. aureus catheter-related bloodstream infection carries the
risk of metastatic abscesses and hence requires treatment with a first-generation
cephalosporin for 5 days [25].
More complex is the patient with a prosthetic heart valve who develops a
catheter-related bloodstream infection. If the signs of infection (fever, increased
CRP, leukocytosis) do not resolve within 24 h of removal of the contaminated
catheter, an aggressive approach is indicated. Combination therapy with vanco-
mycin and gentamicin should be commenced. If ultrasonography reveals vegeta-
tions, antibiotic treatment should be continued for 3 weeks. If after 3 weeks
treatment clinical evaluation reveals signs of persistent infection, surgical
replacement of the prosthetic heart valve is indicated [26].
Abscesses and wound infections require drainage. For localized problems,

prophylactic systemic antibiotics should be given before drainage; prolonged
treatment is not indicated. When the problem is generalized, antibiotics are
administered for 5 days.
The incidence of exogenous lower respiratory tract colonization/infection with
a potential pathogen, e.g., Acinetobacter or Pseudomonas spp., depends on the
hygiene discipline of ICU workers but should be \15%. Colonization, no signs of
infection, and \105 PPM/ml requires nebulized antimicrobials. Infection should be
treated with both systemic (aminoglycoside and beta-lactam antimicrobial) and
nebulized antimicrobials (aminoglycoside or colistine) for 5 days. Replacement of
the endotracheal tube or tracheostomy cannula should be performed on the third
day. Eradication of the source of the PPM is an important component of effective
treatment of respiratory tract infections. Contaminated ventilators, humidifiers,
and sinks are potential sources of external PPM. Breaches of hygiene by care-
givers, particularly during busy periods, may lead to increased transmission of
microorganisms and a higher exogenous infection rate.
Critically ill patients suffer from pronounced immunosuppression, making them
vulnerable for all types of respiratory tract infections. If Aspergillus spp. are
isolated from the tracheal aspirate, immediate antifungal therapy, intravenously
and nebulized should always be given. Combination treatment with continuously
infused flucytosine (loading dose 2500 mg iv followed by continuous infusion
7500 mg/24 h under TDM) and amphotericin (starting dose 30 mg/24 h) under
TDM is safe and effective [27].
The diagnosis of viral pneumonia due to reactivation of herpes or CMV
requires the typical inclusion bodies in the cytology of lower airway secretion
obtained by BAL and a positive blood PCR.
Streptococcus faecalis or faecium bladder infection may occur and is readily
diagnosed. Treatment is replacement of the bladder catheter preceded by a single
dose of 1,000 mg of vancomycin.
Prolonged inflammation must be considered the moment infection has become
unlikely. If the patients recovery is hampered by this continuing inflammatory
response, treatment with a short course of high-dose corticosteroids should be
initiated (see below).
22.4 Managing Infections Caused by MRSA
The classic SDD protocol comprising parenteral cefotaxime and enteral PTA is not
designed to control MRSA infections. Control requires the addition of enterally
administered vancomycin [2830]. The policy of surveillance cultures of the throat
and rectum combined with enterally administered vancomycin in the at-risk
patient is analogous to the way in which aerobic Gram-negative bacillary and
fungal carriage is managed by the enterally administered antimicrobials PTA.
It should be remembered that the principal aim of enterally administered
nonabsorbable antimicrobials is the eradication of carriage of potential pathogen
Box 22.6 MRSA eradication protocol
1. Surveillance samples to detect carriers of MRSA

i. obtain swabs from nose, throat, and rectum
2. Enteral/topical vancomycin to eradicate carriage
i. treatment of MRSA carrier (5 days)
nasal carriage: 2% mupirocin cream 4 times a day or 4% vancomycin cream 4 times a day
oropharyngeal carriage: 4% vancomycin paste (0.5 g) 4 times a day, or 4% vancomycin gel
(0.5 g) 4 times a day, or 5 mg vancomycin lozenges 4 times a day
gastrointestinal carriage: 40 mg of vancomycin/kg per day oral solution in four doses
skin carriage: 4% chlorhexidine bath/shower on alternate days
ii. treatment of colonization/infection (3 days)
tracheostomy, gastrostomy: 4% vancomycin paste twice a day; change foreign body
lower airways: nebulized vancomycin 4 mg/kg per dose, 4 times a day diluted in normal
saline; patient must receive a dose of nebulized salbutamol prior to vancomycin due to the
reported risk of bronchoconstriction
3. Limiting the use of flucloxacillin to lift selection pressure on MRSA
4. High level of antistaphylococcal hygiene, including hand washing and device policy
overgrowth, including MRSA. Thus, enterally administered vancomycin not only

eliminates a prime source of endogenous MRSA infection in at-risk patients, it
also profoundly reduces hand contamination with MRSA and subsequent dis-
semination within the ICU.
Box 22.6 shows MRSA eradication protocol components. These include sur-
veillance samples to detect the asymptomatic oropharyngeal and gut carrier. These
are crucial for controlling antibiotic resistance. Regular surveillance cultures on
admission and throughout treatment in the ICU, e.g., Mondays and Thursdays,
allow detection of asymptomatic MRSA carriers at an early stage, allowing
immediate implementation of isolation, barrier precautions, and enterally admin-
istered vancomycin. Relying solely on diagnostic blood, tracheal aspirate, urine,
and pus samples results in inherent and substantial delay, permitting MRSA dis-
semination to other patients and maintaining endemicity. A shift from diagnostic
toward preemptive surveillance samples is required to avoid wasting time in
controlling MRSA spread. Nasal surveillance must be supplemented with digestive
tract surveillance, as gut carriage of MRSA cannot be ignored [31]. Enterally/
topically administered vancomycin is used to eradicate MRSA carriage: applica-
tion of a 4% vancomycin paste or gel has been found to be effective in eradicating
oropharyngeal carriage; administration of a vancomycin solution (40 mg/kg/day)
through the nasogastric tube readily clears gut carriage; mupirocin or vancomycin
intranasally is indicated to eradicate nasal carriage; 4% chlorhexidine liquid soap
is used to clear skin carriage.
MRSA has an affinity for both intact and damaged skin, particularly when a
plastic device such as a tracheostomy or gastrostomy is present [32, 33]. After
device removal and replacement, a 4% vancomycin paste is required to treat
colonization and/or infection. In lower airway colonization/infection, aerosolized
vancomycin should be delivered via the endotracheal tube. Flucloxacillin should

be avoided to lift selection pressure on MRSA. In general, improved antibiotic use
to limit selective pressure is not that difficult but unfortunately receives little
consideration in therapeutic decision making. Protection of indigenous flora is
required to control overgrowth by MRSA. Cephradine is preferred as the first-line
antistaphylococcal agent, as flucloxacillin disrupts gut ecology to a greater extent
than cephradine [34]. A high level of antistaphylococcal hygiene is important.
Improved adherence to infection-control practicesin particular, hand hygiene
cannot be overstated to control MRSA transmission via the hands of carers. It is
highly likely that hand washing will be more effective in preventing MRSA
transmission in units that implement the new approach. Carriers who receive
vancomycin enterally have significantly less MRSA on their skin, so the risk of
contaminating carers hands is less and the level of contamination is lower, making
hand washing more effective.
Staphylococci, both coagulase positive and negative, have an affinity for plastic
devices. Most patients who require long-term intensive care have indwelling
devices, including intubation tubes, intravascular lines, urinary catheters, and
tracheostomy and/or gastrostomy tubes. The chance that these devices become
contaminated with MRSA is substantial in a patient carrying MRSA in the nose,
throat, gut, or skin [32, 33]. A strict device policy is thus required in the ICU.
Devices are changed immediately if diagnostic samples are positive for MRSA,
e.g., in the case of positive tracheal aspirates, the ventilation tube is replaced; in
the case of a positive blood culture taken through an indwelling vascular line or a
positive vascular catheter site swab, intravascular lines are removed and replaced.
22.5 Corticosteroids
Severity of infection is related to the extent of the inflammatory response.

Excessive and systemic inflammatory response is supposed to play a role in the
development of organ dysfunction that occurs during severe sepsis and septic
shock. Patients with such an excessive inflammatory response to infection need
modulation of this systemic cytokine response. Corticosteroids are potent inhibi-
tors of inflammation by switching off genes that encode proinflammatory cyto-
kines and switching on genes that encode anti-inflammatory cytokines. In a
randomized controlled trial (RCT) in patients with typhus, a beneficial effect of
high-dose steroids was demonstrated [35]. An RCT done in the 1980s investigated
the effect of short courses of high-dose steroids in severe sepsis and septic shock
patients [36]. Reversal of shock was significantly earlier in the treatment group,
leading to a reduced early mortality rate. However, mortality on day 28 was
comparable between controls and study group. The predominant cause of the late
mortality was infection related. SDD eradicates the superinfection problem in the
ICU patient, thus probably preserving the early beneficial effect of the high doses
of corticosteroids.
Based on these considerations, patients with an excessive inflammatory

response resulting in sepsis syndrome, severe sepsis, and septic shock should be
treated with SDD in combination with a short course of high-dose steroids to block
inflammation. Dexamethasone, due to its strong anti-inflammatory action, is pre-
ferred at a dose of 5 mg/kg (equivalent to prednisolone 30 mg/kg) for 3 days.
22.6 Conclusion
The old, dogmatic microbiology has been replaced by the sound, evidence-based
SDD approach for treating and controlling infection in ICU patients. Recom-
mendations in this chapter, particularly regarding the use of antimicrobial drugs,
differ completely from those generated during the numerous consensus meetings
and require a renewed mindset from the reader. The reader should keep in mind
that evidence showing treatment of infections in critically ill patients using the
SDD approach and a limited number of old antimicrobials has reached a grade 1A
recommendation for reducing morbidity and mortality rates and preventing the
emergence of resistance.
References
1. Zandstra DF, van Saene HKF (2011) Selective decontamination of the digestive tract as
infection prevention in the critically ill. A level 1 evidence-based strategy. Minerva
Anaesthesiologica 77:212219
2. Garnacho-Montero J, Garcia-Garmendia JL, Barrero-Almodovar A et al (2003) Impact of
adequate empirical antibiotic therapy on the outcome of patients admitted to the intensive
care unit with sepsis. Crit Care Med 31:27422751
3. Ibraham EH, Sherman G, Ward S et al (2000) The influence of inadequate antimicrobial
treatment of bloodstream infections on patient outcomes in the ICU setting. Chest 118:146155
4. Martin-Loeches I, Lisboa T, Rodriguez A et al (2010) Combination antibiotic therapy with
macrolides improves survival in intubated patients with community-acquired pneumonia.
5. Montravers P, Dupont H, Gauzit R et al (2006) Candida as a risk factor for mortality in
peritonitis. Crit Care Med 34:646652
6. Viviani M, van Saene HKF, Pisa F et al (2010) The role of admission surveillance cultures in
the patients requiring prolonged mechanical ventilation in the intensive care unit. Anaesth
7. Kumar A, Safdar N, Kethireddy S et al (2010) A survival benefit of combination antibiotic
therapy for serious infections associated with sepsis and septic shock is contingent only on
the risk of death: a meta-analytic/meta-regression study. Crit Care Med 38:16511664
8. Chastre J, Wolff M, Fagon JY et al (2003) Comparison of 8 versus 15 days of antibiotic
therapy for ventilator associated pneumonia in adults. A randomised trial. JAMA 290:
25882598
9. Morar P, Makura Z, Jones AS et al (2000) Topical antibiotics on tracheostoma prevents
10. Palmer LB (2009) Aerosolized antibiotics in critically ill ventilated patients. Curr Opinion
Crit Care 115:413418
11. Desai MH, Rutan RL, Heggers JP et al (1992) Candida Infection with and without nystatin
prophylaxis. Arch Surg 127:159162
12. Brown EM (1997) Empirical antimicrobial therapy of mechanically ventilated patients with
nosocomial pneumonia. J Antimicrob Chemother 40:463468
13. Dennesen PJW, van der Ven AJAM, Kessels AGH et al (2001) Resolution of infectious
parameters after antimicrobial therapy in patients with ventilator-associated pneumonia. Am
J Respir Crit Care Med 163:13711375
14. van der Spoel JI, Oudemans-van Straaten HM, Stoutenbeek CP et al (2001) Neostigmine
resolves critical illness-related colonic ileus in intensive care patients with multiple
organ failurea prospective, double-blind, placebo-controlled trial. Intensive Care Med 27:
822827
15. Alvarez-Lerma F, Group ICU-Acquired Pneumonia (1996) Modification of empiric antibiotic
treatment in patients with pneumonia acquired in the intensive care unit. Intensive Care Med
22:387394
16. Abecasis F, Sarginson RE, Kerr S, Taylor N, van Saene HK (2011) Is selective digestive
decontamination useful in controlling aerobic gram-negative bacilli producing extended
spectrum beta-lactamases? Microb Drug Resist 17:1723
17. Marshall JC, Innes M (2003) Intensive care unit management of intra-abdominal infection.
18. Lamme B, Boermeester MA, Reitsma JB et al (2002) Meta-analysis of relaparatomy for
secondary peritonitis. Br J Surg 89:15161524
19. Gormans SP, McCafferty DF, Woolfson AD (1987) Reduced adherence of micro-organisms
to human mucosal epithelial cells following treatment with taurolin, a novel antimicrobial
agent. J Appl Bacteriol 62:315320
20. Weber JM, Sheridan RL, Pasternack ME et al (1997) Nosocomial infections in pediatric
patients with burns. Am J Infect Control 25:195201
CD 000022
digestive tract reduces bacterial bloodstream infections and mortality in critically ill patients.
Systematic review of randomised, controlled trials. J Hosp Infect 65:187203
23. Elliott TSJ, Faroqui MH, Armstrong RF et al (1994) Guidelines for good practice in central
venous catheterisation. J Hosp Infect 28:163176
24. Kurkchubasche AG, Smith MD, Rowe MI (1992) Catheter-sepsis in short bowel syndrome.
Arch Surg 127:2125
25. Naber CK, Baddour LM, Giamarellos-Bourboulis EJ et al (2009) Clinical consensus
conference: survey on Gram-positive bloodstream infections with a focus on Staphylococcus
aureus. Clin Infect Dis 48:S260S270
26. Brush JL (1998) Infective endocarditis in critical care. In: Cunha BA (ed) Infectious diseases
in critical care medicine. Dekker, New York, pp 387434
27. Grayson ML (ed) (2010) Kucers The use of antibiotics, 6th edn. Hodder Arnold, London
Control 30:391399
30. Silvestri L, van Saene HKF, Milanese M et al (2004) Prevention of MRSA pneumonia by oral
vancomycin decontamination: a randomised trial. Eur Respir J 23:921926
31. Coello R, Jimenez J, Garcia M et al (1994) Prospective study of infection, colonization and
carriage of methicillin-resistant Staphylococcus aureus in an outbreak affecting 990 patients.
Eur J Clin Microbiol Infect Dis 13:7481
32. Steinberg JP, Clark CC, Hackman BO (1996) Nosocomial and community-acquired
staphylococcus bacteremias from 1980 to 1993: impact of intravascular devices and
methicillin-resistance. Clin Infect Dis 23:255259
33. Chang FY, Singh N, Gayowski T et al (1998) Staphylococcus aureus nasal colonisation in
34. de Man P, Verhoeven BA, Verbrugh HA et al (2000) An antibiotic policy to prevent
35. Hoffman SL, Punjabi NH, Kumala S et al (1984) Reduction of mortality in chloramphenicol-
treated severe typhoid fever by high-dose dexamethasone. N Engl J Med 310:8288
36. Sprung CL, Caralis PV, Marcial EH et al (1984) The effects of high-dose corticosteroids in
patients with septic shock. New Engl J Med 311:11371143
Part V
Special Topics
The Gut in the Critically Ill: Central
Organ in Abnormal Microbiological 23
Carriage, Infections, Systemic
Inflammation, Microcirculatory
Failure, and MODS
D. F. Zandstra, H. K. F. van Saene

and R. E. Sarginson
23.1 Introduction
For many years, the gut has been proposed to play a central role in the patho-
genesis of infections, multiple organ failure, and other diseases frequently
encountered in the critically ill. Also, it is recognized that therapeutic interventions
in the critically ill, such as stress ulcer prophylaxis, systemic antibiotics and,
vasoconstrictors, can cause inadvertent adverse effects. For example, physiological
balances in microbial colonization and microcirculation are impaired, thereby
contributing to increased susceptibility to bacterial overgrowth (i.e., abnormal
carriage), infections, sepsis, intestinal barrier dysfunction and, ultimately, multiple
organ dysfunction syndrome (MODS) [14].
The gut consists of different components: microcirculation, mucosa, immune
system, enteric nervous system, commensal microflora, andduring critical
illnessacquired microorganisms. All these components interact with each other
during critical illness, and all elements may become disturbed as a consequence of
disease and also its treatment. This may result in alterations in crosstalk between
the different elements and yield both beneficial and pathologic responses. The
latter responses may induce MODS. Clark and Coopersmith propose that the
intestinal epithelium, the intestinal immune system, and the intestines endogenous
bacteria all play vital roles in driving MODS [5]. The complex crosstalk between
these three interrelated parts of the gastrointestinal tract cumulatively makes the
gut a motor of critical illness. The importance of the close relation between
D. F. Zandstra (&)
Department of Intensive Care, Onze Lieve Vrouwe Gasthuis,
Amsterdam, The Netherlands
392 D. F. Zandstra et al.
abnormal microbial carriage, immunity, and intestinal microcirculation in the

critically ill is addressed in this chapter. Therapeutic interventions within this
context are reviewed.
23.2 Microcirculation
Hippocrates described the clinical triad of cold extremities, fever, and thirst, which
implied a bad prognosis. He was probably describing the situation of hypodynamic
septic shock. In the first century A.D, Celsus described the classic signs of
inflammation: rubor, tumor, calor, dolor. Two centuries later, Galen added
the functio laesa of the affected part as the fifth symptom. These signs can be
recognized in systemic inflammatory response syndrome (SIRS), which frequently
occurs in patients after trauma, sepsis, and major surgery. Rubor is the result of
increased blood flow caused by vasodilatation and clinically measured as hypo-
tension. Tumor develops as the consequence of the generalized edema associated
with increased capillary leak. Calor can be recognized systemically as fever
caused by substances such as interleukins and other inflammatory mediators that
alter the set point of the thermoregulatory center. Dolor can be seen as generalized
pain, and functio laesa as the onset of multiple organ dysfunctions.
23.3 Multiple Organ Failure
The description of multiple organ failure is a rather recent event in the history of
medicine. This syndrome was first reported by Tilney et al. in 1973 [6]. They
described the onset of sequential organ failure in patients after surgery for ruptured
abdominal aneurysm and in whom acute renal failure had developed. Sequential
failure of circulation, respiration, liver, and intestines occurred despite intensive
therapy. All patients died. All showed gastric mucosal lesions on postmortem
investigation, and most patients showed renal tubular necrosis. These findings
indicated impaired perfusion of the stomach and kidneys. Postmortem examina-
tions revealed Gram-negative microorganisms in lung slices of all patients. The
relationship between shock, impaired gut perfusion, and the ultimate failure of
organs was evident. However, the role of Gram-negative infections in the patho-
genesis of the syndrome became evident much later [7]. Moreover, the relationship
between sepsis and (gut) mucosal ischemia became apparent. Le Gall et al.
reported in 1976 extensive gastric mucosal erosions seen on endoscopy in patients
with sepsis, whereas patients without sepsis showed normal mucosal surfaces in
the stomach [8].
The acid hypothesis as the major factor in the pathogenesis of gut mucosal
stress ulceration held sway for many years. Impaired microcirculation, as a con-
sequence of sepsis and shock, was subsequently proposed as a major step in the
pathogenesis of gastric mucosal ulceration and bleeding. In a group of high-risk
critically ill patients on prolonged mechanical ventilation, a clinical strategy of
23 The Gut in the Critically Ill 393
optimizing microcirculation of the intestinal mucosa, together with preventing

abnormal carriage of microorganisms by using selective decontamination of the
digestive tract (SDD), resulted in a surprisingly low incidence of stress-ulcer-
related bleeding (SURB) without the use of specific SURB prophylaxis [9].
Similar studies 10 years later showed identical results [10, 11]. In patients
admitted to the ICU with sepsis, sepsis syndrome, and septic shock, a SURB
incidence \1% was shown by focusing on infection prevention using SDD
and optimizing microcirculation with aggressive fluid resuscitation, the use of
vasodilators, and anti-inflammatory agents without using prophylactic antacids,
proton pump inhibitors (PPI), or H2 antagonists [12].
These findings support the hypothesis that the combination of therapeutic
interventions focused on optimizing gut perfusion (fluid resuscitation, vasodila-
tors) and preventing and treating abnormal microbial carriage and organ-site
infections (using SDD) are able to prevent some of the universally reported
problems in the critically ill. In addition to stress ulcerations of the upper intestinal
canal, these problems include loss of colonization resistance, pneumonia, sepsis,
and progression from sepsis to multiple organ failure.
SDD significantly reduces the incidence of pneumonia and bloodstream
infections, the onset of MODS, and mortality rates in the critically ill [7, 13, 14].
Impaired gastrointestinal motility resulting in delayed or absent gastric emptying
and passage of stools are recognized entities in the critically ill [1517].
Compromised caloric intake leads to impaired mucosal integrity and immune
deficiencies. The development of critical-illness-related colonic ileus (CIRCI) may
lead to intestinal ischemia by increasing intra-abdominal pressure and subse-
quently impairing mucosal perfusion. This results in disrupted barrier function
of the mucosa, endotoxemia, and finally SIRS. Untreated gut motility disorders
contribute to stasis, bacterial overgrowth, and carriage of abnormal microorgan-
isms. Timely intervention with neostigmine may prevent this sequence of
events [17].
Commonly used pharmacological agents, such as PPIs, may further contribute
to carriage of abnormal microorganisms. PPIs are designed to shut down the
gastric proton pump of parietal cells, thereby raising the pH of the stomach and
affecting the pH of the colonic content. Although effective in reducing gastric acid
production, a number of side effects have been associated with PPI use.
Naturally occurring bacteria, some of which are acid-producing and contain
adenosine triphosphatase (ATPase) enzymes, have also been found within the oral
cavity, stomach, upper gastrointestinal tract, and colon. Literature reports have
suggested that PPIs may affect these bacteria and fungi in two different ways: by
directly targeting the proton pumps of bacteria and fungi and by indirectly
affecting flora microenvironment via changes in pH. Vesper et al. summarize what
is known about the interactions between PPIs and natural human microbiota [18].
Raising colonic pH [ 6 may inhibit the growth of indigenous anaerobic flora in
the large intestine, contributing to a loss of colonization resistance [19], which in
turn may result in overgrowth by potentially pathogenic microorganisms and
subsequent infections. An increased susceptibility for infectious complications,
such as pneumonia and Clostridium infections, has been reported [2022]. The use
of PPIs should be carefully considered in the critically ill, as inadvertent effects
may cause harm.
23.4 Microcirculation and the Gut
Regulating microcirculatory perfusion in the gut is complex. Gut functional and

structural integrity is maintained under the influence of substances such as cate-
cholamines; hormones; locally released substances, including serotonin, vaso-
pressin, and endothelial vasoactive substances; and food absorption, with the
exclusion of bacteria and their products to prevent them from entering the
bloodstream [4]. It is increasingly recognized that the normalization of blood
pressure during the treatment of septic shock does not automatically result mucosal
microcirculation restoration [2328]. Measuring systemic hemodynamic parame-
ters does not reflect microcirculatory behavior of the intestinal canal [23, 26, 27].
Impairment of effective microcirculation perfusion is the essential step in the
pathogenesis of septic shock [29], and persistently impaired microcirculation in
septic shock is associated with increased risk of mortality [24].
Early studies from Dietzman and coworkers showed that the pillars for treating
(septic) shock were fluid infusion, glucocorticoids, beta agonists, and alpha
blockers [30]. Clinical studies show no benefit from adrenergic infusion alone on
organ perfusion [31, 32]. Breslow et al. found that norepinephrine (NE) did not
prevent reduced renal perfusion in an endotoxin-induced shock model, whereas
Johannes et al. demonstrated that dexamethasone improved renal perfusion and
prevented the onset of acute renal failure [33, 34]. Reduced perfusion in the
pancreas in a septic ovine model could also not be prevented, regardless the agents
used (dobutamine, norepinephrine, dopamine, dopexamine, salbutamol) [35].
Dopamine, however, increased liver and intestinal perfusion [36]. Besides vaso-
constriction, NE causes thrombocyte hyperaggregation. This may result in
impaired microcirculatory flow and subsequently organ dysfunction [37]. NE
levels achieved in humans during moderate exercise can result in hyperaggregation
[38]. However, infusion of the nitric oxide (NO) donor nitroprusside can prevent
NE-induced hyper-aggregation [39]. NE is increasingly used to treat hypotension
associated with septic shock, albeit no clinical study shows a survival benefit [40].
Also, low-dose vasopressin infusion increased mortality rates in an ischemia
reperfusion model in mice [41]. Experimental studies further suggest that NE in
septic shock also leads to MODS activation via the following:
1. ApoptosisNE induces apoptosis of alveolar cells and cardiomyocytes within
24 h [42, 43]; epinephrine had a harmful effect on survival that was signifi-
cantly related to drug dose but not bacterial dose [44].
2. NE impairment of circulating volumeWeil et al. showed in 1975 a sharp
decrease in plasma volume after starting NE infusion [31]. In patients
with increased lactic acidosis as the consequence of septic shock, impaired
peripheral circulation, clinically assessed, identifies these patients [45].
Early goal-directed therapy in patients with septic shock, as proposed by Rivers

et al., reduced mortality rates due to acute hemodynamic collapse. This result
was achieved by the infusion of more volume and blood rather than an
increased administration of catecholamines [46]. Subsequent studies with
similar design confirmed these effects [47].
23.5 Immunology of the Gut
The luminal surface of the gut is continually exposed to dense populations of

bacteria and their products. The intestinal immune system evolved to facilitate
selective absorption of food particles and selective exclusion of bacteria and their
products. The intestinal mucosa is the main interface between the immune system
and the external environment. The role of the mucosal immune system therefore is
to prevent uptake of potentially harmful compounds. A complex system of
immunocompetent cells is organized in and around the gut [gut-associated lym-
phoid tissue (GALT) and mucosa-associated lymphoid tissue (MALT)]. The innate
system is the primary barrier against microorganisms, coordinating a targeted
response to eradicate invading pathogens. An essential component of the innate
system is the presence of T and B-cell receptors and recognition of processed
antigens or presenting antigens. In the intestinal tissues, two main immune systems
can be identified: (1) the initiation compartment (the GALT), and (2) the effector
compartment (consisting of the lamina propria where T-cells reside, migrating
after stimulation in the first compartment) [48]. Despite the redundant design with
several lines of defense of the immune system, bacterial translocation (BT) may
occur and contribute to the development of SIRS and, eventually, MODS. BT has
been related to three main conditions: (1) intestinal bacterial overgrowth,
(2) disruption of the intestinal barrier, and (3) host immune deficiency [13]. These
circumstances are usually encountered in the critically ill patient. BT may occur
via the hematologic and lymphatic routes; it activates GALT and the mesenteric
lymphatic network and seems to be the natural route for immune cells and
inflammatory mediators to reach the bloodstream via portal circulation [49, 50].
Interactions between the enteric nervous system and the immune system occur
via enteric peptides. Many of these peptides originate from the mucosa. However,
they are also produced in the immunocompetent GALT cells, such as macrophages
and mast cells. These peptides may instigate different responses, including
lymph proliferation, cytokine production, and immunoglobulin (Ig) production.
Overgrowth of newly acquired, predominantly Gram-negative, microorganisms
may lead to inappropriate activation of immunocompetent cells in the vicinity of
the gut. This may result in cytokine release, which further enhances permeability
change.
It has been suggested that lymphocyte recruitment to microvessels is affected
by the presence of enterobacteria [51, 52]. Increased expression of adhesion
molecules leads to increased adherence of lymphocytes to the endothelium of the
intestinal microcirculation. This occurs locally in the gut, where enterobacteria are
present within the lumen and enter the lymph system, activating the GALT
immune response.
A relationship between BT and microcirculation injury has been reported in an
experimental setting. Lymph deviated from entering the systemic circulation
prevented the onset of mesenteric microcirculation injury; in non-lymph-deviated
animals, important and long-lasting injury to the mesenteric microcirculation was
observed [50, 53]. It is hypothesized by these authors that the lymphatic route of
BT up to the mesenteric lymph nodes (MLN) is the relevant route for the induction
of immune response, whereas the portal hematologic route might be the major
route for bacteria dissemination from the intestines into systemic organs [49].
23.6 Low Cardiac Output
These events also occur under clinical conditions of low cardiac output, such as
chronic heart failure. Sandek et al. reviewed the impact of chronic heart failure on
gut functions [54, 55]. Chronic heart failure (CHF) results in increased sympa-
thetic tone, hormonal derangements, anabolic/catabolic imbalance, endothelial
dysfunction, and systemic low-grade inflammation affecting various organ sys-
tems. Proinflammatory cytokines appear to play important roles in that context.
There is increasing evidence that the gut has a pathophysiological role for both
chronic inflammation and malnutrition in CHF. Indeed, disturbed intestinal
microcirculation and barrier function in CHF seem to trigger cytokine generation,
thereby contributing to further impairment in cardiac function. On the other hand,
myocardial dysfunction can induce microcirculatory injuries, leading to disruption
in the intestinal barrier. This amplifies the inflammatory response. The increased
number of adherent bacteria on the intestinal mucosa seen in patients with CHF
and elevated systemic levels of antilipopolysaccharide IgA emphasizes this fact.
Therefore, the gut is an interesting target for therapeutic interventions in patients
with CHF, in many of whom attempts to eliminate Gram-negative bacteria and
endotoxins from the gut by using nonabsorbable antibiotics (SDD) improved
vascular reactivity and peripheral circulation. Reducing the intestinal endotoxin
pool in the gastrointestinal tract by SDD led to decreased monocyte CD14
expression and intracellular cytokine production in patients with severe CHF.
The improved peripheral endothelial function could be a marker of the anti-
inflammatory effect of SDD [56].
23.7 Probiotics
Immune suppression and bacterial translocation caused by withholding enteral

nutrition in the presence of parenteral nutrition in an animal experiment was also
reversed and prevented by SDD [57]. In critically ill patients with MODS, the
intestinal application of probiotics resulted in an improved immune response as
shown by systemic IgA and IgM response and reduced intestinal permeability.
The clinical value of these observations is still debatable [58]. However, these
observations emphasize the important interaction between intestinal bacterial
content and immune responses.
Probiotic treatment to improve clinical outcome in critically ill patients with
pancreatitis resulted in increased mortality rates [59]. The beneficial role of pro-
biotics in the critically ill to prevent infections is still under debate [60, 61].
Eliminating aerobic Gram-negative bacilli (AGNB) from the intestinal canal using
SDD with nonabsorbable antibiotics has been shown to reduce the severity and
incidence of MODS [7, 62].
23.8 Conclusion
In the critically ill patient, overgrowth with acquired Gram-negative bacteria,

disruption of the intestinal barrier, and immune deficiency are often problematic.
Therapeutic efforts may be directed at preventing abnormal microbial colonization
by SDD; aggressive shock management using early goal-directed therapy and
vasodilators to maintain mucosal barrier (i.e., reducing translocation and pre-
serving mucosal integrity); quick restoration of intestinal motility with early
feeding; and, if needed, prokinetics. The combination of these interventions
contributes to fewer infections, fewer MODS, and a lower mortality rate. The role
of probiotics in preventing infections is still under debate.
References
1. Deitch EA (1990) Bacterial translocation of the gut flora. J Trauma 30:S184S189
2. MacFie J, OBoyle C, Mitchell CJ et al (1999) Gut origin of sepsis: a prospective study
investigating associations between bacterial translocation, gastric micro flora, and septic
morbidity. Gut 45:223228
3. Berg RD (1992) Bacterial translocation from the gastrointestinal tract. J Med 23:217244
4. Marston A, Bulkley GB, Fiddian Green RC (eds) (1989) Splanchnic ischaemia and multiple
organ failure. Edward Arnold, London
5. Clark JA, Coopersmith CM (2007) Intestinal crosstalk: a new paradigm for understanding the
gut as the motor of critical illness. Shock 28:384393
6. Tilney NL, Bailey GL, Morgan AP (1973) Sequential system failure after rupture of
abdominal aortic aneurysms: an unsolved problem in the postoperative care. Ann Surg
178:117122
7. Stoutenbeek ChP, van Saene HKF, Zandstra DF (1996) Prevention of multiple organ failure
by selective decontamination of the digestive tract in multiple trauma patients. In: Faist E,
Baue AE, Schildberg FW (eds) The immune consequences of trauma shock and sepsis.
Mechanisms and therapeutic approaches, vols 1, 2. Pabst Science, Berlin, pp 10551066
8. Le Gall JR, Mignon FC, Rapin M et al (1976) Acute gastroduodenal lesions related to severe
sepsis. Surg Gynecol Obstet 142:377380
9. Zandstra DF, Stoutenbeek CP (1994) The virtual absence of stress-ulceration related bleeding
in ICU patients receiving prolonged mechanical ventilation without any prophylaxis.
A prospective cohort study. Intensive Care Med 20:335340
10. Zandstra DF, van der Voort PH (2004) A more appropriate critical appraisal of the available
evidence? Crit Care Med 32:21662167
11. Zandstra DF, van der Voort PH (2004) Comment on surviving sepsis campaign guidelines for
the management of severe sepsis and septic shock by dellinger et al. Intensive Care Med 30:1984
12. Van Spreuwel-Verheijen M, Bosman RJ, Oudemans-Van Straaten HM et al (2006) Is the
surviving sepsis campaign for stress ulcer prophylaxis justified? Intensive Care Med
32(Suppl 1):S23
13. Silvestri L, van Saene HK, Milanese M et al (2007) Selective decontamination of the
digestive tract reduces bacterial bloodstream infection and mortality in critically ill patients.
Systematic review of randomized, controlled trials. J Hosp Infect 65:187203
15. Herbert MK, Holzer P (2008) Standardized concept for the treatment of gastrointestinal
dysmotility in critically ill patientscurrent status and future options. Clin Nutr 27:2541
16. van der Spoel JI, Oudemans-van Straaten HM, Kuiper MA et al (2007) Laxation of critically
ill patients with lactulose or polyethylene glycol: a two-centre randomized, double-blind,
placebo-controlled trial. Crit Care Med 35:27262731
resolves critical illness-related colonic ileus in intensive care patients with multiple organ
failurea prospective, double-blind, placebo-controlled trial. Intensive Care Med 27:822827
18. Vesper BJ, Jawdi A, Altman KW et al (2009) The effect of proton pump inhibitors on the
human microbiota. Curr Drug Metab 10:8489
19. Louis P, Flint H (2009) Diversity, metabolism and microbial ecology of butyrate-producing
bacteria from the human large intestine. FEMS Microbiol Lett 294:18
20. Dalton BR, Lye-Maccannell T, Henderson EA et al (2009) Proton pump inhibitors increase
significantly the risk of clostridium difficile infection in a low-endemicity, non-outbreak
hospital setting. Aliment Pharmacol Ther 29(6):626634
21. Sultan N, Nazareno J, Gregor J (2008) Association between proton pump inhibitors and
respiratory infections: a systematic review and meta-analysis of clinical trials. Can J
Gastroenterol 22:761766
22. Vakil N (2009) Acid inhibition and infections outside the gastrointestinal tract. Am J
Gastroenterol 104(Suppl 2):S17S20
23. Spronk PE, Ince C, Gardien MJ et al (2002) Nitroglycerin in septic shock after intravascular
volume resuscitation. Lancet 360(9343):13951396
24. Spronk PE, Zandstra DF, Ince C (2004) Bench to bedside review: Sepsis is a disease of the
microcirculation. Crit Care 8:462468
25. Sakr Y, Dubois MJ, De Backer D et al (2004) Persistent microcirculatory alterations are
associated with organ failure and death in patients with septic shock. Crit Care Med 32:
18251831
26. De Backer D, Creteur J, Preiser JC et al (2002) Microvascular blood flow is altered in patients
with sepsis. Am J Respir Crit Care Med 166:98104
27. Boerma EC, van der Voort PH, Spronk PE et al (2007) Relationship between sublingual and
intestinal microcirculatory perfusion in patients with abdominal sepsis. Crit Care Med
35:10551060
28. Boerma EC, Kuiper MA, Kingma WP et al (2008) Disparity between skin perfusion and
sublingual microcirculatory alterations in severe sepsis and septic shock: a prospective
observational study. Intensive Care Med 34:12941298
29. Lush CW, Kvietys PR (2000) Micro vascular dysfunction in sepsis. Microcirculation 7:83101
30. Dietzman RH, Manax WG, Lillehei RC (1967) Shock: mechanisms and therapy. Can
Anaesth Soc J 14:276286
31. Weil MH, Shubin H, Carlson R (1975) The treatment of circulatory shock. Use of
sympathomimetic and related vasoactive agents. JAMA 231:12801286
32. Dubin A, Oposo M, Casabella CA et al (2009) Increasing arterial blood pressure with
norepinephrine does not improve microcirculatory blood flow: a prospective study. Crit Care
13:r92
33. Breslow MJ, Miller CF, Parker SD et al (1987) Effect of vasopressors on organ blood flow
during endotoxin shock in pigs. Am J Physiol 252:H291H300
34. Johannes T, Mik EG, Klingel K et al (2009) Low-dose dexamethasone-supplemented fluid
resuscitation reverses endotoxin-induced acute renal failure and prevents cortical
microvascular hypoxia. Shock 31(5):521528
35. Bersten AD, Hersch M, Cheung H et al (1992) The effects of various sympathomimetics on
the regional circulations in hyperdynamic sepsis. Surgery 112:549561
36. Priebe HJ, Noldge GF, Ambruster K et al (1995) Differential effects of dobutamine,
dopamine and noradrenaline on splanchnic hemodynamics and oxygenation in the pig. Acta
Anesthesiol Scand 39:10881096
37. Marsunaga T, Fujisaki Y, Yamamoto K et al (1985) Norepinephrine in cochlear
microcirculation of guinea pigs. Am J Otolaryngol 6:226230
38. Ikaruga H, Taka T, Nakajima S et al (1999) Norepinephrine, but not epinephrine enhances
platelet reactivity and coagulation after exercise in humans. J Appl Physiol 86:133138
39. Dietrich GV, Heesen M, Boldt J et al (1996) Platelet function and adrenoceptors during and
after induced hypotension using nitroprusside. Anesthesiology 85:13341340
40. Muellner M, Urbanek B, Havel C et al (2004) Vasopressors for shock (Cochrane review).
Cochrane Library, issue 4. Wiley, Chichester
41. Indrambarya T, Boyd JH, Wang Y et al (2009) Low-dose vasopressin infusion results in
increased mortality and cardiac dysfunction following ischemia-reperfusion injury in mice.
Crit Care 13:98
42. Communal C, Singh K, Pimentell DR et al (1998) Norepinephrine stimulates apoptosis in
adult rat ventricular myocytes by activation of the B-adrenergic pathway. Circulation
98:13291334
43. Dincer HE, Gangopadhyay N, Wang R et al (2001) Norepinephrine induces alveolar
epithelial apoptosis mediated by alfa-, beta-, and angiotensin receptor activation. Am J
Physiol Lung Cell Mol Physiol 281:L624L630
44. Minneci PC, Deans KJ, Banks SM et al (2004) Different effects of epinephrine,
norepinephrine, and vasopressin on survival in a canine model of septic shock. Am J
Physiol Heart Circ Physiol 287:H2545H2554
45. Lima A, Jansen TC, van Bommel J et al (2008) The prognostic value of the subjective
assessment of peripheral perfusion in critically ill patients. Crit Care Med 34:12941298
46. Rivers E, Nguyen B, Havstad S et al (2001) Early goal-directed therapy collaborative group.
Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med
345:13681377
47. Rivers EP, Coba V, Whitmill M (2008) Early goal-directed therapy in severe sepsis and
septic shock: a contemporary review of the literature. Curr Opin Anaesthesiol 21:128140
48. Cheroutre H, Madakamutil L (2004) Acquired and natural memory T cells join forces at the
mucosal front line. Nat Rev Immunol 4:290300
49. Koh IHJ, Liberatore AMA, Menchaca-Diaz JL et al (2006) Bacterial translocation,
microcirculation injury and sepsis. Endocr Metab Immune Disord Drug Targets 6:143150
50. Koh IH, Menchaca-Diaz JL, Farsky SH (2002) Injuries to the mesenteric microcirculation
due to bacterial translocation. Transplant Proc 34:10031004
51. Rath HC (2003) The role of endogenous bacterial flora: bystander or the necessary
prerequisite? Eur J Gastroenterol Hepatol 15:615620
52. Takebayashi K, Hokari R, Kurihara C et al (2009) Oral tolerance induced by enterobacteria
altered the process of lymphocyte recruitment to intestinal microvessels: roles of endothelial
cell adhesion molecules, TGF-beta and negative regulators of TLR signaling.
Microcirculation 16:251264
53. Ruiz-Silva M, Silva RM (2006) Can bacterial translocation be a beneficial event? Transplant
Proc 38:18361837
54. Sandek A, Anker SD, von Haehling S (2009) The gut and intestinal bacteria in chronic heart
failure. Curr Drug Metab 10:2228
55. Sandek A, Rauchhaus M, Anker SD et al (2008) The emerging role of the gut in chronic heart
failure. Curr Opin Clin Nutr Metab Care 11:632639
56. Conraads VM, Jorens PG, De Clerck LS et al (2004) Selective intestinal decontamination in
57. Spth G, Hirner A (1998) Microbial translocation and impairment of mucosal immunity
induced by an elemental diet in rats is prevented by selective decontamination of the
digestive tract. Eur J Surg 164:223228
58. Alberda C, Gramlich L, Meddings J et al (2007) Effects of probiotic therapy in critically ill
patients: a randomized, double-blind, placebo-controlled trial. Am J Clin Nutr 85:816823
59. Besselink MG, van Santvoort HC, Buskens E et al (2008) Probiotic prophylaxis in predicted
severe acute pancreatitis: a randomised, double-blind, placebo-controlled trial. Lancet
371:651659
60. van Silvestri L, van Saene HK, Gregori D et al (2010) Probiotics to prevent ventilator-
associated pneumonia: no robust evidence from randomized controlled trials. Crit Care Med
38:16161617
61. Morrow LE (2009) Probiotics in the intensive care unit. Curr Opin Crit Care 15:144148
62. Silvestri L, van Saene HKF, Zandstra DF et al (2010) Impact of selective decontamination of
Nonantibiotic Measures to Control
Ventilator-Associated Pneumonia 24
A. Gullo, A. Paratore and C. M. Celestre
24.1 Introduction
Ventilator-associated pneumonia (VAP) is associated with increased intensive care

unit (ICU) stay, higher rates of morbility and mortality, pressure on critical care
capacity and increased costs [1]. Preventing hospital-acquired pneumonia (HAP)
and implementing cost-effective strategies to reduce risk and improve patient
outcomes should be a priority [2]. Multiple risk factors for VAP involve complex
host factors and ubiquitous pathogens, which require several different types of
prevention strategies [3]. Risk factors for VAP can be differentiated into modifi-
able and nonmodifiable conditions, and their identification allows the development
of strategies for prevention and the design of treatment protocols. Nonmodifiable
risk factors may be patient related (such as male sex, preexisting pulmonary dis-
ease, or multiorgan system failure). Prevention efforts should focus on modifiable
conditions [4], which are crucial targets that can reduce patient mortality and
morbidity rates and promote the cost-effective use of health care resources. There
is a growing number of evidence-based strategies for VAP prevention that, if
applied in practice, may reduce the incidence of this serious nosocomial infection.
A variety of measures has been suggested for HAP prevention, depending on
the setting and the individual risk profile [5]. Effective strategies include the use of
strict infection control, hand hygiene, microbiological surveillance with avail-
ability of data on local drug-resistant pathogens, monitoring and early removal of
invasive devices, and programs to reduce or alter antibiotic prescribing practices.
The nonantibiotic prevention strategies include conventional infection control
A. Gullo (&)
Department of Anesthesia and Intensive Care, School of Medicine,
University Hospital Catania, Catania, Italy
402 A. Gullo et al.
Table 24.1 Recommended nonantibiotic strategies to prevent nosocomial pneumonia in

mechanically ventilated patients
Strategy Recommendation
Conventional infection control Handwashing and use of protective gowns and gloves and
measures chlorhexidine oral rinse
Strategies related to the Respiratory airway care; design of endotracheal tubes;
artificial airway continuous subglottic aspiration
Strategies related to mechanical Maintenance of ventilator equipment; heat and moisture
ventilation exchangers; sedation adjustment; noninvasive mechanical
ventilation
Strategies related to the Stress-ulcer prophylaxis; gastric overdistension: nasogastric
gastrointestinal tract tubes, enteral nutrition
Strategies related to patient Semirecumbent position; rotational bed therapy
placement
measures, related to correct care of the artificial airway, and strategies related to
the position of intubated patients, the maintenance of mechanical ventilators and
equipment, and the gastrointestinal tract (Table 24.1). The Canadian Critical Care
trials groups report comprises three classes of evidence: (1) recommended
strategies are based on strong rationale and suggestive evidence, (2) strategies
may be supported by suggestive clinical or epidemiologic studies, and (3) no
recommendations are given for practices for which insufficient evidence or con-
sensus regarding efficacy exists. The report asks three questions related to HAP
prevention of HAP: What is not controversial? What is still controversial? What
should be investigated? [6].
24.2 Nonantibiotic Management
Table 24.2 shows the preventive measures for VAP with insufficient evidence or
consensus regarding efficacy.
24.2.1 Conventional Infection Prevention and Control Measures
24.2.1.1 Multidisciplinary Team Approach

Use of proper infection control practices is the cornerstone for preventing
nosocomial pneumonia. Prevention efforts must be part of an evidence-based,
multidisciplinary prevention program, with a team that sets benchmarks, estab-
lishes goals and time lines, and provides staff education and training. Colonization
of carers hands is always a concern, as it increases the risk of nosocomial
infection by cross-colonization during procedures such as tracheal suctioning,
manipulation of ventilatory circuits, and bronchoscopy. Cross-contamination via
the inoculation of bacteria into upper and lower airways (contamination of
24 Nonantibiotic Measures to Control Ventilator-Associated Pneumonia 403
Table 24.2 Nonantibiotic measures to prevent ventilator-associated pneumonia still under

debate
Preventive strategies
Infrastructure
Multidisciplinary team Programs developed by team consensus are more
effective. Input by critical care staff and respiratory
therapists is crucial
Target staff education Staff education/awareness programs reduce VAP.
Such programs are adaptable to local needs and are
cost effective
Adequate staffing Critical for maintaining patient safety and
adherence to protocols. Particularly important in
critical care units; current nursing shortages exist
Patient
Do not routinely change the breathing Not controversial
circuit more frequently than every week
Humidification system: heat and moisture Still controversial
exchangers versus heated humidification
Handwashing and protective gowns and Recommended
gloves
Chlorhexidine oral rinse Should be considered
Stress-ulcer prophylaxis Still controversial, should be investigated
Avoid gastric overdistension Recommended
Semirecumbent body position and head of Recommended
bed elevation to 3045
Postural changes by rotating beds Should be considered
Enteral nutrition Should be investigated
Avoid deep sedation Still controversial, should be investigated
Closed-system suction catheter versus open- Still controversial, should be investigated
system catheter
Orotracheal instead of nasotracheal Not controversial
intubation
Cuff-pressure optimization Not controversial
Subglottic secretion drainage Recommended
Noninvasive mechanical ventilation Still controversial, should be investigated
Early tracheostomy Still controversial, should be investigated
respiratory equipment, condensed water in ventilator-circuit tubing, excessive

manipulation of ventilator circuits) is an exogenous mechanism in VAP patho-
genesis. Cross-infection is an important source of acquiring antibiotic resistant
404 A. Gullo et al.
organisms (AROs), and hands or gloves of hospital personnel are potential res-
ervoirs for spread [7].
24.2.1.2 Handwashing
Infection control programs, such as hand disinfection; handwashing; and use of
protective gowns and gloves, aprons, and masks, to avoid contact with patient
secretions have repeatedly demonstrated efficacy in reducing infection rates. Hand
washing is widely accepted as the cornerstone of infection control in the ICU.
Literature reports show that handwashing and using protective gowns and gloves
during patient contact do not significantly reduce the rate of acquired nosocomial
infections, especially when handling respiratory secretions or during patient con-
tact when the patient carries an antibiotic-resistant pathogen. However, poor staff
compliance is not the only reason for this failure. Although handwashing alone
reduces transmission, it does not eliminate it, as transmission is dependent on the
bacterial load on health care workers hands [8]. The lack of easily reachable
appropriate physical facilities (sinks, bathrooms) has led many institutions to
alcohol-based gels, and clinical data indicate that rates of all nosocomial infection
may be significantly reduced by their use. A randomized clinical trial of ICUs is
required to support handwashing as the cornerstone of infection control.
24.2.1.3 Modulating Bacterial Colonization
Colonization of the oropharynx with pathogenic organisms is an important risk
factor leading to subsequent HAP/VAP. Host-related factors reported in the
literature that predispose to oropharyngeal colonization include renal dysfunction,
diabetes, coma, shock, advanced age, underlying lung disease, and thoracic or
upper abdominal surgery [9]. Oral care has been recommended in several studies,
and adequate daily oral hygiene using topical antiseptic agents yielded mixed
results.
Topical oral application of antiseptics such as chlorhexidine, an antiseptic
solution for controlling dental plaque, or povidoneiodine to the oral mucosa to
prevent VAP was studied in randomized controlled trials (RCTs) with conflicting
results [10]. Several RCTs examined the influence of chlorhexidine in preventing
nosocomial lower respiratory tract infection. Bacteria accumulated in dental pla-
que have been implicated as VAP pathogens when aspirated to lower airways.
Preventive oral washes with chlorhexidine therefore seem reasonable in selected
high-risk patients given its easy administration and reasonable cost. The prophy-
lactic use of chlorhexidine prevention strategies is still controversial and is sug-
gested in selected risk patients [11, 12]. The majority of meta-analyses concluded
that oral antiseptic rinses seem to be effective in reducing VAP. However, RCTs
and meta-analyses should be interpreted with caution, as it seems that these
antiseptics may be effective for preventing lower respiratory tract infection only in
patients who receive mechanical ventilation no longer than 48 h [13]; their use did
not significantly reduce mortality rates. It seems that chemical decontamination
with chlorhexidine as a solitary intervention may be insufficient to significantly
decrease the risk of pneumonia and that thorough mechanical cleaning is still
necessary. Oral decontamination with chlorhexidine did not result in significant

reduction in the incidence of nosocomial pneumonia or mortality rates in patients
receiving mechanical ventilation.
24.2.1.4 Probiotics
Previous reviews showed no benefit of probiotic administration in critically ill
patients, but they did not focus on VAP. Probiotics normally function as colonizers
and contribute to the overall health of their hosts by multiple mechanisms,
including immune and antibacterial effects. Enteral administration of probiotics
may modify the gastrointestinal environment in a manner that preferentially favors
growth of minimally virulent species. No adverse events related to probiotic
administration were identified [14]. There is no clinical evidence to support the use
of probiotics to restore normal human flora in critically ill patients and reduce
HAP rates [15]. Literature reports suggest that probiotics (e.g., Lactobacillus
rhamnosus GG) are safe and recommended in a select, high-risk ICU populations,
but administration is not associated with lower incidence of VAP [16, 17].
24.3 Strategies Related to the Artificial Airway

and Mechanical Ventilation
24.3.1 Airway, Ventilator Circuit, and Secretion Care
24.3.1.1 Subglottic Secretions Drainage

Accumulation of oropharyngeal secretions above the cuff of the endotracheal tube
is thought to increase the risk of secretion aspiration and thus pneumonia. Two
meta-analyses of RCTs showed that removing these pooled secretions through
suctioning of the subglottic region may reduce the risk of VAP and mortality rates
[18, 19]. Not only gross aspiration but also microaspiration to lower airways can
facilitate the development of VAP despite the presence of an artificial airway. It is
therefore important to maintain adequate tube-cuff pressure to reduce microaspi-
ration. Stagnant oropharyngeal secretions pooled above the cuff can easily gain
access to lower airways when the pressure decreases spontaneously or there is a
temporal deflation of the cuff, providing a direct route for tracheal colonization and
bolus aspiration. Oropharyngeal secretions may descend into the trachea, accu-
mulate above the endotracheal cuff, and later progress to the lower respiratory
tract, causing VAP. Investigators have attempted to preemptively remove these
secretions to reduce microaspiration and VAP risk and found that continuous
aspiration of subglottic secretion reduced the incidence of VAP by half and
shortened ICU stay, and thus it is recommended [20]. Continuous aspiration using
a specially designed endotracheal tube significantly reduced the incidence of early-
onset VAP in several studies but may be a less effective strategy for preventing
late-onset disease, which carries a greater risk of ARO infection and higher
mortality and morbidity rates. Subglottic secretion drainage is recommended in
patients expected to require [72 h of mechanical ventilation, and its regular use
406 A. Gullo et al.
should be encouraged in intubated patients. In the meta-analysis by Dezfulian

et al., which involved five RCTs, secretion drainage appeared effective in pre-
venting early-onset VAP; no impact on mortality was demonstrated [21, 22].
Subglottic secretion drainage appears effective in preventing early-onset pneu-
monia due to normal flora with no significant impact on survival. More studies are
needed to assess its impact on late-onset VAP, although exogenous infections are
an inherent limitation of the maneuver.
24.3.1.2 Silver-Coated Tubes

Preventing biofilm formation in endotracheal tubes is necessary in VAP prophy-
laxis. Altered surface characteristics of silver-coated endotracheal tubes interfere
with the ability of bacteria to adhere to the tubes. Silver-coated tubes can prevent
the bacterial colonization requisite for biofilm formation [23, 24], but further
investigations are needed. The silver-coated tube showed its greatest effect during
the first 10 days of mechanical ventilation, thus primarily affecting microorgan-
isms present in the patients admission flora, i.e., controls primary endogenous
VAP. Inactivation of silver ions by proteins, saliva, mucosal cells, and leukocytes
may explain why the silver-coated tubes failed to control secondary endogenous
VAP that occurred late during ICU stay. Additionally, exogenous VAP, in which
microorganisms are introduced directly into the lower airways due to poor
hygiene, bypassing the oropharynx, may be an inherent limitation of the silver-
coated tubes. Finally, there was no impact of silver-coated endotracheal tubes on
survival. Neither subglottic drainage nor silver-coated tubes have been associated
with a survival benefit [25].
24.4 Strategies Related to Patient Placement
Implementing a ventilator care bundle can significantly reduce the incidence of

VAP. The original high-impact intervention ventilation care bundle, updated in
2007, consisted of elevating the head of the patients bed to 3045 and daily
sedation hold [26].
24.4.1 Semirecumbent Position
Aspiration of upper-airway secretions is common, even in healthy adults, in the

supine position. Supine patient positioning facilitates aspiration; semirecumbent
positioning decreases it. Infection in patients in the supine position was associated
with simultaneous administration of enteral nutrition and an increased risk of
aspiration of gastric contents. Gastroesophageal reflux occurs less frequently in the
semirecumbent position. Thus, it is recommended that intubated patients be
managed in a semirecumbent position ([30), particularly during feeding. Some
studies found this position to be associated with lower levels of aspiration into the
lower airways but not lower VAP incidence, especially in patients receiving
enterally administered nutrition. Several authors and influential scientific societies

claim the semirecumbent position prevents VAP [27]. Results were based on
experimental studies with radioactive-labeled enteral feeding, which suggested
that endotracheal aspiration of gastric content occurred more frequently in supine
patients than in patients at a 45 angle. However, clinical data supporting the
statement are not robust. Three RCTs and three meta-analyses demonstrated that
semirecumbency did not significantly reduce the odds of either microbiological or
clinically suspected VAP, and mortality rates were not statistically reduced [28].
The semirecumbent position is a low-cost, easily accessible intervention and may
be a more practical and tolerable approach than rotational beds (3045, partic-
ularly during enteral feeding, continues to be recommended). Patients should thus
be nursed in a semirecumbent position [29].
24.5 Conclusion
Despite an increased understanding of HAP/VAP pathogenesis and advances in

diagnosis and treatment, risk, cost, morbility and mortality remain unacceptably
high [30]. Implementing preventive measures and cost-effective strategies to
reduce patient risk and improve outcomes should be a priority. A variety of
measures has been suggested [31], depending on the setting and the individual risk
profile; here we focused on nonantibiotic strategies. The gold standard is based on
a multidisciplinary team approach. To prevent VAP, authors recommend the
orotracheal route of intubation for intubation; a new ventilator circuit for each
patient; circuit changes if the circuit becomes soiled or damaged, but no scheduled
changes; changing heat and moisture exchangers every 57 days or as clinically
indicated; using a new closed endotracheal suctioning system for each patient and
as clinically indicated; subglottic secretion drainage in patients expected to be
mechanically ventilated [72 h; elevating the bed head to 45 when impossible, or
as near to 45 as possible. Rotating beds and oral antiseptic rinses should be
considered. Bacterial filters and antimicrobial agents such as iseganan are not
recommended. We make no recommendations regarding a systematic search for
sinusitis, airway humidification type, tracheostomy timing, prone positioning,
aerosolized antibiotics, mupirocin intranasally, topically and/or intravenously
applied antibiotics, because they are still under debate in medical literature and not
universally recommended. In conclusion, prevention is the most important step
toward improve standard of care and cost-effectiveness related to HAP/VAP [32].
References
1. Restrepo MI, Anzueto A, Arroliga AC et al (2010) Economic burden of ventilator-associated
pneumonia based on total resource utilization. Infect Control Hosp Epidemiol 31:509515
2. Diaz E, Lorente L, Valles J, Rello J (2010) Mechanical ventilation associated pneumonia.
Med Intensiva 34(5):318324
408 A. Gullo et al.
3. Torres A, Ewig S, Lode H, Carlet J, European HAP Working Group (2009) Defining, treating
and preventing hospital acquired pneumonia: European perspective. Intensive Care Med
35:929
4. Omrane R, Eid J, Perreault MM et al (2007) Impact of a protocol for prevention of ventilator-
associated pneumonia. Ann Pharmacother 41:13901396
5. Lorente L, Blot S, Rello J (2010) New issue and controversies in the prevention of ventilator-
associated pneumonia. Am J Respir Crit Care Med 182(7):870876
6. Muscedere J, Dodek P, Keenan S et al, VAP Guidelines Committee and the Canadian Critical
Care Trials Group (2008) Comprehensive evidence-based clinical practice guidelines for
ventilator-associated pneumonia: prevention. J Crit Care 23:126137
7. Rotstein C, Evans G, Born A et al (2008) Clinical practice guidelines for hospital-acquired
pneumonia and ventilator-associated pneumonia in adults. Can J Infect Dis Med Microbiol
19(1):1953
big measure with modest effects. J Hosp Infect 59(3):172179
9. Feider LL, Mitchell P, Bridges E (2010) Oral care practices for orally intubated critically ill
adults. Am J Crit Care 19:175183
10. Panchabhai TS, Dangayach NS, Krishnan A, Karnad DR (2009) Effect of oropharyngeal
cleansing with 0.2% chlorhexidine in critically ill patients: an open label randomized
controlled trial. Chest 135:11501156
11. Chan EY, Ruest A, Meade MO, Cook DJ (2007) Oral decontamination for prevention of
pneumonia in mechanically ventilated adults: systematic review and meta-analysis. BMJ
334(7599):889
12. Tantipong H, Morkchareonpong C, Jaiyindee S, Thamlikitkul V (2009) Randomized
controlled trial and meta-analysis of oral decontamination with 2% chlorhexidine solution for
the prevention of ventilator-associated pneumonia. Infect Control Hosp Epidemiol 30:
101102
13. Kola A, Gastmaier P (2007) Efficacy of oral chlorhexidine in preventing lower respiratory
tract infections. Meta-analysis of randomized controlled trials. J Hosp Infect 66:207216
14. Morrow LE, Kollef MH, Casale TB (2010) Probiotic prophylaxis of ventilator-associated
pneumonia: a blinded, randomized, controlled trial. Am J Respir Crit Care Med 182(8):1058
1064
15. Isakow W, Morrow LE, Kollef MH (2007) Probiotics for preventing and treating nosocomial
infections: review of current evidence and recommendations. Chest 132:286294
16. Silvestri L, van Saene HK, Gregori D et al (2010) Probiotics to prevent ventilator-associated
pneumonia: no robust evidence from randomized controlled trials. Crit Care Med 38(7):
16161617
17. Siempos II, Ntaidou TK, Falagas ME (2010) Impact of the administration of probiotics on the
incidence of ventilator-associated pneumonia: a meta-analysis of randomized controlled
trials. Crit Care Med 38(3):954962
18. Dezfulian C, Shojania K, Collard HR et al (2005) Subglottic secretion drainage for
preventing ventilator associated pneumonia: a meta-analysis. Am J Med 118:1118
19. Silvestri L, Milanese M, Piacente N et al (2008) Impact of subglottic secretion drainage on
ventilator-associated pneumonia and mortality. Systematic review of randomized controlled
trials. In: Proceedings of the 21st anesthesia and ICU symposium Alpe Adria, pp 2629
20. Lorente L, Lecuona M, Jimnez A et al (2007) Influence of an endotracheal tube with
Med 176:10791083
21. Overend TJ, Anderson CM, Brooks D et al (2009) Updating the evidence-base for suctioning
adult patients: a systematic review. Can Respir J 16(3):e6e17
22. Van Saene HKF, Zandstra DF, Petros AJ et al (2009) Infections in ICU: an ongoing
challenge. In: Gullo A, Besso J, Lumb PD, Williams GF (eds) Intensive and critical care
medicine. Springer, Milan, pp 261272
23. Niederman MS (2010) Fighting vampires and ventilator-associated pneumonia: is silver the
magic bullet? Chest 135:10071009
24. Kollef MH, Afessa B, Anzueto A et al, NASCENT Investigation Group (2008) Silver-coated
endotracheal tubes and incidence of ventilator-associated pneumonia: the NASCENT
randomized trial. JAMA 300:805813
25. Silvestri L, van Saene HKF, de la Cal MA, de Gaudio AR (2009) Carriage classification of
pneumonia rather than time improves survival. Chest 136(4):11881189
26. Westwell S (2008) Implementing a ventilator care bundle in an adult intensive care unit. Nurs
Crit Care 13:203207
27. Alexiou VG, Ierodiakonou V, Dimopoulos G, Falagas ME (2009) Impact of patient position
on the incidence of ventilator-associated pneumonia: a meta-analysis of randomized
controlled trials. J Crit Care 24(4):515522
28. Silvestri L, Gregori D, van Saene HK et al (2010) Semirecumbent position to prevent
ventilator-associated pneumonia is not evidence based. J Crit Care 25(1):152153
29. Niel-Weise BS, van den Broek PJ (2009) Semi-recumbent position or not? In: Dutch working
party for infection prevention (WIP). http://www.wip.nl/systrev.asp?nr=12
30. Rosenthal VD, Maki DG, Jamulitrat S et al, for the INICC Members (2010) International
Nosocomial Infection Control Consortium (INICC) report, data summary for 20032008,
issued June 2009. Am J Infect Control 38:95104
31. Efrati S, Deutsch I, Antonelli M et al (2010) Ventilator-associated pneumonia: current status
and future recommendations. J Clin Monit Comput 24(2):161168
32. van Saene HKF, Silvestri L, de la Cal MA, Baines PB (2009) The emperors new clothes; the
fairy tale continues. J Crit Care 24:149152
Impact of Nutritional Route
on Infections: Parenteral Versus 25
Enteral
A. Gullo, C. M. Celestre and A. Paratore
25.1 Introduction
The significance of nutrition in the intensive care unit (ICU) cannot be overstated
[1]. Malnutrition is a marker of poor outcomes and is correlated with longer
hospital stays, nutrition-related complications during and after hospitalization, and
other adverse outcomes. There is a clear association between malnutrition and
postoperative complications. Nutritional status also worsens during hospitalization
in surgical patients and during critical illness; malnutrition rates were higher at
discharge than at admission. Among seriously ill patients, malnutrition is associ-
ated with increased infectious morbidity and prolonged hospital stay. Critical
illness is typically associated with a catabolic stress state in which patients com-
monly demonstrate a systemic inflammatory response. This response is coupled
with complications of increased infectious morbidity, multiorgan dysfunction,
prolonged hospitalization, and disproportionate mortality rates.
Nosocomial infection in critically ill patients is associated with higher mor-
bidity and mortality rates, prolonged ICU and hospital stay, and consequent higher
health care cost [2]. Providing nutritional support has become a standard of care
for critically ill patients. A post hoc analysis suggested preoperative administration
as the most important period. Preoperative supplementation is as effective as
perioperative supplementation in improving outcome [3]. Early nutritional
A. Gullo (&) C. M. Celestre (&) A. Paratore (&)

Department of Anesthesia and Intensive Care,
School of Medicine, University Hospital Catania, Catania, Italy
C. M. Celestre
A. Paratore
412 A. Gullo et al.
support, defined as initiation within the first 2448 h of ICU care, is recommended
by clinical practice guidelines as the first-line nutritional therapy in the ICU.
Nutrition administered enterally (EN) and parenterally (PN) should be initiated
if the caloric goal will be difficult to attain. The most important goal is to con-
tinuously supply the enteric mucosa with useful immunonutrients such as gluta-
mine and fiber to preserve the barrier effect, the mucus layer, and immunological
status of the mucosa, with consequent reduced infection rates [4].
25.2 Critical Illness and the Immune System
Critically ill patients are at high risk for nosocomial infections, which can lead to
organ dysfunction and death. Thus, the benefits and risks of nutritional therapies in
preventing and managing infectious diseases are highly relevant [5, 6]. A major
methodological problem is related to the term critically ill, as it does not refer to
homogenous populations. The prevalence of malnutrition among critically ill
patients, especially those with a protracted clinical course, has remained largely
unchanged over several years and has implications on hospital length of stay, illness
course, and morbidity rates. The profound and stereotypic metabolic response to
critical illness and failure of carers to provide optimal nutritional support therapy
during a patients ICU stay are the principal factors contributing to malnutrition.
Critically ill patients should not be allowed to remain in a state of unopposed
starvation, because this increases morbidity and mortality, particularly in the
setting of multi-organ-system failure [7]. Immunosuppression occurs as a result of
malnutrition, and protein-energy malnutrition has been cited as the major cause of
immunodeficiency worldwide. Critical illness results in derangements of all
components of the acute immune response, which is organized and executed by
innate immunity influenced by the neuroendocrine system. This response starts
with sensing danger by pattern-recognition receptors on immunocompetent cells
and endothelium [8]. The sensed danger signals, through specific signalling
pathways, activate nuclear transcription factor kappa-B and other transcription
factors and gene regulatory systems, which up-regulate proinflammatory mediator
expression. Plasma cascades are also activated, which together with proinflam-
matory mediators further stimulate inflammatory biomarker production. The acute
inflammatory response underlies the pathophysiological mechanisms involved in
the development of multiorgan dysfunction syndrome (MODS). The inflammatory
mediators directly affect organ function mediating the production of nitric oxide
(NO), leading to mitochondrial anergy and cytopathic hypoxia, a condition of
cellular inability to use oxygen.
Understanding the mechanisms of acute immune responses in critical illness is
necessary for the development of therapeutic strategies, understanding molecular
and biological effects of nutrients in maintaining homeostasis has made expo-
nential advances. Perioperative immune modulation using specialized enteral diets
containing specific immunonutrients may improve postoperative outcomes [9].
Some studies investigated the role of nutrition as a modifier of the immune
25 Impact of Nutritional Route on Infections 413
response in specific clinical settings, especially the use of preoperative oral sup-
plementation with immunonutrients in comparison with standard nutrition in
surgical and critically ill patients [10]. Several specific nutrients have been shown
in laboratory and clinical studies to influence nutritional, immunological, and
inflammatory parameters. Immunonutrients are defined as nutrients that provide
specific benefits to the immune system and include glutamine, arginine, long-chain
n-3 polyunsaturated fatty acids, and nucleotides, either alone or in combination.
Usually provided in combination, these nutrients, when added to a standard enteral
formula, seem to improve outcomes by reducing infection rates.
The influence of malnutrition on immunity is complex. Studies investigated the
effects of immunonutrition on morbidity and mortality rates in critical ill patients,
but results are conflicting in terms of study design, population heterogeneity,
treatment timing, and suboptimal delivery of nutrients. In selected patient groups,
the immunonutrition, can be efficacious to reduce infection and mortality rates,
and hospital length of stay. However whether these immunonutrients are benefi-
cial, or should even be used, in critically ill patients remains controversial [11].
Immunonutrition formulae are indicated in specific subgroups of critically ill
patients (e.g. patients with trauma, mild sepsis, surgical patients); this conclusion
is supported by meta-analyses and recent guidelines.
25.3 Infection, Bacterial Translocation, and Sepsis
Nutritional support for ICU patients is important, as adverse effects of malnutrition

are multiple and common and infection is a serious complication [13]. Malnutrition is
accompanied by progressive atrophy and disruption of the intestinal mucosa,
resulting in a protein-losing enteropathy. Loss of protein promotes an edematous
state, complicating patient care, particularly regarding drug administration. Fur-
thermore, lack of nutrition allows translocation of enteric pathogens across the bowel
mucosa and into the circulation, leading to sepsis [14]. Adequate EN or PN nutri-
tional support not only improves clinical outcomes but is essential for recovery from
critical illness. EN also may be important in maintaining normal gut structure and
function, thereby decreasing bacterial translocation (BT) and the risk of systemic
infection. EN is particularly beneficial for promoting gut-barrier integrity and
reducing infection and mortality rates. Failure of gastrointestinal tract (GIT) barrier
function may fuel systemic inflammatory response syndrome (SIRS), sepsis, and
organ failure. In these conditions, BT due to gut-barrier failure has been targeted as
the trigger of a self-sustaining process for systemic infection and MODS.
In critically ill patients with severe homeostasis disorders, many different
factors are involved in the pathophysiology of bacteria and endotoxin transloca-
tion, either in conditions of anatomically intact bowel barrier or in conditions of
altered intestinal mucosa. Preventing BT can be attained both by improving
intestinal function and the host defense mechanism. Therapies against transloca-
tion include nutrition, EN, and selective decontamination of the digestive tract
(SDD) to increase oxygen delivery and avoid hypoperfusion.
414 A. Gullo et al.
A retrospective study demonstrated no significant decrease in the incidence of

fungal infections in critically ill patients receiving SDD between those receiving EN
and total PN (TPN). SDD significantly reduced overall bloodstream infections
(BSI), Gram-negative BSI [15, 16] and overall mortality, without affecting Gram-
positive BSI [17]. The full protocol of systemic SDD reduces mortality rates in
critically ill patients, particularly when successful decontamination is achieved [18].
EN is indicated in postoperative patients (GI surgery), reducing the compli-
cation rate and hospital stay, and in severe pancreatitis promotes resolution of
inflammation and reduces the incidence of infection. Nutrition via the enteral route
is often preferred over central venous or TPN due to its relative ease of admin-
istration and lower cost. As demonstrated in the literature, it is not only safer and
less expensive than PN but modulates an exaggerated cytokine response related to
surgical trauma that leads to increased intestinal permeability, BT, and infection.
Nutrients administered enterally can reach the bowel lumen where enterocytes
draw upon their fuel, preserving the barrier effect and modulating the cytokine
response. Parenteral supply does not achieve this target, as the blood supply of
nutrients is not as important as the luminal supply. It is only via the enteral route
that the barrier effect can be preserved.
Experimental studies show that TPN (enteric starvation) results in rapid and
severe atrophy of gut-associated lymphoid tissue (GALT) and increases BT.
In patients with an intact GIT, early EN is the preferred route of nutritional support
[19]. PN is immunosuppressive and proinflammatory and may be deleterious in
patients with pancreatitis. GALT is the source of most mucosal immunity in
humans. In addition, TPN is associated with impaired B- and T-cell lymphocyte
function, altered leucocyte chemotaxis, and impaired bacterial and fungal killing.
Lack of enteral feeding results in atrophy of the GI mucosa, bacterial overgrowth,
increased intestinal permeability, and translocation of bacteria or bacterial prod-
ucts into the circulation. TPN may therefore promote BT [20].
Two vital components of critical care are the use of central venous catheters and
TPN. The most severely ill patients often require both for survival and recovery.
In the ICU, GI dysfunction associated with multiorgan failure and shock or with
abdominal surgery is not uncommon. Nutrition therefore must not be compromised.
When TPN is suggested for a patient, the risk of infectious complications, especially
infection related to central venous catheters, is often thought greater than potential
benefits. Of all the potentially devastating infectious complications, catheter-related
infection remains the major concern associated with TPN. However, because early
provision of nutritional support improves outcomes in critically ill patients, avoiding
or delaying TPN administration of solutions is potentially harmful [21].
25.4 Assessing Nutritional Status and Score Index
Many tools are used to assess patients nutritional status. Traditional nutrition
assessment tools (albumin, prealbumin, and anthropometry) are not valid in critical
care. Most nutritional assessment techniques are based on their ability to predict
clinical outcomes [22]. However, none of these techniques to accurately measure

nutritional risk has been validated. Using nutritional assessment to predict clinical
outcome can be problematic, because the interaction between malnutrition and
other factors that influence outcomes makes it difficult to isolate any putative
contribution from malnutrition alone. Recognizing specific prognostic factors
might lead to interventions or increased postoperative surveillance, which could
improve outcome. However, the validity of these techniques has also not been
proven.
Two methods can be applied for nutritional assessment: (1) Subjective Global
Assessment (SGA) is used to classify patients into one of three categories of
nutritional status: (a) well nourished, (b) moderately malnourished, or (c)
severely malnourished (2) Nutritional Risk Index (NRI) is a simple equation that
uses serum albumin levels and recent weight loss. An NRI [100 indicates no
malnourishment, 97.5100 mild malnourishment, 83.5\97.5 moderate mal-
nourishment, and \83.5 severe malnourishment. Whereas the NRI uses serum
albumin concentrations, which are influenced by nutritional status in the pre-
sence of inflammatory stress due to a disease, the SGA is not influenced by
serum proteins. On the other hand, NRI uses some laboratory examination,
which requires laboratory costs, but SGA uses only clinical examination, which
can easily be done in several minutes. One might say SGA is the more cost-
effective means of assessing nutritional risk. Both these nutrition tests are,
however, predictive for malnutrition and postoperative complications (infectious-
complications-included pneumonia, intra-abdominal abscess, sepsis, wound
infection, urinary tract infection, pneumonia, atelectasis, pulmonary complica-
tions, anemia, cardiac arrhythmia), length of hospital and ICU stay, and
mechanical ventilation duration [23].
Malnutrition is also associated with a delayed recovery from illness and an
increased rate of complications. Heart failure, respiratory diseases, impaired
immune function, and postoperative wound healing are all influenced by nutri-
tional status. Nutritional assessment includes patient history, physical examination,
anthropometric measurements, laboratory data, and changes in immunocompe-
tence. The validity and sensitivity of the parameters, i.e. serum proteins, albumin,
transferrin, and retinol-binding protein, to assess nutritional status are diminished
for the individual patient. Albumin is commonly thought of as a good indicator of
nutritional status and visceral proteins [24]. Although a variety of nutritional
indices have been found to be valuable in predicting patient outcome when used
alone, there is no consensus on the best method for assessing the nutritional status
of hospitalized patients. There are no sensitive markers available to assess the
influence of malnutrition on the immunocompetence of an individual patient for
clinical purposes. Traditionally, nutritional support in the critically ill population
has been regarded as adjunctive care designed to provide exogenous fuels to
support the patient during the stress response.
416 A. Gullo et al.
25.5 Enteral Versus Parenteral Administration Route
Nutritional support had three main objectives: to preserve lean body mass, to
maintain immune function, and to avert metabolic complications. Nutrition ther-
apy therefore specifically aims at attenuating the metabolic response to stress,
preventing oxidative cellular injury, and favorably modulating the immune
response. Because EN and PN carry both risks and benefits, in the average patient
in the ICU who has no contraindications, the choice of route for nutritional support
may be influenced by several factors.
Enteral tube feeding was first employed in the 1600s and was made popular in
the medical profession by the famous British surgeon John Hunter at the end of the
eighteenth century. The indication for early EN is supported by guidelines
published by the European Society for Clinical Nutrition and Metabolism and
American and Canadian guidelines, which recommend starting administration
within the first 2448 h of admission to the ICU [25]. Short-term access is usually
achieved using nasogastric (NG) or nasojejunal (NJ) tubes; percutaneous endo-
scopic gastrotomy (PEG) or jejunostomy should be considered if feeding is
planned for longer than 1 month. Early EN is recommended for critically ill
patients, with special formulas indicated in specific patient subgroups. Early EN
enhances immunocompetence, reduced clinical infection rates, and maintained gut
structure and function by preserving gut structure/function integrity, balancing
intestinal microflora, maintaining effective local and systemic immunocompe-
tence, and potentially attenuating catabolic stress responses in patients after sur-
gery. There is strong evidence that early enteral feeding prevents infections in a
variety of traumatic and surgical illnesses [26]. There is, however, little support for
similar early feeding in medical illnesses.
Recommendations are to initiate EN as soon as possible whenever the GIT is
functioning. The disadvantage of enteral support is that insufficient energy and
protein coverage can occur. Evidence shows that EN can result in underfeeding
and that nutritional goals are frequently reached only after 1 week. Several
observational studies in long-term ICU patients note that cumulative energy deficit
is related to increased infectious morbidity (infection rate, wound healing,
mechanical ventilation, length of stay, duration of recovery), and costs [27].
Morbidity and mortality rates seem to be linked to such an energy deficit, which
often occurs during the first week of stay. Supplemental PN combined with EN can
be considered to cover energy and protein targets when EN alone fails to achieve
the caloric goal [28]. EN is believed be safer and less expensive than PN. How-
ever, total enteral feeding (TEN) is associated with complications such as diarrhea,
abdominal distention and cramps, and contamination and infection of an enteral
feeding system. In fact EN provides an ideal environment for the development of
bacteria, and the ICU team thus plays a vital role in implementing and maintaining
appropriate standards of care and minimizing risks of bacterial contamination.
Again, if EN is insufficient or fails, PN should be instituted, respecting the often
reduced demand for exogenous substrates in critically ill patients [29].
PN therapy is primarily initiated in patients with a contraindication to use of the

GIT and when infection is one of its frequent and severe complications. PN is
indicated in patients who are unable to obtain adequate nutrients by oral or enteral
routes (EN is not feasible). Other indications are short-gut syndrome, high-output
fistula, prolonged ileus, or bowel obstruction. Indications for TPN are patients
without a functioning GIT or who have disorders requiring complete bowel rest,
such as some stages of Crohns disease or ulcerative colitis, bowel obstruction,
certain pediatric GI disorders (congenital GI anomalies, prolonged diarrhea
regardless of cause, short-bowel syndrome due to surgery). Short-term PN may be
used if a persons digestive system has shut down (peritonitis) and their weight is
low enough to cause concerns about nutrition during an extended hospital stay.
Long-term PN is occasionally used to treat people suffering the extended conse-
quences of an accident, surgery, or digestive disorder. The nutrient solution con-
sists of water and electrolytes (glucose, amino acids, lipids); essential vitamins,
minerals, and trace elements are added or given separately. Previously, lipid
emulsions were given separately. In critical patients in perioperative care, PN
increases the risk of infection when compared with EN or delayed nutrition [30].
A European meta-analysis showed that PN is superior to delayed EN in criti-
cally ill patients. Additional PN thus seems to be the way to avoid cumulative
energy deficit associated with insufficient or no EN. Guidelines issued by the
American Society for Parenteral and Enteral Nutrition (ASPEN) and the European
Society for Clinical Nutrition and Metabolism (ESPEN) are mainly based on
observational studies showing a strong correlation between negative energy bal-
ance and morbiditymortality rates. ASPEN guidelines recommend administration
of PN to nonmalnourished ICU patients receiving some but not adequate EN
during the first 710 days after admission [31]; ESPEN guidelines recommend
compensating the deficit by adding PN after 2448 h [32].
The energy deficit accumulated by underfed ICU patients during the first days
of stay may play an important role in ICU and hospital outcomes for long-term
patients. To reach caloric requirements by artificial nutritional support without
harming the patient is still a subject of debate. Frequent questions related to
artificial nutrition are: Is there significant benefit or risk associated with the route
chosen for nutrition delivery? Would TPN increase the bacteremia rate? Some
studies showed that the nutritional support route in severely ill patients in an ICU
does not affect the rate of infectious complications [33]; however, comorbid
medical conditions and the need of ICU support are more important parameters for
determining the risk of infectious complications. Although many studies have
reported that catheter-associated infective complications are more frequently
elicited by TPN, some studies report that TPN-associated infections can be
attributed to hyperglycemia and caloric overload and that insulin therapy can
alleviate these infections. Others consider PN to be an independent risk factor for
centralvenous catheter-related infection [34]. Studies evaluating the impact of
feeding route and intestinal permeability on BSI and systemic immune responses
concluded that systemic proinflammatory response decreases with increasing EN
and PN weaning [35]. In the majority of critically ill patients, it is practical and
418 A. Gullo et al.
Table 25.1 European Society for Clinical Nutrition and Metabolism (ESPEN) guidelines
modified. Enteral nutrition (EN): intensive care
Recommendations
Indications for EN All patients who are not expected to be on a full oral diet within
3 days should receive EN
Application of early EN The ESPEN committee recommends that hemodynamically stable
critically ill patients who have a functioning gastrointestinal tract
should be fed early (\24 h) using an appropriate amount of feed.
Exogenous energy supply:
During the acute and initial phase of critical illness: in excess of
2025 kcal/kg body weight/day may be associated with a less
favorable outcome
During the anabolic recovery phase, the aim should be to provide
2530 kcal/kg body weight/day
Patient with severe undernutrition should receive EN up 2530 total
Kcal/kg body weight/day. If these target values are not reached,
supplementary parenteral nutrition (PN) should be given. Consider
i.v. administration of metoclopramide or erythromycin in patients
with intolerance to enteral feeding (e.g., with high gastric residuals)
Route Use EN in patients who can be fed via the enteral route
There is no significant difference in the efficacy of jejunal versus
gastric feeding in critically ill patients
Use supplemental PN in patients who cannot be fed sufficiently via the
enteral route
safe to use EN instead of PN. The beneficial effects of EN when compared with PN
are well documented in numerous prospective randomized controlled trials
involving a variety of patient populations in critical illness, including trauma,
burns, head injury, major surgery, and acute pancreatitis. Whereas few studies
show a differential effect on mortality rates, the most consistent outcome effect
from EN is reduced infectious morbidity. When TPN is recommended, the risk of
infectious complications, especially infection related to central venous catheters, is
often thought greater than potential benefits. However, because early provision of
nutritional support improves outcomes in critically ill patients, avoiding or
delaying TPN administration is potentially harmful (Table 25.1).
When selecting the appropriate enteral formulation for the critically ill patient,
the clinician must first decide if the patient is a candidate for a specialty immu-
nomodulating formulation, which are those supplemented with agents such as
arginine, glutamine, nucleic acid, Omega-3 fatty acids, and antioxidants, and
which should be used for the appropriate patient population (major elective sur-
gery, trauma, burns, head and neck cancer, critically ill patients on mechanical
ventilation) [36]. Results strengthen the indication for a special formula in acute
respiratory distress syndrome (ARDS) and acute lung injury [37]. Administration
of probiotic, a combination of antioxidant vitamins, and trace minerals in specific
critically ill patient populations is still being debated (Table 25.2).
Table 25.2 European Society for Clinical Nutrition and Metabolism (ESPEN) guidelines
modified. Special enteral nutrition (EN) formulations for the appropriate patient population
EN formulae and patient population

Type of Whole-protein formulae are appropriate in most patients because no clinical
formula advantage of peptide-based formulae could be shown
Immune-modulating formulae (formulae enriched with arginine, nucleotides, and
x-3 fatty acids) are superior to standard enteral formulae
In elective upper GI surgical patients (see guidelines surgery)
In patients with a mild sepsis (APACHE II \15)
In patients with severe sepsis, however, immune-modulating formulae may
be harmful and are therefore not recommended
In patients with trauma (see guidelines surgery)
In patients with ARDS (formulae containing x-3 fatty acids and
antioxidants)
No recommendation for immunomodulating formulae can be given for burn
patients due to insufficient data. In burn patients, trace elements (Cu, Se, Zn)
should be supplemented in a higher-than-standard dose
ICU patients with very severe illness who do not tolerate [700 ml enteral
formulae per day should not receive an immunomodulating formula enriched with
arginine, nucleotides and x-3 fatty acids
Glutamine should be added to standard enteral formula in:
Burn patients
Trauma patients
There are not sufficient data to support glutamine supplementation in surgical or
heterogenous critically ill patients
GI gastrointestinal tract, APACHE Acute Physiology and Chronic Health Evaluation II, ARDS
acute respiratory distress syndrome, Cu copper, Se selenium, Zn zinc
EN via tube feeding is the preferred method of feeding the critically ill patient,
particularly on those who develop a severe inflammatory response, (i.e. patients
who have failure of at least one organ during ICU stay) and an important means of
counteracting the catabolic state induced by severe diseases. Evidence from pri-
marily low-quality trials shows that EN reduces infections, septic complications,
and length of hospital stay, resulting in a better outcome compared with total PN,
but not does affect noninfectious complication or hospital mortality rates. That
combined nutritional support provides additional benefit on overall outcomes has
yet to be proven in further studies, including outcomes of physical and cognitive
functioning, quality of life, cost-effectiveness, and cost utility. Meta-analyses of
ICU studies showed that PN is not related to a greater mortality rate and may even
be associated with improved survival [38].
25.6 Conclusion
The significance of nutritional support in the hospital setting cannot be overstated

and is part of the standard of care for the critically ill adult patient. Nutrition is a
cornerstone for improved outcomes [39]. Nutritional modulation of the stress
420 A. Gullo et al.
response to critical illness includes early EN using the enteral route, which is seen as a
proactive therapeutic strategy that may reduce disease severity, diminish compli-
cations, decrease length of ICU stay, and favorably impact patient outcome.
Guidelines and studies confirm that EN versus PN, early EN initiation, enteral and
parenteral glutamine administration, and intensive insulin therapy are all associated
with reduced infectious morbidity in critically ill patients [40, 41]. EN compared
with PN results in an important decrease in the incidence of infectious complications
in the critically ill and may be less costly, and thus should be the first choice for
nutritional support in the critically ill. EN reduces infections, septic complications
and length of hospital stay, resulting in a better outcome compared with total PN, but
it not does affect noninfectious complication or hospital mortality rates.
Guidelines for using TPN while avoiding catheter-related infection may markedly
improve its outcome. Appropriately training personnel to care for central venous
catheters is imperative and is an effective method of reducing devastating compli-
cations of infection [42]. Catheters impregnated with antiseptics and coated with
antibiotics are now available. Conceivably, as our understanding of interventions
such as central venous catheters and TPN continues to improve, these catheters can
be used in situations in which the risk of catheter-related infection is high, such as
TPN administration. Evidence-based critical care nutrition clinical practice guide-
lines recommend EN over PN; time to initiate EN; use of formulas enriched with fish
oils; glutamine supplementation; glycemic control; arginine-enriched formulas;
motility agents; timing of supplemental PN when appropriate; delivery of hypoca-
loric PN when appropriate; and adoption of a feeding protocol [43, 44].
Nutritional assessment of the critically ill patient is crucial, as the deterioration
of nutritional status is a key factor in surgical and critically ill patient outcomes.
Perioperative nutrition with specialized enteral diets improves outcome when
compared with standard formulas. One meta-analysis suggests that antioxidant
supplementation is associated with no improvement in infectious complications
but is associated with increased survival rates. New prospects may be possible in
the fight against surgical infections by adding probiotics to EN in order to improve
the microenvironment of the colon. Many unanswered questions remain, however,
the last but not the least of which is the advocated proactive posture for metabolic
support in the ICU [45, 46].
References
1. Cahill NE, Dhaliwal R, Day AG, Jiang X, Heyland DK (2010) Nutrition therapy in the
critical care setting: what is best achievable practice? An international multicenter
observational study. Crit Care Med 38(2):395401
2. Farber MS, Moses J, Korn M (2005) Reducing costs and patient morbidity in the enterally fed
intensive care unit patient. JPEN J Parenter Enteral Nutr 29:S62S69
3. Martindale RG, Maerz LL (2006) Management of perioperative nutrition support. Curr Opin
Crit Care 12:290294
4. Gianotti L (2006) Nutrition and infections. Surg Infect (Larchmt) 7(Suppl 2):S29S32
5. Elia M, Engfer M, Green C, Silk DB (2010) Letter to the editor on the new guidelines for
adult critically ill patients. JPEN J Parenter Enteral Nutr 34(1):105
6. Taylor B, Krenitsky J (2010) Nutrition in the intensive care unit: year in review 20082009.
JPEN J Parenter Enteral Nutr 34(1):2131
7. Saka B, Kaya O, Ozturk GB et al (2010) Malnutrition in the elderly and its relationship with
other geriatric syndromes. Clin Nutr74 29(6):745748
8. Marshall JC, Charbonney E, Gonzalez PD (2008) The immune system in critical illness. Clin
Chest Med 29(4):605616
9. Mizock BA (2010) Immunonutrition and critical illness: an update. Nutrition 26(78):701707
10. Helminem H, Raitanen M, Kellosalo J (2007) Immunonutrition in elective gastrointestinal
surgery patients. Scand J Surg 96:4650
11. Calder PC (2007) Immunonutrition in surgical and critically ill patients. Br J Nutr 98 (Suppl 1):
S1339
12. Klek S, Kulig J, Sierzega M et al (2008) The impact of immunostimulating nutrition on
infectious complications after upper gastrointestinal surgery: a prospective, randomized,
clinical trial. Ann Surg 248(2):212220
13. Dhaliwal R, Heyland DK (2005) Nutrition and infection in the intensive care unit: what does
the evidence show? Curr Opin Crit Care 11(5):461467
14. Kang W, Kudsk KA (2007) Is there evidence that the gut contributes to mucosal immunity in
humans? JPEN J Parenter Enteral Nutr 31(3):246258
15. Silvestri L, van Saene HK, Milanese M, Gregori D (2005) Impact of selective
of randomized controlled trials. Intensive Care Med 31(7):898910
16. Silvestri L, Zandstra DF, van Saene HK, Petros AJ et al (2008) Antifungal prophylaxis in
critically ill patients. Crit Care 12(3):420
17. Silvestri L, van Saene HK, Milanese M et al (2007) Selective decontamination of the
digestive tract reduces bacterial bloodstream infection and mortality in critically ill patients.
Systematic review of randomized, controlled trials. J Hosp Infect 65(3):187203
18. Silvestri L, van Saene HK, Weir I, Gullo A (2009) Survival benefit of the full selective
digestive decontamination regimen. J Crit Care 24(3):474.e7474.e14
19. Doig GS, Heighes PT, Simpson F, Sweetman EA (2010) Early enteral nutrition reduces
mortality in trauma patients requiring intensive care: a meta-analysis of randomised
controlled trials. Injury 42(1):5056
20. Hanna N, Bialowas C, Fernandez C (2010) Septicemia secondary to ileus in trauma patients:
a human model for bacterial translocation. South Med J 103(5):461463
21. Ziegler TR (2009) Parenteral nutrition in the critically ill patient. N Engl J Med
361(11):10881097; comment: N Engl J Med 362(1):83; author reply 8384
22. Kuzu MA, Terzioglu H, Gen V et al (2006) Preoperative nutritional risk assessment in
predicting postoperative outcome in patients undergoing major surgery. World J Surg
30(3):378390
23. Paillaud E, Herbaud S, Caillet P et al (2005) Relations between undernutrition and
nosocomial infections in elderly patients. Age Ageing 34(6):619625
24. Lohsiriwat V, Lohsiriwat D, Boonnuch W et al (2008) Pre-operative hypoalbuminemia is a
major risk factor for postoperative complications following rectal cancer surgery. World J
Gastroenterol 14(8):12481251
25. Kreymann KG (2008) Early nutrition support in critical care: a European perspective. Curr
Opin Clin Nutr Metab Care 11(2):156159
26. Artinian V, Krayem H, DiGiovine B (2006) Effects of early enteral feeding on the outcome of
critically ill mechanically ventilated medical patients. Chest 129:960967
27. Singer P, Pichard C, Heidegger CP, Wernerman J (2010) Considering energy deficit in the
intensive care unit. Curr Opin Clin Nutr Metab Care 13(2):170176
28. Wernerman J (2008) Paradigm of early parenteral nutrition support in combination with
insufficient enteral nutrition. Curr Opin Clin Nutr Metab Care 11(2):160163
422 A. Gullo et al.
29. Nagata S, Fukuzawa K, Iwashita Y et al (2009) Comparison of enteral nutrition with

combined enteral and parenteral nutrition in post-pancreaticoduodenectomy patients: a pilot
study. Nutr J 8:24
30. Tennenberg D, Peris A, di Valvasone S et al (2010) Parenteral nutrition in the critically ill
patient. N Engl J Med 362(1):8182
31. Huhmann MB, August DA (2008) Review of American Society for Parenteral and Enteral
Nutrition (A.S.P.E.N.) clinical guidelines for nutrition support in cancer patients: nutrition
screening and assessment. Nutr Clin Pract 23(2):182188
32. Plauth M, Cabr E, Campillo B for ESPEN (2009) Guidelines on parenteral nutrition. Clin
Nutr 28(4):436444
33. Selcuk H, Kanbay M, Korkmaz M et al (2006) Route of nutrition has no effect on the
development of infectious complications. J Natl Med Assoc 98(12):19631966
34. Hartl WH, Jauch KW, Parhofer K, Rittler P for the Working Group for Developing the
Guidelines for Parenteral Nutrition of the German Association for Nutritional Medicine
(2009) Complications and monitoringGuidelines on parenteral nutrition, Ger Med Sci
18;7:Doc17
35. Cole CR, Frem JC, Schmotzer B et al (2010) The rate of bloodstream infection is high in
infants with short bowel syndrome: relationship with small bowel bacterial overgrowth,
enteral feeding, and inflammatory and immune responses. J Pediatr 156(6):941947
36. Aiko S, Yoshizumi Y, Tsuwano S et al (2005) The effects of immediate enteral feeding with a
formula containing high levels of x-3 fatty acids in patients after surgery for esophageal
cancer. JPEN J Parenter Enteral Nutr 29(3):141147
37. Singer P, Theilla M, Fisher H et al (2006) Benefit of an enteral diet enriched with
eicosapentaenoic acid and gamma-linolenic acid in ventilated patients with acute lung injury.
38. Heidegger CP, Darmon P, Pichard C (2008) Enteral vs. parenteral nutrition for the critically
ill patient: a combined support should be preferred. Curr Opin Crit Care 14(4):408414
39. Hise ME, Halterman K, Gajewski BJ et al (2007) Feeding practices of severely ill intensive
care unit patients: an evaluation of energy sources and clinical outcomes. J Am Diet Assoc
107(3):458465
40. Martindale RG, McClave SA, Vanek VW et al (2009) Guidelines for the provision and
assessment of nutrition support therapy in the adult critically ill patient: Society of Critical
Care Medicine and American Society for Parenteral and Enteral Nutrition: executive
summary. Crit Care Med 37(5):17571761
41. Btaiche IF, Chan L-N, Pleva M, Kraft MD (2010) Critical illness, gastrointestinal
complications, and medication therapy during enteral feeding in critically ill adult patients.
Nutr Clin Pract 25(1):3249
42. Sriram K, Cyriac T, Fogg LF (2010) Effect of nutritional support team restructuring on the
use of parenteral nutrition. Nutrition 26(78):735739
43. Zhou M, Martindale RG (2007) Arginine in the critical care setting. J Nutr 137(6 Suppl
2):1687S1692S
44. Dos Santos RG, Viana ML, Generoso SV et al (2010) Glutamine supplementation decreases
intestinal permeability and preserves gut mucosa integrity in an experimental mouse model.
JPEN J Parenter Enteral Nutr 34(4):408413
45. Scurlock C, Mechanick JI (2008) Early nutrition support in the intensive care unit: a US
perspective. Curr Opin Clin Nutr Metab Care 11(2):152155
46. Bistrian BR (2010) Comment on guidelines for the provision and assessment of nutrition
support therapy in the adult critically ill patient. JPEN J Parenter Enteral Nutr 34(3):348349;
author reply 350352
Gut Mucosal Protection
in the Critically Ill Patient: Toward 26
an Integrated Clinical Strategy
D. F. Zandstra, P. H. J. van der Voort, K. Thorburn

26.1 Introduction
Traditionally, the critically ill patient is considered at risk for the development of
stress-ulcer-related bleeding (SURB) from the intestinal canal. Back-diffusion of
H+ is considered the most important mechanism in the etiology of SURB [1].
In the intensive care unit (ICU), routine administration of specific prophylaxis using
antacids, histamine-2 (H2) receptor antagonists, and cytoprotective agents has been
practiced for the past 40 years. Several meta-analyses have shown a reduction from
15 to 5% in SURB after administration of antacids and H2 receptor antagonists
[25]. Despite this reduced incidence, overt SURB contributes to both morbidity
and mortality [6]. The risk of death is increased only when the bleeding occurs
for [4 weeks after ICU admission, which suggests a different pathophysiology
for early- and late-onset bleeding [7]. Early bleeding is associated with acute
hemodynamic disturbances, such as shock and incomplete resuscitation, whereas
late bleeding is due to sepsis and multiple organ dysfunction syndrome (MODS).
The reported incidence of SURB in adult patients who do not receive prophylaxis
has also fallen, from 60% in 1978 to 0.65% in 1994 [8]. Since 1994, the incidence
of SURB has remained constant (15%), irrespective of whether the patient
D. F. Zandstra (&)
Onze Lieve Vrouwe Gasthuis, Amsterdam,
The Netherlands
received prophylaxis or not [8]. Reduction in mortality rates has never been
demonstrated using a prophylactic regimen [9, 10].
26.3 SURB Pathogenesis in the Critically Ill
SURB pathogenesis is complex. Under normal conditions, the mucosa is protected

against potentially aggressive factors (e.g., gastric acid, pepsin, and bile) by a
mucus layer. The most important mechanisms involved mucosal damage are
vascular injury, gastric acid, ischemia, sepsis, endotoxin and inflammation in
serious infection, Helicobacter pylori gastritis, and parenteral feeding.
26.3.1 Vascular Injury
Time-sequence studies in animal experiments have shown that vascular injury due to
various substances is the rate-limiting step in the pathogenesis of mucosal injury.
Vascular injury of the superficial mucosal capillaries may lead to reduced or even
absent blood flow, with subsequent edema and congestion of the mucosal layer.
Whereas the blood supply remains intact, the self-restoring capacity of the mucosal
layer is enormous. When the vascular injury is minimal or absent, the lesions of the
epithelial surface are covered by migrating cuboidal cells within 60 min [11].
26.3.2 Gastric Acid
Back-diffusion of H+ ions was considered to be of the utmost importance in SURB

development [1]. Recent work questions the pivotal role of gastric acid, however.
A meta-analysis showed that ranitidine is no better than placebo in reducing the
incidence of SURB [12]. H2 receptor antagonists are contraindicated in the critically
ill, as they contribute to infection [13]. The H2 receptors are not located on the acid-
producing cells of the stomach but on the surface of the immunocompetent cells in the
mucosal lining [14]. These cells orchestrate the local immune system in the gut (enteric
minibrain). H2 receptor antagonists downregulate the gut immune system, explaining
the higher incidence of infections [13]. Critically ill patients often have gastric exo-
crine failure, with a subsequent gastric pH [4 [15]. SURB is frequently observed in
this particular condition where the critically ill are unable to produce gastric acid;
therefore gastric acid can be considered as a secondary factor in SURB pathogenesis.
The primary step is impairment of mucosal barrier functions due to ischemia.
26.3.3 Ischemia
Mucosal ischemia plays a key role in SURB pathogenesis [16, 17] and occurs
during shock, sepsis, and endotoxemia. The fundus and corpus of the stomach are
particularly sensitive to ischemia. Mucosal cells do not have the ability to store
26 Gut Mucosal Protection in the Critically Ill Patient 425
glycogen as an energy substrate. During ischemia, these cells cannot maintain

cellular function by anaerobic glycolysis [18]. The reduced splanchnic blood flow
leads to decreased oxygen delivery and an energy deficit that causes impairment
of the barrier function of the mucosa. Back-diffusion of acid may occur, with
subsequent mucosal damage and erosions or ulcers.
Hypovolemia, low cardiac output, and vasoconstrictive medication may all
reduce splanchnic perfusion. Hypovolemia, particularly in surgical patients, is
associated with an increased risk of gastrointestinal complications [19, 20]. There
is a relationship between endotoxemia and the central venous pressure [21].
Mechanical ventilation on its own is associated with impaired intestinal perfusion
in a high percentage of patients and may lead to impaired oxygenation [20].
Circulation therapy aimed at adequate microcirculation (flow) rather than blood
pressure remains crucial in SURB prevention. Ischemia of the mucosa may persist
due to arteriovenous shunting in the submucosa. Consequently, reperfusion injury
may occur. Several experimental studies have shown that radical oxygen scav-
engers attenuate mucosal damage [22, 23].
26.3.4 Sepsis
The link between SURB and infection becomes increasingly clear in clinical
practice. The incidence of SURB was 20% in patients with ineffectively treated
pneumonias, whereas it was \10% in adequately treated patients [24]. There-
fore, pneumonia can be considered a major factor in SURB pathogenesis.
Ventilator-associated pneumonia in the critically ill patient can effectively be
prevented by selective decontamination of the digestive tract (SDD) [25]. The
potential role of sepsis in the pathogenesis of coagulation disorders that impact
on SURB should be emphasized [10]. Experimental studies that focused on the
role of sepsis in SURB pathogenesis are scarce. However, clinical experience
shows that sepsis is the most important risk factor of SURB [7]. Various
mechanisms are involved: (1) release of vasoactive substances, including
serotonin, histamine, adrenaline, and noradrenaline, promotes vasoconstriction
following endotoxin release; (2) hemodynamic changes may cause hypotension
during the early phase of sepsis and may lead to a redistribution of blood flow
between and within organs [26]. During sepsis, a substantial deficit in nutrient
flow to the mucosa has been observed despite fluid resuscitation [27]. Increased
arteriovenous shunting may cause tissue hypoxemia irrespective of increased
blood flow [28].
26.3.5 Endotoxin and Inflammation in Serious Infections
Many clinical conditions are associated with increased intestinal permeability and
may lead to endotoxemia, which is related to the degree of permeability [21].
Increased permeability can be prevented by vasodilators and adequate
intravascular volume [21]. This inflammatory insult involves activation of

leukocytes, which may then adhere to the endothelium of the venular side of
the microcirculation. This may block microcirculation by means of subsequent
stasis and ulceration [29]. Serious infections such as pneumonia have been
shown to result in high levels of interleukin (IL)-6 and tumor necrosis factor
(TNF). These inflammatory mediators increase the expression of adhesion
molecules on endothelium and leukocytes [30]. Expression of the adhesion
molecules can be prevented with steroids. Several anti-inflammatory agents
reduce mucosal damage by preventing leukocyte adherence to the venular
endothelium [31, 32]. There is good evidence for the role of activated leuko-
cytes in SURB development. Steroids prevent adhesion of leukocytes and
thereby mucosal damage. Although steroids were believed to promote SURB,
risk analysis failed to substantiate this [10].
26.3.6 Gastritis by Helicobacter pylori
H. pylori infection is the most important factor in gastric and duodenal ulceration
pathogenesis in the non-ICU patient. In this population, the prevalence of H. pylori
infection is about 30%. Its potential role in SURB pathogenesis in critically ill
patients was investigated in detail in our unit [33, 3537]. The following endpoints
were studied:
1. a new method for detecting H. pylori in mechanically ventilated patients using
a urea breath test;
2. the incidence of H. pylori infection in patients requiring acute admission to the
ICU using the urea breath test and serology;
3. the impact of parenterally and enterally administered antibiotics in SDD on
H. pylori eradication;
4. the incidence and risk factors of mucosal damage by direct endoscopy at ICU
admission;
5. H. pylori infection as an ICU occupational hazard.
Serological tests are unreliable for identifying H. pylori infections ventilated
ICU patients; the laser-assisted ratio analyzer (LARA) 13C-urea breath test
(Alimenterics, NJ, USA) is more reliable [34]. The incidence of H. pylori infec-
tions in acute patients admitted to the ICU is 40% [33]. A relationship was found
between the degree of gastric mucosal lesions on admission and the presence of
active infection [35]. H. pylori infection of the stomach causes extensive micro-
vascular leucocyte adhesion and migration into the tissue parenchyma, and sig-
nificant inflammatory cell infiltration is found [36]. Antimicrobials may be
effective in treating H. pylori infection of the gastric mucosa, e.g., enterally
administered nonabsorbable antibiotics [37]. SDD reduces SURB, in part by
eliminating H. pylori. Transmission of H. pylori from infected patients to nursing
staff was shown in our study [38].
26.3.7 Feeding
Enteral feeding can cause infectious morbidity [39]. Delay in initiating enteral
feeding promotes mucosal vasoconstriction due to reduced prostaglandin synthe-
sis. Stressful conditions combined with food deprivation increases the incidence of
ulceration. Enteral feeding protects the mucosa by: (1) neutralizing acid, (2)
stimulating perfusion, and (3) providing intraluminal substrate as fuel for the
colonic mucosal cell. Three studies show that enteral feeding has a beneficial effect
on the incidence of stress ulceration [4042]. As the critically ill patient is
deprived of adequate swallowing abilities, impaired salivary flow into the stomach
results in extreme low gastric levels of nitric oxide (NO). Nitrate-rich saliva
enhances bactericidal effects of gastric juices [43, 44]. Dietary nitrate increases
gastric mucosal blood flow and mucosal bactericidal defence [45]. Future studies
are needed to demonstrate whether NO donors applied topically or systemically
prevent SURB.
26.4 Clinical Approach to Controlling SURB
The decreased incidence of SURB in ICU patients without prophylaxis is due to

improved treatment in recent decades. The main factors are: optimizing the
microcirculation, effective infection control, early enteral feeding, and controlling
inflammation/infection due to H. pylori (Table 26.1). Aggressive correction of
hypovolemia is achieved by adequate fluid replacement and preventing vasocon-
striction using vasodilators. Vasoconstricting agents should be administered
cautiously. Treating low-cardiac-output syndrome is based inodilator administra-
tion following diagnosis. Steroids prevent the release of adhesion molecules, so
leukocytes do not adhere to the endothelium.
H2 receptor antagonists are contraindicated in the critically ill, as they are risk
factors for infection. The risk of ventilator-associated pneumonia increases with
increased severity of critical illness [46, 47], and the subsequent hyperinflamma-
tory status predisposes the patient to SURB. Effective prevention of serious
infection using SDD reduces the incidence of SURB [48] and significantly reduces
infectious morbidity and mortality rates.
Enteral feeding protects the gut mucosa and contributes to infection control.
SDD using enterally and parenterally administered antibiotics controls H. pylori
infections, and SDD clears fecal endotoxin, thus preventing endotoxin absorption
and subsequent inflammation. Prokinetic use can also promote rapid elimination of
the fecal endotoxin pool [49].
Clinical measures are important to optimize microcirculation, prevent noso-
comial infections, and reduce hyperinflammation, which in turn reduces the
incidence of SURB. This strategy, comprising continuous vasodilators to prevent
mucosal ischemia, SDD to control infection and reduce intestinal endotoxin, and
steroids to reduce leukocyte adherence, was associated with a SURB incidence of
Table 26.1 Protocol for preventing stress-ulcer-related bleeding (SURB)
Circulation
Prevent microcirculatory stasis by aggressive correction of hypovolemia and of low cardiac
output
Prevent arteriovenous shunting using vasodilators
Preven corpuscular endothelial adhesion using steroids
Prevent infection
Prevent serious infections, including pneumonia and septicemia, by selective decontamination
of the digestive tract and immunonutrition
Intestinal contents
Remove fecal endotoxins using selective decontamination of the digestive tract
Prevent stasis of intestinal contents using enema and neostigmine
Enteral feeding to ensure mucosal energy supply
Control gastritis due to Helicobacter pylori
Eliminate H. pylori using enterally and parenterally administered antimicrobials in selective
decontamination of the digestive tract protocol
Provide nitric oxide donors enterally or systemically
Consider side effects of proton-pump inhibitors on gut ecology
0.6% in a 7-year cohort of critically ill ICU patients needing mechanical venti-
lation [48 h [35, 48]. These data, in combination with the lack of efficacy of
specific prophylaxis in most recent studies, support the concept that SURB pre-
vention is not only a matter of intragastric-acid control. The proposed guidelines in
the Surviving Sepsis Campaign for preventing stress ulceration therefore can be
challenged [50, 51]. Also, the potential role of NO donors via saliva into the
stomach or systemically administered contributes to improved gastric mucosal
microcirculation and increased intragastric microbial killing of potentially path-
ogenic microorganisms [43]. This challenges the perceived pivotal role of gastric
acid in the pathogenesis of SURB in the critically ill. Enhancement of endogenous
cytoprotective mechanisms by NO donors as a strategy for SURB prevention,
however, needs further evaluation.
Increased susceptibility of potentially the hazardous side effect of increased
transmucosal gastric leak caused by of proton-pump inhibitors (PPIs) and the effects
on the human (protective) microbiota may further contribute to a reconsideration of
PPIs being used to prevent SURB [52, 53]. PPIs impair intestinal colonization
resistance and increase susceptibility for Clostridium difficile infection [53].
26.5 Pediatric Experience
There is a paucity of information concerning the incidence of clinically significant

SURB in pediatric practice, with the reported incidence ranging from 0.4 to 1.6%
[54, 55]. The basic pathophysiology is similar in children and adults [56]. There is
no clear evidence of improvement with prophylaxis [5]. In a large pediatric ICU in
Liverpool, UK, there were only three cases of clinically important SURB in
3,238 admissions ([85% ventilated), including 772 postcardiopulmonary bypass
(personal communication). No prophylaxis is used other than infection prevention

by SDD in combination with a policy of early enteral feeding. A meta-analysis
suggests a weak benefit from PPIs, but high-quality evidence to guide clinical
practice is still limited [57].
26.6 Conclusions
During the past 40 years, clinical strategies controlling SURB have shifted from
interventions to neutralize intragastric acid (antacids) and inhibit acid synthesis
(H2 receptor antagonists, PPIs) toward maneuvers aimed at maintaining and
improving microcirculatory perfusion (aggressive circulatory support, including
vasodilators). The use of anti-inflammatory drugs to prevent microcirculatory
sludging is important to maintain adequate microcirculation. Infections are a
pivotal pathogenetic step in SURB and MODS development [58]. These infec-
tions, however, can be prevented by SDD (25). This therapy also suppresses
H. pylori gastritis, which plays a role in SURB pathogenesis. Implementing SDD
guarantees a consistently low incidence (\1%) of SURB in patients requiring a
minimum of 2 days of mechanical ventilation. Dietary nitrate is considered
important to maintain gastric mucosal blood flow and mucosal defence, including
antimicrobial defence. A better understanding of the side effects of PPIs and H2
antagonists should lead to a more restrictive use of these classes of drugs in the
critically ill, as overuse is acknowledged [59, 60].
References
1. Skillman JJ, Gould SA, Chung W et al (1970) The gastric mucosal barrier: clinical and
experimental studies in critically ill and normal man, and in the rabbit. Ann Surg 172:
564582
2. Cook DJ, Witt LJ, Cook RJ (1991) Stress ulcer prophylaxis in the critically ill: a meta-
analysis. Am J Med 91:519527
3. Shuman RB, Schuster DP, Zuckerman GR (1987) Prophylactic therapy for stress bleeding: a
reappraisal. Ann Intern Med 106:562567
4. Lacroix J, Infante-Rivard C, Jecinek M et al (1989) Prophylaxis of upper gastrointestinal
bleeding in intensive care units. Crit Care Med 17:862869
5. Tryba M (1991) Der Einfluss praeventiver Massnahmen auf Morbiditaet und Mortalitaet von
Intensivpatienten. Anaesthesiol Intensivmed Notfallmed Schmerzther (Suppl)1:4253
6. Cook DJ, Griffith LE, Walter S et al (2001) Canadian Critical Care Trials Group. The
attributable mortality and length of intensive care unit stay of clinically important
gastrointestinal bleeding in critically ill patients. Crit Care 5:368375
7. Zandstra DF, Stoutenbeek CP, Oudemans-van Straaten HM (1989) Pathogenesis of stress
ulcer bleeding. In: van Saene HKF et al (eds) Update in intensive and emergency medicine.
Springer, Berlin, pp 166172
8. Zandstra DF (1995) Stress ulceration in the critically ill. No longer a problem? Proceedings
SMART. Springer, Milan, pp 3032
9. Tryba M (1991) Stress bleeding prophylaxis 1990a meta-analysis. Clin J Gastroenterol
13(Suppl 2):4455
10. Cook DJ, Fuller HD, Guyatt GH et al (1994) Risk factors for gastrointestinal bleeding in
11. Lacey ER, Ito S (1984) Rapid epithelial restitution of the rat gastric mucosa after ethanol
injury. Lab Invest 51:573583
12. Messori A, Trippoli S, Vaiani M et al (2000) Bleeding and pneumonia in intensive care
patients given ranitidine and sucralfate for prevention of stress ulcer: metaanalysis of
randomized controlled trials. BMJ 321:11031106
13. OKeefe GE, Gentilello LM, Maier RV (1998) Incidence of infectious complications
associated with the use of histamine2-receptor antagonists in critically ill trauma patients.
Ann Surg 227:120112
14. Mezey E, Palkovits M (1992) Localisation of targets for anti-ulcer drugs in cells of the
immune system. Science 258:16621665
15. Stannard VA, Hutchinson A, Morris DL et al (1988) Gastric exocrine failure in critically ill
patients: incidence and associated features. BMJ 296:155156
16. Fiddian-Green RG, McCough E, Pittenger G et al (1983) Predictive value of intramural pH
and other risk factors for massive bleeding from stress ulceration. Gastroenterology 85:
613620
17. Hottenrott C, Seufert RM, Becker H (1978) The role of ischaemia in the pathogenesis of
stress induced gastric lesions in piglets. Surg Gynecol Obstet 146:217220
18. Menguy R, Desbaillets L, Masters YF (1974) Mechanisms of stress: influence of
hypovolaemic shock on energy metabolism in the gastric mucosa. Gastroenterology 66:4655
19. Christenson JT, Schmuziger M, Maurice J et al (1994) Gastrointestinal complications after
coronary artery bypass grafting. J Thorac Cardiovasc Surg 108:899906
20. Love R, Choe E, Lipton H et al (1995) Positive end-expiratory pressure decreased mesenteric
blood flow despite normalization of cardiac output. J Trauma 39:195199
21. Oudemans-van Straaten HM, Jansen PG, Velthuis H et al (1996) Endotoxaemia and
postoperative hypermetabolism in coronary artery bypass surgery: the role of ketanserin. Br J
Anaesth 77:473479
22. Bhattacharjee M, Bhattacharjee S, Gupta A et al (2002) Critical role of an endogenous gastric
peroxidase in controlling oxidative damage in H. pylori-mediated and nonmediated gastric
ulcer. Free Radic Biol Med 32:731743
23. Biswas K, Bandyopadhyay U, Chattopadhyay I et al (2003) A novel antioxidant and
antiapoptotic role of omeprazole to block gastric ulcer through scavenging of hydroxyl
radical. J Biol Chem 278:10991001
24. Alvarez Lerma F, ICU Acquired Infection Group (1996) Modification of empiric antibiotic
treatment in patients with acquired pneumonia in the ICU. Intensive Care Med 22:387394
25. Liberati A, DAmico R, Pifferi S et al. (2009) Antibiotic prophylaxis to reduce respiratory
4:CD000022
26. Lang CH, Bagby GJ, Ferguson JL et al (1984) Cardiac output and redistribution of organ
blood flow in hypermetabolic sepsis. Am J Physiol 246:331
27. Kreimeier U, Yang ZH, Messmer K (1988) The role of fluid replacement in acute endotoxin
shock. In: Kox W, Bihari D (eds) Shock and the adult respiratory distress syndrome.
Springer, Berlin
28. Bowen JC, LeDoux JC, Harkin GV (1979) Evidence for pathophysiologic arteriovenous
shunting in the pathogenesis of acute gastric mucosal ulceration. Adv Shock Res 1:3542
29. McCafferty DM, Granger DN, Wallace JL (1995) Indomethacin induced gastric injury and
leucocyte adherence in arthritic versus healthy rats. Gastroenterology 109:11731180
30. Cush JJ, Rothlein R, Lindley HB et al (1993) Increased levels of circulating intercellular
adhesion molecules in the sera of patients with rheumatoid arthritis. Arthritis Rheum
36:10981102
31. Low J, Grabow D, Sommers C et al (1995) Cytoprotective effects of CI-959 in the rat gastric
mucosa; modulation of leukocyte adhesion. Gastroenterology 109:12241223
32. Santucci L, Firucci S, Giansanti M et al (1994) Pentoxifylline prevents indomethacin induced

acute mucosal damage in rats: role of TNF. Gut 35:909915
33. van der Voort PHJ, van der Hulst RWM, Zandstra DF et al (1999) Detection of Helicobacter
pylori in mechanically ventilated patients: the LARA-13C-Urea breath test and serology.
Clin Intensive Care 10:9195
34. van der Voort PH, van der Hulst RW, Zandstra DF et al (2001) Prevalence of Helicobacter
pylori infection in stress-induced gastric mucosal injury. Intensive Care Med 27:6873
35. van der Voort PHJ, van der Hulst RW, Zandstra DF et al (2001) Suppression of Helicobacter
pylori infection during intensive care stay: related to stress ulcer bleeding incidence? J Crit
Care 16:182187
36. van der Voort PH, van der Hulst RW, Zandstra DF et al (2000) In vitro susceptibility of
Helicobacter pylori to, and in vivo suppression by antimicrobials used in selective
decontamination of the digestive tract. J Antimicrob Chemother 46:803805
37. Suzuki H, Suzuki M, Imeada H et al (2009) Helicobacter pylori and microcirculation.
Microcirculation 16:547558
38. van der Voort PH, van der Hulst RW, Zandstra DF et al (2001) Gut decontamination of
critically ill patients reduces Helicobacter pylori acquisition by intensive care nurses. J Hosp
Infect 47:4145
39. Beale RJ, Bryg DJ, Bihari DJ (1999) Immunonutrition in the critically ill: a systematic review
of clinical outcome. Crit Care Med 27:27992805
40. Ephgrave KS, Scott DL, Ong A et al (2000) Are gastric, jejunal, or both forms of enteral
feeding gastroprotective during stress? J Surg Res 88:17
41. Ephgrave KS, Kleiman-Wexler RL, Adair CG (1990) Enteral nutrients prevent stress
ulceration and increase intragastric volume. Crit Care Med 18:621624
42. Raff T, Germann G, Hartmann B (1997) The value of early enteral nutrition in the
prophylaxis of stress ulceration in the severely burned patient. Burns 23:313318
43. Bjorne H, Petersson J, Phillipson H et al (2004) Nitrite in saliva increases gastric mucosal
bloodflow. J Clin Invest 113:106114
44. Bjorne H, Weitzberg E, Lundberg JO (2006) Intraqastric generation of antimicrobial nitrogen
oxides from saliva. Physiological and therapeutic considerations. Free Radic Biol Med
41:14041412
45. Petersson J, Phillipson M, Jansson EA (2007) Dietary nitrate increases gastric mucosal
bloodflow and mucosal defence. Am J Physiol Gastrointest Liver Physiol 292:g718g724
46. Vincent JL, Rello J, Marshall J et al (2009) International study of the prevalence and
outcomes of infection in intensive care units. JAMA 302:23232329
47. Zandstra DF, Petros AJ, van Saene HK (2009) The final gasp from the European experts.
48. Zandstra DF, Stoutenbeek CP (1994) The virtual absence of stress-ulcer related bleeding in
ICU patients receiving prolonged mechanical ventilation without any prophylaxis: a
prospective cohort study. Intensive Care Med 20:335340
resolves critical illness-related colonic ileus in intensive care patients with multiple organ
failurea prospective, double-blind, placebo-controlled trial. Intensive Care Med 27:
822827
50. Van SpreuwelVerheijen M, Bosman RJ, Oudemansvan Straaten HM et al (2006)
Is the surviving sepsis campaign for stress ulcer prophylaxis justified? Intensive Care Med
32(Suppl 1):s23
51. Zandstra DF, van der Voort PH (2004) A more appropriate critical appraisal of the available
evidence? Crit Care Med 32:21662167
52. Vesper BJ, Jawdi A, Altman KW et al (2009) The effect of proton pump inhibitors on the
human microbiota. Curr Drug Metab 10:8489
53. Murray LJ, Gabello M, Rudolph DS et al (2009) Transmucosal gastric leak induced by proton
pump inhibitors. Dig Dis Sci 54:14081417
54. Lacroix J, Nadeau D, Laberge S et al (1992) Frequency of upper gastrointestinal bleeding in a

pediatric intensive care unit. Crit Care Med 20:3542
55. Chaibou M, Tucci M, Dugas M-A et al (1998) Clinically significant upper gastrointestinal
bleeding in a pediatric intensive care unit: a prospective study. Pediatrics 102:933938
56. Crill CM, Hak EB (1999) Upper gastrointestinal bleeding in critically ill pediatric patients.
Pharmacotherapy 19:162180
57. Reveiz L, Guerrero-Lozano R, Camacho A et al (2010) Stress ulcer, gastritis, and
gastrointestinal bleeding prophylaxis in critically ill pediatric patients: a systematic review.
Pediatr Crit Care Med 11:124132
58. Silvestri L, van Saene HK, Zandstra DF et al (2010) Impact of selective decontamination of
59. Heidelbaugh JJ, Goldberg KL, Inadomi JM (2009) Overutilization op protonpump inhibitors:
a review of cost-effectiveness and risk in PPI. Am J Gastroenterol 104:527532
60. Farrel CP, Mercogliano G, Kuntz CL (2009) Overuse of stress ulcer prophylaxis in the
critical care setting and beyond. J Crit Care 25(2):214220
Selective Decontamination
of the Digestive Tract: Role 27
of the Pharmacist
N. J. Reilly, A. J. Nunn and K. Pollock
27.1 Introduction
Selective decontamination of the digestive tract (SDD) is an antimicrobial

prophylactic strategy aimed at preventing both endogenous and exogenous
infections of the lower airways and bloodstream in patients requiring treatment in
the intensive care unit (ICU) [1]. SDD therapy is based on the concept of microbial
carriage and is directed at limiting infections in ICU patients caused by the
carriage of 15 potentially pathogenic microorganisms (PPMs) (Table 27.1).
The SDD concept was introduced in 1984 by Stoutenbeek et al. [2] with the aim
of controlling primary endogenous, secondary endogenous and exogenous infec-
tions, caused by both normal and abnormal PPMs, by means of a parenterally
administered antibiotic, topical nonabsorbable antimicrobials, high levels of
hygiene and surveillance cultures [3]. To date, there have been 60 randomised
controlled trials (RCTs) evaluating SDD [417] and ten meta-analyses assessing
its efficacy [1824], providing evidence that SDD significantly reduces infection in
patients in the ICU and improves patient survival. Based on the available evidence,
the British Society for Antimicrobial Chemotherapy recently recommended that
SDD be considered for ICU patients expected to require mechanical ventilation
C48 h in order to prevent ventilator-associated pneumonia (VAP) [25].
Due to the lack of commercially available topical formulations, application of
the SDD concept to hospital practice has required substantial input from phar-
macists in both the development and extemporaneous preparation of SDD
formulations. This chapter aims to provide the reader with comprehensive infor-
mation on the pharmaceutical technology involved in implementing SDD and how
N. J. Reilly (&)
Pharmacy Department, Alder Hey Childrens NHS Foundation Trust,
434 N. J. Reilly et al.
Table 27.1 Potentially pathogenic microorganisms (PPMs) causing infection in intensive care
unit (ICU) patients
Previously healthy host (normal PPM) Host with severe underlying disease (abnormal PPM)
1. Streptococcus pneumoniae 7. Klebsiella spp.
2. Haemophilus influenzae 8. Proteus spp.
3. Moraxella catarrhalis 9. Morganella spp.
4. Escherichia coli 10. Enterobacter spp.
5. Staphylococcus aureus 11. Citrobacter spp.
6. Candida albicans 12. Serratia spp.
13. Acinetobacter spp.
14. Pseudomonas spp.
15. Methicillin-resistant Staphylococcus aureus (MRSA)
the role of the pharmacist is essential if this concept is to be successfully applied to

reduce carriage, colonisation and infection rates in ICU patients.
27.2 Nonabsorbable Antimicrobials: The PTA Regimen
SDD aims to convert the abnormal carrier state into normal carriage using nonab-
sorbable antimicrobials. A paste or gel is applied into the inside of the lower cheeks
to prevent or eradicate pre-existing oral carriage of abnormal PPMs, i.e., to decon-
taminate the oropharynx. Suspensions are administered via the nasogastric tube
to decontaminate the stomach and gut and control abnormal carriage of aerobic
Gram-negative bacilli (AGNB), particularly Pseudomonas aeruginosa. Nonab-
sorbable antimicrobials are given throughout ICU treatment to control morbidity and
mortality associated with late secondary endogenous infection, which is caused by
abnormal bacteria not present in admission flora but acquired during ICU treatment.
It generally occurs after 1 week in the ICU and accounts for 30% of infections.
Topical antimicrobials are also used to control exogenous infections, e.g., by
applying paste to a tracheostomy site to control lower airway infections. Grafted
burn wounds may also be targeted by mixing topical antimicrobial agents with a
translucent hydrogel (Aquaform) and applying it in thin layers over fine mesh
gauze to the wound site [26]. Exogenous infection is caused by abnormal potential
pathogens never carried in the patients digestive tract but introduced directly into
the patient. Exogenous infection may occur at any time during ICU treatment and
accounts for 15% of infections.
The ideal SDD regimen should use antimicrobials that are nontoxic, inexpen-
sive, palatable and microbiologically active in the presence of faeces, saliva or
antacids. The most widely used SDD regimen is that of the Groningen group [2],
who devised a protocol using polymyxin E, tobramycin and amphotericin B
27 Selective Decontamination of the Digestive Tract 435
(PTA), which is applied as an oral paste and suspension to treat both the throat and
gut, respectively. In the UK, polymyxin E is known as colistin.
The combination of polymyxin E and tobramycin is synergistic against Proteus
and Pseudomonas spp. It is the most potent antipseudomonal combination asso-
ciated with an effective clearance of Pseudomonas from the gut. Emergence of
resistance to polymyxin is rare. Although there are bacteria-producing tobramycin-
inactivating enzymes, polymyxin is thought to protect tobramycin from being
destroyed by these enzymes [27]. Tobramycin is the preferred aminoglycoside
because it is intrinsically most active against Pseudomonas and is minimally
inactivated by saliva and faeces. It also has useful activity against Staphylococcus
aureus [27]. Both agents absorb endotoxins released by AGNB in the gut. This
feature is important because endotoxin can be absorbed from the gut of seriously
ill patients, producing fever, release of inflammatory mediators and shock. In order
to control yeast overgrowth, amphotericin B is also included in the regime. It is
intrinsically the most potent antifungal, but there is a high rate of inactivation in
the gut requiring the use of high doses [27]. By design, the PTA regimen is
inactive against indigenous flora, such as streptococci viridans, enterococci,
coagulase-negative staphylococci (CNS) and anaerobes [27], each of which is
necessary for normal physiological gut function. Modifications to the PTA regime
may need to be made. For example, SDD was not designed to cover methicillin-
resistant Staphylococcus aureus (MRSA), as MRSA was not a significant problem
in the early 1980s. Therefore, in cases of MRSA endemicity, enterally adminis-
tered vancomycin should be added to the PTA regime [28, 29]. In the case of
endemicity of AGNB producing extended-spectrum beta lactamases (ESBL)
resistant to tobramycin, tobramycin may have to be replaced by paromomycin, and
if Serratia endemicity is present, both polymyxin and tobramycin should be
replaced by paromomycin [29].
27.3 Indications
In the paediatric ICU (PICU) setting, SDD is indicated to [30]:

1. control infection: in patients expected to require ventilation for [48 h, SDD
reduces the risk of endogenous infections;
2. control inflammation: following cardiac surgery (and particularly cardiopulmo-
nary bypass), patients develop the systemic inflammation response syndrome
(SIRS) characterized by high levels of cytokines, leucocytes and increased
C-reactive protein (CRP); polymyxin E and tobramycin may neutralize endo-
toxin and therefore reduce gut overgrowth;
3. prevent resistance: SDD eradicates multi-drug-resistant microorganisms
detected from surveillance swabs of throat and rectum.
Once initiated, SDD therapy should be continued until the patient has been
extubated.
27.4 Application Method
27.4.1 SDD of the Oropharynx
SDD gel containing 2% PTA is used. A pea-sized application is evenly smeared

inside the lower cheeks four times a day(a 5-g tube lasts approximately 5 days).
27.4.2 SDD of the Gastrointestinal Tract
1. The antimicrobials are given orally or through the nasogastric tube. When the
patient requires gastric suction, the nasogastric tube is clamped and the suction
is discontinued for 1 h.
2. When the normal anatomy of the gastrointestinal tract is disrupted (gastro- or
intestinal-fistulae or colostomy), each (blind) loop should be separately treated
with approximately one half of the oral PTA dose in an adequate volume [30].
27.4.3 SDD of Tracheostomy and Gastrostomy Sites
When being ventilated in the PICU, patients with tracheostomies or gastrostomies

should have SDD paste containing 2% PTA applied four times a day to these sites,
in addition to oropharyngeal and gastrointestinal SDD treatments. Preparation
dosages used are shown in Table 27.2 [30].
27.5 Pharmaceutical Technology
The development of SDD medication has depended upon close collaboration

between pharmacists and microbiologists. For decontamination of the gut in the
unconscious adult and in children, liquid preparations for administration via a
nasogastric tube are required. For decontamination of the oropharynx, formula-
tions of oral gels, pastes, pastilles and lozenges have been developed and prepared
by hospital and academic pharmacists [31]. Solutions or suspensions of the three
antimicrobials are used for oral and nasogastric administration [30]. Coated tablets
of colistin have also been prepared for positioned release into the colon [32].
No commercial company has yet shown more than a passing interest in the further
development, licensing or marketing of these preparations, and this may have
influenced the rate of development of the SDD concept.
As all preparations are made extemporaneously in the hospital pharmacy or
manufacturing units, they are classed as hospital specials and as such are
unlicensed. In using these unlicensed products in clinical practice, responsibility
for safety and efficacy lies with the prescriber, and responsibility for quality lies
with the pharmacist. Ideally, to reduce risks to the patient, it would be preferable to
Table 27.2 Selective decontamination of the digestive tract (SDD): doses [30]
Age Medication Dosage

Over 12 years Polymyxin E/Colistin 100 mg : 3,000,000
units four times daily
Tobramycin base 80 mg four times daily
Amphotericin B 500 mg four times daily
SDD gel A pea-sized application
four times daily
512 years Polymyxin E/Colistin 50 mg : 1,500,000
units four times daily
four times daily
\5 years Polymyxin E/Colistin 25 mg : 750,000 units
four times daily
four times daily
The dose administered is dependent on gut volume. To be administered 30 min before feeds and
not with feeds. Sucralfate significantly reduces concentrations of colistin, tobramycin and
amphotericin B; therefore, separate administration by 24 h [34]
NB Colistin doses in milligrams are specified as colistin base, where 1 mg = 30,000 units, 1 mg
tobramycin base is equivalent to 1.5 mg tobramycin sulphate, 1 mg colistin base is equivalent to
1.5 mg colistin sulphate
use medicines that have been appropriately researched and subjected to the
scrutiny of the medicines licensing process. One other problem with extempora-
neous production in the UK is that pharmacy departments are limited in the
quantities of products that they can prepare, unless they have a manufacturers
licence (specials) issued by the medicines and healthcare products regulatory
agency (MHRA).
Because SDD medication is not commercially available or supported by the
usual manufacturers marketing activity, scientific background to the formulations
(assay, rheology) is limited. To date, little work has been undertaken to develop
assays for PTA ingredients when combined in mixtures or formulations for local
and oral application. Assays should indicate activity of the constituent antimi-
crobials, and therefore, a microbiological assay is preferred to techniques such as
high performance liquid chromatography. Suitable microorganisms must be
selected for their resistance to the other PTA components, lack of reversion to
sensitivity must be demonstrated and diffusion from the gel to agar plate must be
matched to that from standard antimicrobial solutions.
Despite commercial apathy, hospital-based research is continuing into areas

such as modifications to the gel, development of colonic-positioned release
products and production of placebo products for trial work. It should be noted,
however, that extemporaneous production of SDD formulations does lead to an
increase in a hospital pharmacy departments workload and that involvement in
research programmes requires a considerable amount of research and develop-
ment, which would ideally be best met by commercial support.
27.6 Choice of Formulation
27.6.1 Oropharynx
To abolish PPM carriage in the oropharynx, a contact time of at least 20 min is

required for effective decontamination [31]. Pastes, gels, pastilles and lozenges
therefore offer advantages over suspensions, aerosols and oral rinses, as they have
a longer contact time [33]. The ideal formulation for use in the oropharynx should
therefore have a prolonged contact time with the oral mucosa, should release the
antimicrobials into the oropharynx throughout the contact period, should be
pharmaceutically stable and should be acceptable to the patient [31].
There are four suitable formulations recommended for use in the oropharynx:
paste, gel, lozenge and pastille. It should be noted that the shelf life for these products
has been assigned on the basis of in vivo microbiological experience and not by
traditional pharmaceutical methodology. For use in our PICU, the majority of these
preparations are formulated and supplied to us by specialist hospital pharmacy
manufacturing units at the Western Infirmary in Glasgow, Scotland, and Stepping
Hill Hospital (Stockport Pharmaceuticals), Manchester, UK; however, products may
be available from other manufacturing units. La Botica de Argensola is a manu-
facturing unit in Madrid, Spain, that can supply SDD products to all countries in the
European Union and to South America (www.famaciamagistral.com).
27.6.1.1 Paste
Paste has advantages in that it is easy to produce, has good adhesion to the mucosa,
has a prolonged release of medicament, is sugar free, is stable and has a well-
proven formula [33] (Table 27.3). Although paste effectively eliminates AGNB, it
has several drawbacks. It has an unpleasant taste and appearance, can cause
considerable drying of the oral mucosa and can be difficult to remove, occasionally
causing trauma to the mucosa. Because of this, it has poor acceptability with
patients, staff and relatives. We restrict use of the paste to applications around
tracheostomy and gastrostomy sites, where the adherence and barrier properties of
the paste are particularly useful.
Table 27.3 Formula for selective decontamination of the digestive tract (SDD) paste [33]
Ingredients g
Amphotericin B powder Ph. Eur 2 (adjusted for potency)
Tobramycin sulphate USP 2
Colistin sulphate Ph. Eur 2
Liquid paraffin Ph. Eur 10

Orabase paste (ConvaTec) to 100
Shelf life 1 month; store at room temperature
SDD paste is prepared by mixing each of the antimicrobial powders with 10% w/w liquid paraffin
and gradually incorporating Orabase . Vigorous mixing causes the Orabase to crack
Ph. Eur = European Pharmacopoeia
USP = United States Pharmacopoeia
Table 27.4 Formula selective decontamination of the digestive tract (SDD): pastilles [33]
Ingredients
Gelatine Ph. Eur 500 g
Glycerol Ph. Eur 700 g
Sucrose Ph. Eur 100 g
Sodium benzoate Ph. Eur 4g
Distilled water 600 ml
Lemon oil Ph. Eur 2 ml
Blackcurrant powder 10 g
Amphotericin B powder Ph. Eur 15.88 g (adjusted for potency)
Colistin Sulphate Ph. Eur 12.95 g
Tobramycin sulphate USP 12.95 g
To prepare SDD pastilles, soak the gelatine and water and heat to melt. Add most of the glycerol
and the other ingredients, except the antibiotics, and mix well. Heat for 30 min and then add the
antibiotics, wetting the amphotericin with glycerine
Shelf life 6 months; store at room temperature
A 1.5 g pastille contains 12 mg amphotericin and 10 mg tobramycin and colistin
27.6.1.2 Pastille
The advantages of the pastille are that it can be flavoured easily, it has good release
characteristics, it is easy to use and it is acceptable to the conscious patient
(Table 27.4). Studies in cancer patients have demonstrated that SDD pastilles are
effective in eradicating the carriage of AGNB and yeasts, reducing the incidence of
radiation mucositis and yeast stomatitis [34]. The pastille, however, has limited,
use as it cannot be used in comatose patients, it has high sugar content and
therefore cannot be used in diabetics and it is unsuitable for young children. It is
Table 27.5 Formula for selective decontamination of the digestive tract (SDD): lozenge [36]
Ingredients mg
Antibiotic mixture
Amphotericin B powder Ph. Eur 10
Colistin Sulphate Ph. Eur 2
Tobramycin sulphate USP 1.6
Basic mixture
Citric acid 40
Calcium diphosphate 150
Saccharine 795
Shelf life 3 months; store at room temperature
To prepare SDD lozenges, two powder mixtures are prepared. After sieving of the powders, the
total mixture is mixed in a Turbula mixer (90 rpm) for 15 min. The total powder mixture is then
moistened with 25 ml water, and thereafter, 25 ml sodium carboxymethylcellulose (low vis-
cosity) is added. Further mixing then takes place for 10 min, after which the moistened powder is
dried for a minimum of 4 h at 40C. The dried mixture is then mashed through a 0.75-mm sieve,
and the resulting granulate is sieved further through a 0.4 mm sieve to eliminate the fine powder.
Prior to the final tableting stage, the granulate is mixed with 0.5% magnesium stearate and 2.5%
talc in the Turbula mixer for 2 min
difficult to produce, as many hospital pharmacy departments do not have appro-

priate facilities, and therefore this preparation needs to be made in a specialised
manufacturing unit.
27.6.1.3 Lozenge
The advantages and disadvantages of the lozenge are very similar to that of the
pastille (Table 27.5). In cancer patients, eradication of AGNB and yeasts by SDD
lozenges has been shown by Spijkervet et al. [35] to take up to 3 weeks, therefore
comparing poorly with eradication rates of 34 days that have been achieved in
ICU patients with paste. One explanation for these differing eradication rates
would be that patients in the ICU are usually unconscious, permitting proper
application of sticky paste, whereas patients with head and neck cancer suck their
lozenges four times daily and eat normal, unsterilised food. Poor compliance
within this group of patients, a lower standard of personal hygiene and an altered
oropharyngeal anatomy may also contribute to the longer eradication times.
Lozenges, when sucked, will take approximately 15 min to dissolve in vivo.
Hence, they do not achieve the same length of contact time with the buccal mucosa
as the paste or gel, and this is therefore another factor contributing to poorer
eradication rates. These results would suggest a need for new formulations to be
developed to allow a more protracted and hence more effective delivery of the
antimicrobials to the buccal mucosa of ambulant patients [36].
Table 27.6 Formula for selective decontamination of the digestive tract (SDD): 2% gel [33]
Ingredients
Sodium carboxymethylcellulose Ph. Eur 10 g
Glycerol Ph. Eur 60 ml

Nipasept 600 mg
Concentrated peppermint water BP 10 ml
Distilled water 200 ml
Amphotericin B powder Ph. Eur 6 g (adjusted for potency)
Colistin sulphate Ph. Eur 6g
Tobramycin sulphate USP 6g
Shelf life one month; store at 28C
A gel base is prepared from sodium carboxymethylcellulose, propylene glycol or glycerol and
Nipasept solution. Peppermint water is added for flavour; 2% by weight each of amphotericin
B, colistin sulphate and tobramycin sulphate are stirred into the cold gel base, and the resulting
SDD gel is packed into aluminium tubes using a syringe and tube to aid filling. Colistin
sulphomethate sodium has been used in the gel in some centres, where a stringy texture has
been noted when using the sulphate. This method involves using the commercial powder for
injection (Colomycin injection, manufactured by Pharmax)
BP = British Pharmacopoeia
27.6.1.4 Gel
Gel, as with paste, is sugar free, but it is an improvement on paste in that it is more
palatable, much easier to remove and does not dry the oropharyngeal mucosa [33]
(Table 27.6). The efficacy of this formulation appears to be equal to that of paste,
but patient acceptability with the gel is higher and therefore compliance is better.
The gel presents problems in that it is difficult to produce and at present its long-
term stability is unknown.
27.6.2 Gastrointestinal Tract
Decontamination of the gut is not difficult. Most ICU patients have gut stasis, there
is good contact time between antimicrobials and organisms and it can be dem-
onstrated by surveillance culture that decontamination of the oesophagus, stomach
and small intestine occurs within 3 days. However, to clear PPMs from the large
intestine, there must be functioning gut motility. Due to this controlling factor,
decontamination of the colon and rectum may be longer and may take up to
7 days. A formulation for use in the gastrointestinal tract should ideally release the
antimicrobials in the terminal ileum and provide high concentrations of these
antimicrobials in the colon and rectum. The product should also be easy to use,
acceptable to patients and have good pharmaceutical stability [30]. To date, only
oral suspensions and solutions have been used (Tables 27.6, 27.7, 27.8 and 27.9).
Table 27.7 Formula for amphotericin suspension 100 mg/ml [33]
Ingredients
Amphotericin B Ph. Eur 500 g (adjusted for potency)
Sodium Citrate Ph. Eur 25 g
Sodium carboxymethylcellulose Ph. Eur 12.5 g
VEEGUM K 25 g
Citric acid monohydrate Ph. Eur 5.95 g to adjust pH to 5.5
Saccharin powder Ph. Eur 470 mg

Lycasin 750 ml
Nipasept 6.5 g
Concentrated peppermint water BP 125 ml
Distilled water to 5330 g
Shelf life 6 months; store at 28C
A suspension is prepared with sodium carboxymethylcellulose as suspending and thickening
agent, distilled water, and Lycasin as the sweetener. VEEGUM K (hydrated magnesium
aluminium silicate) is then added as the anticaking agent. Amphotericin B powder is added
gradually to this mixture, stirring after each addition. Nipasept used as a preservative is
dissolved in the concentrated peppermint water and then added to the suspension. Peppermint
water is added to mask the metallic taste of amphotericin. Finally, saccharin powder is added to
improve palatability and citric acid monohydrate to adjust the pH of the suspension to 5.5. The
remaining water is then added to make up to the final weight
Although Crome [31], in 1988, suggested that research was progressing into the
development of colon-positioned release tablets/capsules for use in conscious
patients, this research does not appear to have lead to the availability or wide-
spread use of these preparations [37]. We used coated colonic colistin capsules
only once [32] in conscious, immunocompromised patients who had functioning
guts. The aim of coated colonic preparations is that they should allow release of
the capsule contents at a pH of approximately 77.2, resulting in disintegration in
the ascending colon. The resultant local delivery of antibacterial agents into the
colon is thought to achieve faecal flora suppression and that, by bypassing the
oesophagus and stomach, gastrointestinal side effects such as nausea and vomiting
should be reduced.
The widespread use of the orally administered solutions and suspensions in
PICU therefore continues. The advantages of these products is that they are stable
[33], easy to produce and can be given via a nasogastric tube and are therefore
suitable to give to an unconscious patient [31]. Problems with poor taste, however,
particularly of colistin, may decrease compliance in conscious patients. Diarrhoea
reportedly may be a problem with SDD treatment [5], but a recent publication [38]
observed that, in fact, patients receiving SDD had more days with normal stools
than days with diarrhoea.
Table 27.8 Constituents of

Ingredients
colistin oral solution 100 mg
base (3 megaunits) in 1 ml Colistin sulphate powder Ph. Eur
(formula intellectual property
Polysorbate 80 Ph. Eur
of Stockport Pharmaceuticals,
Manchester, UK) [42] Sodium methyl hydroxybenzoate (Nipagin M Sodium) REG.
SPEC
Sodium propyl hydroxybenzoate (Nipasol M Sodium) REG.
SPEC
Orange syrup BP
Dilute hydrochloric acid 10%
Purified water
100 mg colistin base : 150 mg colistin sulphate
Shelf life 12 months
Colistin for SDD is prescribed as base; use within 4 weeks of
opening; store \25C
A solution is prepared by adding polysorbate 80, followed by
colistin powder, to purified water; dissolution is achieved by
gentle mixing with an SL2T mixer at 1,100 rpm. Sodium
hydroxybenzoate and sodium propyl hydroxybenzoate are
dissolved separately in heated purified water, and then these
two solutions are mixed together and filtered. The filtrate is
added very slowly to the gently mixing colistin solution.
Orange syrup is then added to flavour, and hydrochloric acid is
added to adjust the pH to 4.55.5. (A commercial preparation
of colistin is available (Colomycin ) from Forest. It contains
250,000 units of colistin in 5 ml and is stable for 2 weeks when
reconstituted)
REG SPEC = Regional Specification
There have been isolated reports of accumulation of enterally administered

antibiotics as an adverse effect of SDD resulting in obstructive masses developing in
either the oesophagus or jejunum of affected patients [39]. Accumulation of residual
buccally applied SDD paste may be the reason for oesophageal obstruction, whereas
residual SDD paste and/or suspension could be the cause of jejunal masses. The
incidence of clinically relevant obstructive accumulation is probably very low, but to
avoid this adverse effect, it is recommended that there be thorough removal of
residual SDD paste before each new application. The likelihood of accumulation
may also be reduced by using an SDD gel rather than paste and liquids.
In neonates, vomiting and other gastrointestinal problems have been seen,
particularly if concentrated solutions are administered to an empty stomach [40].
The osmolality of amphotericin suspension is high, approximately 2,000 mOsm/l
compared with an osmolality of approximately 430 mOsm/l with a colistin
25 mg/ml solution and 60 mOsm/l with a 20 mg/ml solution of tobramycin sul-
phate. Because hyperosmolar medications have been associated with an increased
Table 27.9 Constituents of

Ingredients
tobramycin sulphate oral
solution 122 mg in 1 ml Tobramycin sulphate powder USP
(equivalent to tobramycin
Sodium methyl hydroxybenzoate (Nipagin M sodium) REG.
base 80 mg in 1 ml) [42].
SPEC
(Formula intellectual
property of Stockport Sodium propyl hydroxybenzoate (Nipasol M sodium) REG.
Pharmaceuticals, Manchester, SPEC
UK)
Orange syrup BP
Dilute hydrochloric acid 10%
Purified water
Tobramycin for selective decontamination of the digestive
tract (SDD) is prescribed as base. Shelf life 2 years; store at
28C; use within 4 weeks of opening
Sodium methyl hydroxybenzoate and sodium propyl
hydroxybenzoate are dissolved separately in heated, purified
water, and the resultant solution is then filtered. Tobramycin
sulphate powder is also dissolved in heated purified water
using a Silverson mixer (3,800 rpm) to aid dissolution, and
then the hydroxybenzoates filtrate solution is added very
slowly to the tobramycin solution into the vortex created by the
mixer. The mixer speed is increased to 6,000 rpm, orange
syrup is added, followed by purified water to make up to
volume. Finally, dilute hydrochloric acid 10% is added to
adjust the pH to between 4.6 and 5.1
REG SPEC = Regional Specification
incidence of necrotising enterocolitis (NEC) [40], it is therefore recommended that

all three PTA ingredients be diluted in water.
Observers also suggest that PTA may delay gastric emptying/absorption of feeds,
and the binding of colistin and tobramycin by food proteins is a factor known to reduce
the lethal intestinal antibiotic levels required to eradicate AGNB [41]. This interfer-
ence with food has been shown as the reason for SDD failure when oral contaminated
feeds were given to a premature neonate [41]. In our PICU, we recommend that SDD
doses be administered 30 min before feeds and not with feeds [30].
Although there are a limited number of commercially available products that
could be used for SDD, problems with pack sizes, expiry dates and strengths of the
commercial products mean that most ICUs prefer to use extemporaneously pre-
pared products. For example, colistin syrup is available commercially at a strength
of 250,000 units in 5 ml (Colomycin, Forest). This means that to give a dose of
3,000,000 units (:100 mg) for a patient [12 years of age, 60 ml would be
required to be administered per dose, i.e., 240 ml/day. The same dose using an
extemporaneously prepared colistin 100 mg/ml oral solution would equate to a
volume of 1 ml, i.e., 4 ml/day. Recent discontinuations of some of the products
that were available commercially (Fungilin suspension and tablets) may have
also hindered the application of the SDD concept to clinical practice.
Table 27.10 Costs of selective decontamination of the digestive tract (SDD) hospital prepara-
tions (June 2010)
Drug Supplier Pack Cost Total daily dose Cost/

size () [12 years of age day ()
Amphotericin 100 mg/ml Western Infirmary, 100 ml 80 2,000 mg 16
suspension Glasgow
Colistin oral solution Stockport 30 ml 42.37 400 mg 5.65
100 mg/ml Pharmaceuticals,
Manchester
Tobramycin base oral Stockport 60 ml 51 320 mg 3.4
solution 80 mg/ml Pharmaceuticals,
Manchester
SDD gel Western Infirmary, 5g 9.50 A pea-sized 1.9
Glasgow application four
times a day
SDD paste Western Infirmary, 5g 14.70 A pea-sized 2.94
Glasgow application four
times a day
SDD pastilles Western Infirmary, 28 45.15 1 pastille four 6.45
Glasgow times a day
27.7 Costs of Decontaminating Agents
Formal costbenefit analyses of SDD in ICU patients have not been performed.
In theory, successful prevention of infection may make ICU more cost effective in
that reduced infection rates secondary to SDD may lead to a shorter patient ICU
stay and lower ICU, parenterally administered antibiotic usage and microbiology
laboratory costs. It has not been firmly established whether these potential savings
offset the additional costs that the SDD regimen incurs through the use of non-
absorbable antimicrobials, systemic antimicrobials and additional microbiological
cultures [43].
The costs of drugs used in the PTA regimen using products prepared by hospital
manufacturing units are listed in Table 27.10, and the costs of commercially
available products are listed in Table 27.11 [44]. The cost of treating a patient
[12 years of age using hospital or commercial products is also shown, although a
complete daily cost comparison cannot be made due to the lack of commercial
products.
It should be noted that the costs of hospital-made products have escalated
dramatically over the past few years, perhaps because of further research and
development costs in the production of more stable products, lack of competition
from commercial products in driving prices down and problems affecting the
Table 27.11 Costs of commercially available preparations (British National Formulary March
2010) [44]
Drug Supplier Pack Cost Total daily Cost/

size () dose [12 years of age day ()
Colomycin syrup (colistin sulphate Forest 80 ml 3.48 400 mg(:12,000,000 10.44
250,000 units/5 ml) units)
Colomycin tablets (colistin sulphate Forest 50 58.28 400 mg(:12,000,000 9.32
1.5 million unit tablets) units)
Fungilin lozenges (amphotericin Squibb 60 3.53 1 lozenge four times a 0.24
10 mg lozenges) day
Table 27.12 Possible sources of ingredients
Ingredient Supplier
Sodium carboxymethylcellulose Ph. Eur; sodium benzoate Ph. Eur, Fagron UK
tobramycin sulphate powder USP, amphotericin B powder Ph. Eur;
Tel/fax: +44-845-6522525
glycerol Ph. Eur; citric acid monohydrate Ph. Eur; sodium citrate
Ph. Eur; liquid paraffin Ph. Eur; peppermint concentrate BP www.fagron.co.uk
Colistin sulphate powder Ph. Eur. Duchefa Farma
Haarlem, The Netherlands.
Tel: +31-235-319093

Nipasept Clariant UK Ltd
Calverley Lane, Horsforth,
Leeds, UK.
LS18 4RP
Tel: +44-113-2584646
Orange syrup BP J.M. Loveridge Ltd
Southbrook Rd,
Southampton, UK.
SO15 1BH
Tel: +44-170-3228411
Fax:+44-170-3639836
global economy. However, due to the commercial unavailability of tobramycin

oral solution, an SDD gel andmore recentlyamphotericin suspension and
tablets (discontinued by the manufacturers), it would not be possible to follow the
PTA regimen without the provision of hospital-made products.
One factor that may make it difficult for a hospital pharmacy to begin manu-
facturing preparations for SDD regimes is sourcing the raw ingredients necessary
to make the products. Table 27.12 therefore provides useful information to
overcome this problem.
27.8 Conclusion
Although application of the SDD concept to intensive care medicine has been
proven to reduce ICU-related morbidity and mortality rates, and despite a publi-
cation validating SDD as an evidence-based medicine manoeuvre [25], the SDD
approach is still not widely used in ICUs. Proposed reasons for this may be:
1. SDD is contrary to the traditional concept that prophylaxis creates resistance.
2. A primacy of opinion over evidence.
3. Opinion leaders control the medical media.
4. SDD formulations are not marketed by the pharmaceutical industry.
5. There is little physicianpharmaceutical industry interaction to stimulate
industry interest in manufacturing SDD products.
In a climate with a lack of commercial products, the necessary extemporaneous
production of SDD formulations must be undertaken by a pharmacy department
that is able to commit to the additional workload that this entails. This means that,
at present, the formulation and supply role of the hospital pharmacist is vital in
order to facilitate application of the SDD concept to clinical practice.
References
1. Silvestri L, de la Cal MA, van Saene HKF (2009) Selective decontamination of the digestive
tract [SDD]. Twenty five years of European experience. In: Gullo A, Besso J, Lumb PD,
Williams GF (eds) Intensive care and critical care medicine. Springer, Milan, pp 273283
2. Stoutenbeek CP, Van Saene HKF, Miranda DR, Zandstra DF (1984) The effect of selective
3. Silvestri L, Mannucci F, van Saene HK (2000) Selective decontamination of the digestive
tract: a life saver. J Hosp Infect 45:185190
4. Abdel-Razek SM, Abdel-Khalek AH, Allam AM et al (2000) Impact of selective
gastrointestinal decontamination on mortality and morbidity in severely burned patients.
Ann Burns Fire Disasters 13:213216
in severely burned paediatric patients. Burns 27:439445
pneumonia by oral decontamination: a prospective, randomised, double- blind, placebo-
randomised, placebo-controlled study. Crit Care Med 30:3843
8. Camus C, Bellissant E, Sebille V et al (2005) Prevention of acquired infections in intubated
patients with the combination of two decontamination regimens. Crit Care Med 33:307314
9. de Jonge E, Schultz MJ, Spanjaard L et al (2003) Effects of selective decontamination of the

11. Farran L, Llop J, Sans M, Kriesler E et al (2008) Efficacy of enteral decontamination in the
prevention of anastomotic dehiscence and pulmonary infection in esophagogastric surgery.
Dis Esophagus 21:159164
12. Hellinger WC, Yao JD, Alvarez S et al (2002) A randomised, prospective, double-blinded
evaluation of selective bowel decontamination in liver transplantation. Transplantation 73:
19041909
topical antibiotic prophylaxis on the incidence of infections, organ dysfunctions and mortality
in critically ill surgical patients: a prospective, stratified, randomised, double-blind, placebo-
controlled clinical trial. Am J Respir Crit Care Med 166:10291037
14. Pneumatikos I, Koulouras V, Nathanail C (2002) Selective decontamination of subglottic area
in mechanically ventilated patients with multiple trauma. Intensive Care Med 28:432437
verses selective bowel decontamination: a controlled trial in liver transplant recipients.
16. Stoutenbeek CP, van Saene HK, Little RA, Whitehead A, Working Group on Selective
Decontamination of the Digestive Tract (2007) The effect of selective decontamination of the
digestive tract on mortality in multiple trauma patients: a multicenter randomised controlled
trial. Intensive Care Med 33:261270
digestive tract to prevent postoperative infection: a randomised placebo-controlled trial in
18. Safdar N, Said A, Lucey MR (2004) The role of selective decontamination for reducing
infection in patients undergoing liver transplantation: a systematic review and meta-analysis.
Liver Transpl 10:817827
19. Liberati A, DAmico R Pifferi S et al (2004) Antibiotic prophylaxis to reduce respiratory
tract infections and mortality in adults receiving intensive care (Cochrane Review).
In: The Cochrane Library Issue 1, Chichester, UK: Wiley
20. Silvestri L, van Saene HKF, Milanese M, Gregori D (2005) Impact of selective
decontamination of the digestive tract on fungal carriage and infection; systematic review
of randomised controlled trials. Intensive Care Med 31:898910
digestive tract reduces bloodstream infections and mortality in critically ill patients: a
systematic review of randomised controlled trials. J Hosp Infect 65:187203
22. Silvestri L, van Saene HKF, Casarin AL et al (2008) Impact of selective decontamination of
the digestive tract on carriage and infection due to gram-negative and gram-positive bacteria.
Systematic review of randomised controlled trials. Anaesth Intensive Care 36:324338
23. Silvestri L, van Saene HKF, Weir I, Gullo A (2009) Survival benefit of the full selective
digestive decontamination regimen. J Crit Care 24:474.e714
tract infections and mortality in adults receiving intensive care. Cochrane Database Syst Rev,
CD000022
25. Masteron RG, Galloway A, French G et al (2008) Guidelines for the management of hospital-
acquired pneumonia in the UK: report of the working party on hospitalacquired pneumonia
of the British Society for antimicrobial chemotherapy. J Antimicrob Chemother 62:534
26. Rommes JH, Selby A, Zandstra DF (2005) Therapy of infection. In: van Saene HKF, Silvestri
L, de la Cal MA (eds) Infection control in the intensive care unit, 2nd edn edn. Springer,
Milan, pp 515533
27. Sanchez M, Pizer BP, Alcock SR (2005) Enteral antimicrobials. In: van Saene HKF, Silvestri
L, de la Cal MA (eds) Infection control in the intensive care unit, 2nd edn edn. Springer,
Milan, pp 171187
Control 30:391399
29. van Saene HKF, Petros AJ, Sarginson RE et al (2009) Is selective decontamination of the
digestive tract a solution to the anti-microbial problem in the UK? JICS 10(2):8689
30. Guidelines for the use of selective digestive decontamination (SDD) (1999, updated 2010).
Alder Hey Childrens NHS Foundation Trust, Liverpool, UK
31. Crome D (1989) Pharmaceutical technology in selective decontamination. In: van Saene
HKF, Stoutenbeek CP, Lawin P, McA Ledingham I (eds) Infection control by selective
decontamination. Springer, Berlin, pp 109112
32. Data on file. Alder Hey Childrens NHS Foundation Trust, Liverpool, UK
33. Data on file. Western Infirmary, Glasgow, UK
34. Symonds RP, Mcillroy P, Khorrami J et al (1996) The reduction of radiation mucositis by
selective decontamination antibiotic pastilles: a placebo-controlled double-blind trial. Br J
Cancer 74:312317
35. Spijkervet FKL, van Saene HKF, van Saene JJM et al (1991) Effect of selective elimination
of the oral flora on mucositis in irradiated head and neck cancer patients. J Surg Oncol
46:167173
36. Data on file. Organon, Oss, The Netherlands
37. van Saene JJM (1990) Colonic delivery of polymyxin E and four quinolones for flora
suppression. PhD thesis, University of Groningen. PAL, Amsterdam
38. van der Spoel JI, Schultz MJ, van der Voort PHJ, de Jonge E (2006) Influence of severity
of illness, medication and selective decontamination on defecation. Intensive Care Med 32:
875880
39. Smit MJ, van der Spoel JI, de Smet AMGA et al (2007) Accumulation of oral antibiotics as
an adverse effect of selective decontamination of the digestive tract: a series of three cases.
Intensive Care Med 33: 20252026
40. Jew RK, Owen D, Kaufman D, Balmer D (1997) Osmolality of commonly used medications
and formulas in the neonatal intensive care unit. Nutr Clin Practice 12:158163
41. van Saene HKF, Stoutenbeek CP, Faber-Nijholt R, van Saene JJM (1992) Selective
decontamination of the digestive tract contributes to the control of disseminated intravascular
coagulation in severe liver impairment. J Pediatr Gastroenterol Nutr 14(4):436442
42. Personal Communication (2010) Stockport pharmaceuticals, Stepping Hill Hospital,
Manchester, UK
43. Bonten MJ, Kullberg BJ, Van Dalen R et al (2000) Selective digestive decontamination in
patients in intensive care. The Dutch Working Group on Antibiotic Policy. J Antimicrob
44. British National Formulary (BNF) No 59 (2010) British Medical Association and the Royal
Pharmaceutical Society of Great Britain, March 2010
Antimicrobial Resistance
28
N. Taylor, I. Cortes Puch, L. Silvestri, D. F. Zandstra
28.1 Introduction
Resistance to antimicrobial drugs is a serious risk factor for infection-related

morbidity and mortality in critically ill patients. Gut overgrowth by potentially
pathogenic microorganisms (PPMs) is a major risk for de novo development of
drug-resistant microorganisms. Selective decontamination of the digestive tract
(SDD), in which enteral administration of antimicrobial agents is the essential
element, significantly reduces the risk of gut overgrowth and the subsequent
development of drug-resistant pathogens.
28.2 Definition
SDD is a prophylactic antimicrobial strategy aiming to reduce infection-related

morbidity and mortality in the critically ill patient. It is a four-component protocol
consisting of enteral, nonabsorbable antimicrobial administration (polymyxin
E/tobramycin/amphotericin B (PTA)) and a short course of parenterally admin-
istered antibiotics (cefotaxime), together with a high standard of hygiene and
surveillance cultures [1].
N. Taylor (&)
452 N. Taylor et al.
28.3 Philosophy: Carriage
SDD is based on the observation that critical illness changes body flora by two
methods: qualitatively and quantitatively. Carriage of normal flora is eroded and
then replaced by abnormal flora [2], and the density of growth shifts from low to
high grade (gut overgrowth) [3]
1. Normal flora consists of six potential pathogens that are commonly found in
the oropharynx and/or gut of healthy people. Streptococcus pneumoniae,
Haemophilus influenzae, Moraxella catarrhalis, which are found in the oro-
pharynx only; Escherichia coli in the gut; and Staphylococcus aureus and
Candida albicans in both the oropharynx and gut.
2. Abnormal flora consists of nine abnormal potential pathogens. Eight aerobic
Gram-negative bacilli (AGNB) and methicillin-resistant S. aureus (MRSA),
found in both the oropharynx and gut. The eight AGNB are Klebsiella,
Enterobacter, Citrobacter, Proteus, Morganella, Serratia, Acinetobacter, and
Pseudomonas spp.
Critical illness changes the normal carrier state into the abnormal carrier state,
which is the persistent presence of abnormal flora in the patients oropharynx and/
or gut [2, 4].
28.3.1 Low- and High-Grade Carriage
Low-grade to high-grade carriage is the quantitative change brought about by

critical illness. Low-grade carriage is defined as \105 potential pathogens per
millilitre of saliva or gram of faeces (or \2+), whereas high-grade carriage is
defined as C105 potential pathogens per millilitre of saliva or gram of faeces (or
C2+) [5]. High-grade carriage is synonymous with gut overgrowth and is the
crucial event that precedes endogenous infection [6].
28.4 Infection Types
Infection, when classified according to the carrier state, can be categorised into
three different groups [7]: primary endogenous, secondary endogenous and
exogenous. All require a different prophylactic manoeuvre (Table 28.1).
1. Primary endogenous infection is the most frequent type, causing approximately
55% of infections in the intensive care unit (ICU). It occurs when the causative
microorganism is present in the patients admission flora in overgrowth concen-
trations, generally within 1 week of admission. The component of SDD that
prevents primary endogenous infections is the short course of parenterally
administered antibiotics given when the patient is admitted and for the next 4 days.
2. Secondary endogenous infection occurs when the causative microorganism is
not present in the patients admission flora but the abnormal flora is acquired
28 Antimicrobial Resistance 453
Table 28.1 Carriage classification of severe infections of lower airways and bloodstream
Infection PPM Timing Frequency Manoeuvre

(%)
1 6 normal \1 week 55 Parenteral antimicrobials
9 abnormal
2 9 abnormal [1 week 30 Hygiene and enteral
antimicrobials
3 9 abnormal Anytime during ICU 15 Hygiene and topical
1 primary endogenous, 2 secondary endogenous, 3 exogenous, PPM potentially pathogenic
microorganism, 6 normal PPM, Streptococcus pneumoniae, Haemophilus influenzae, Moraxella
catarrhalis, Candida albicans, Staphylococcus aureus, Escherichia coli, 9 abnormal PPM,
Klebsiella, Enterobacter, Citrobacter, Proteus, Morganella, Serratia, Acinetobacter, Pseudomonas
spp. and methicillin-resistant Staphylococcus aureus (MRSA), ICU intensive care unit
during treatment in the ICU, and the patient subsequently develops the carrier
state in overgrowth concentrations, which precedes the infection. Secondary
endogenous infection occurs after a week of ICU treatment at a frequency of
30%. The manoeuvre to impact this type of infection is enterally administered
nonabsorbable antimicrobials and high standards of hygiene.
3. Exogenous infection may occur at anytime during ICU treatment and develops
when the abnormal flora is introduced directly into the patient, bypassing the
carrier state. Approximately 15% of all ICU infections are exogenous. The
manoeuvre designed to combat this infection is hygiene and topical application
of antimicrobials. Enteral administration of antimicrobials also contributes to
exogenous infection control by reducing the reservoir of potential pathogens in
the unit that may be introduced into a patient
Surveillance cultures, the fourth component of SDD, are essential for three
reasons:
1. They are required to distinguish primary endogenous from secondary endog-
enous and exogenous infections.
2. They allow ongoing monitoring of eradication of gut overgrowth during SDD.
3. They are the most sensitive sampling methods for detecting antimicrobial
resistance.
28.5 Mechanisms of Action Explaining Efficacy: Control

of High-Grade Carriage or Overgrowth
The reason for SDD efficacy requires the understanding of two points
1. Acknowledgement of carriage classification [8]: In order to accept the concept
of SDD, it must be understood that carriage occurs. Primary carriage is when
the patient is admitted to the ICU with flora in the digestive tract; secondary
Table 28.2 Four components of selective decontamination of the digestive tract (SDD)
Target PPM and antimicrobials Total daily dose (4 times daily)

1. Parenteral antimicrobials: normal PPM 150/kg 200/kg 4,000
cefotaxime (mg)
2. Hygiene with enteral antimicrobials:
abnormal PPM
A. Oropharynx
AGNB: polymyxin E with tobramycin 2 g of 2% paste or gel
MRSA: Vancomycin 2 g of 4% paste or gel
Yeasts: amphotericin B or nystatin 2 g of 2% paste or gel
B. Gut
AGNB: polymyxin E (mg) with tobramycin (mg) 100 200 400
80 160 320
MRSA: vancomycin (mg) 2040/kg 2040/kg 5002,000
Yeasts: amphotericin B (mg) or nystatin units 500 1,000 2,000
2 9 106 4 9 106 8 9 106
3. Hygiene with topical antimicrobials: 2% PTA paste/4% vancomycin paste
abnormal PPM
4. Surveillance swabs of throat and rectum on
admission, Monday, Thursday
PPM potentially pathogenic microorganisms, AGNB aerobic Gram-negative bacilli, MRSA
methicillin-resistant Staphylococcus aureus, PTA polymyxin E/tobramycin/amphotericin B
carriage when the patient acquires flora in the digestive tract during treatment in
the ICU.
2. The antimicrobials selected for SDD were selected for specific reasons and
invariably clear high-grade carriage (Table 28.2):
parenteral cefotaxime clears carriage of normal bacteria; it is active against
normal and against some abnormal pathogens and, moreover, has a good
safety profile;
enterally administered polyenes clear normal fungal carriage;
enterally administered polymyxin/tobramycin with or without vancomycin
clears carriage of abnormal bacteria AGNB and MRSA; these antimicrobials
are nonabsorbable and manage to achieve high intraluminal concentrations.
SDD antimicrobials are required in high concentrations against prevailing
microorganisms (Table 28.3). These concentrations are deemed more important
than sparing the colonisation resistance [9]. Colonisation resistance is the mech-
anism whereby indigenous flora are a barrier against the abnormal flora acquired,
e.g. from food, and then carried in the digestive tract.
Gut overgrowth [10] harms the critically ill in four main ways:
1. Infection. There is a quantitative relationship between surveillance and diag-
nostic samples. As soon as there is overgrowth in surveillance samples, the
Table 28.3 Effective concentrations against prevailing microorganisms are more important than
sparing the colonisation-resistant flora
Antimicrobials selected for SDD Concentrations (mg/L) in
Saliva Bile Faeces

Cefotaxime 6 20
Polymyxin E 161,000
Tobramycin 100
Amphotericin B 60
or Nystatin \100
Vancomycin 3,00024,000
SDD selective decontamination of the digestive tract
diagnostic samples become positive, which is the first stage in infection

development [11].
2. Immunosuppression. Overgrowth of abnormal AGNB (and associated endo-
toxin) impairs systemic immunity due to generalised inflammation following
absorption of AGNB and/or endotoxin [12].
3. Inflammation. Overgrowth of abnormal AGNB and/or endotoxin leads to
cytokinaemia and inflammation of major organ systems [13].
4. Resistance. The abnormal carrier state in overgrowth concentrations guarantees
increased spontaneous mutation, leading to polyclonality and antibiotic resis-
tance [14].
SDD is a prophylactic measure using selected antimicrobials to control gut
overgrowth, thereby reducing the four harmful overgrowth side effects, i.e. infec-
tion [6], immunosuppression [15], inflammation [16] and resistance [17].
28.6 Efficacy
SDD is one of the most investigated clinical interventions in critically ill patients.
Its efficacy in preventing infection-related morbidity and mortality has been
assessed in 11 meta-analyses [1828] covering 65 randomised controlled trials
(RCTs) [2993] (Table 28.4). Of the 11 meta-analyses, lower airway infection was
the endpoint in six [1820, 24, 26, 28]. All meta-analyses invariably demonstrated
a significant reduction in lower airway infections (odds ratio (OR) 0.28, 95%
confidence interval (CI) 0.200.38). Bloodstream infection was the endpoint in
three meta-analyses [2224] and was significantly reduced (OR 0.63, 95% CI
0.460.87). When assessing bloodstream infection, AGNB septicaemias were
significantly reduced, Gram-positive septicaemias were increased but not sig-
nificantly and fungaemia was reduced but not significantly due to the low incidence
in the control group (Table 28.4). Multiple organ dysfunction syndrome (MODS)
456
Table 28.4 Efficacy of selective decontamination of the digestive tract (SDD): 65 randomised controlled trials (RCTs) and 11 meta-analyses
Author No. RCTs Sample size Lower airway Bloodstream Multiple organ Mortality rate
infection infection dysfunction syndrome OR (95% CI)
OR (95% CI) OR (95% CI) OR (95% CI)
Vandenbroucke-Grauls [18] 6 491 0.12, 0.080.19 NR 0.92, 0.451.84
DAmico [19] 33 5727 0.35, 0.290.41 NR 0.80, 0.690.93
Safdar [20] 4 259 NR NR 0.82, 0.222.45
Liberati [21] 36 6922 0.35, 0.290.41 NR 0.78, 0.680.89
Silvestri [22] yeasts 42 6075 NR 0.89, 0.164.95 NR
Silvestri [23] 51 8065 NR 0.63, 0.460.87 0.74, 0.610.91
Silvestri [24] 54 9473
Gram 0.07, 0.040.13 0.36, 0.220.60 NR
Gram+ 0.52, 0.340.78 1.03, 0.751.41 NR
Silvestri [25] 21 4902 NR NR 0.71, 0.610.82
Liberati [26] 36 6914 0.28, 0.200.38 NR 0.75, 0.650.87
Silvestri [27] 7 1270 NR NR 0.50, 0.340.74 0.82, 0.511.32
Silvestri [28] 12 2252 0.54, 0.420.69 NR NR
OR odds ratio, CI confidence interval, NR not reported
N. Taylor et al.
was the endpoint in one meta-analysis [27]; the relative reduction of 50% was
significant. Mortality was the endpoint in nine meta-analyses [1821, 23, 2528].
SDD consistently reduced mortality rates as long as the sample size was large
enough; the sample size was too small in three meta-analyses [18, 20, 27].
28.7 Safety
SDD safety relies on the long-term level of resistance not emerging against the
SDD antimicrobials [94]. The concept of exposing vast numbers of critically ill
patients to broad-spectrum multiple drug cocktails runs counter to existing theo-
retical models (and dogma), as it is related to the genesis and promotion of
antimicrobial resistance in pathogens acquired in the ICU [95]. The dynamics of
resistance are driven by three mechanisms:
1. Import. The patient comes into the ICU with resistant microorganisms in
overgrowth concentrations in the gut [3].
2. Transmission. 33% of patients admitted as normal carriers to a mixed ICU
developed abnormal carriage of multi-drug-resistant K. pneumoniae and/or
A. baumannii, the two abnormal AGNB endemic in the ICU at the time of the
1-year prospective observational study [96]. A higher Severity of Illness score
on admission was a significant risk factor: the Simplified Acute Physiologically
Score (SAPS) was 13 4.6 in carriers versus 11.3 5 in noncarriers
(p = 0.0006).
3. De novo development. Gut overgrowth defined as C105 potential pathogens per
gram of faeces has been identified as a risk factor for de novo resistance
development [14, 97]. The gut of the critically ill patient with microbial
overgrowth is the ideal site for de novo development of new clones following
increased spontaneous mutation, termed hypermutation. In hypermutation,
microbial populations start mutating vigorously at random, presumably as an
adaptive mechanism that may cause a mutant to arise that would enable them to
overcome the unfavourable surroundings, resulting in polyclonality. A high
proportion of patients who require long-term ICU treatment receive parenter-
ally administered antimicrobials, which are invariably excreted via the bile into
the gut. Although low and fluctuating, the antibiotic levels will kill the sensitive
clones, promoting the emergence of clones resistant to the antibiotics.
Each mechanism is responsible for a third of antimicrobial resistance in the
ICU. The common denominator of all three mechanisms is gut overgrowth.
There are four categories in which antimicrobial resistance is a problem in the
ICU:
1. AGNB
(a) Sensitive to decontaminating agents polymyxin/tobramycin. de Jonge et al.
conducted a prospective open-label RCT in which 934 critically ill adult
patients were randomly assigned on admission to either a medical/surgical
ICU using routine SDD or a similar ICU in the same hospital that did not
use SDD [43]. Study participants were patients with expected duration of
mechanical ventilation of at least 48 h and/or ICU stay [3 days. Surveil-
lance cultures from the throat and rectum were obtained at ICU admission
and at discharge, weekly during ICU treatment and for the first week
postdischarge. The in-hospital mortality rate was significantly lower for
SDD recipients than for control patients (24 vs. 31%; p = 0.02). Carriage
of AGNB resistant to polymyxin E, tobramycin, ceftazidime, ciprofloxacin
and imipenem was significantly reduced in SDD patients compared with
controls (16 vs. 26%; p = 0.001). Similar results were observed by de Smet
et al. in their cluster RCT [45]. Monthly point-prevalence surveys for
carriage of multi-drug-resistant AGNB were obtained. The proportion of
rectal swabs with resistant AGNB was lower for SDD compared with
standard-care patients.
(b) Resistant to decontaminating agents polymyxin/tobramycin:
there is only one potential pathogen intrinsically resistant to polymyxin/
tobramycinSerratia species. In case of Serratia endemicity, polymyxin/
tobramycin must be replaced by paromomycin [98, 99].
extended-spectrum beta-lactamase (ESBL) producing AGNB are often resistant
to tobramycin but always sensitive to polymyxin [100]. In case of ESBL pro-
ducing AGNB endemicity, tobramycin needs to be replaced by another amino-
glycoside, e.g. neomycin, paromomycin [39, 93, 99].
2. MRSA
(a) Sensitive to glycopeptides. Practically all MRSA strains are sensitive to
glycopeptides such as vancomycin and teicoplanin. All RCTs [34, 52, 59,
60, 69, 80] and prospective studies [6, 101, 102] assessing the efficacy of
enterally administered vancomycin on the abnormal MRSA carrier state
demonstrated its efficacy. Therefore, it is recommended to add vancomycin
to PTA in cases of MRSA endemicity.
(b) Resistant to glycopeptides. MRSA is now endemic in many hospitals
throughout the world. Several clusters of glycopeptide-resistant MRSA
have been reported. A 39-case outbreak, defined as involving any patient
carrying or infected by a strain of MRSA with reduced or intermediate
susceptibility to glycopeptides, was recorded in a hospital in Paris, France,
from October 1998 to March 1999 [103]. Another outbreak from Lyon,
France, included 15 patients [104]. MRSA has been acknowledged as a gut
rather than a nasal bacterium [6, 101, 102]. Patients requiring long-term
ICU treatment invariably develop gut overgrowth of MRSA when it is
endemic. MRSA gut overgrowth precedes endogenous MRSA infection,
whichin generalis treated with glycopeptides. The combination of low
antimicrobial concentrations in the faeces following biliary excretion and
overgrowth leads to increased spontaneous mutation, polyclonality and
antimicrobial resistance against glycopeptides [14].
3. Azole-resistant Candida spp.
Practically all critically ill patients who require long-term ICU treatment have
fungal overgrowth [105]. Parenteral azoles administered either prophylactically
or therapeutically fail to clear fungal overgrowth, because the faecal levels are
not fungicidal and/or fluctuating following biliary excretion [106]. In contrast,
fungal overgrowth promotes increased spontaneous mutation, polyclonality and
antifungal resistance [107, 108].
4. Vancomycin-resistant enterococci (VRE)
Enterococci are normal residents of the large bowel but not the oropharynx.
Enterococci are organisms of low-level pathogenicity present in the gut of
healthy people at concentrations between 1036 enterococci/gram of faeces.
Enterococcus faecalis and E. faecium are the two most common enterococci
associated with infection: E. faecalis is carried by 80% of healthy individuals at
concentrations between 1057 CFU/g of faeces; E. faecium is carried by 30% of
healthy individuals at lower concentrations. Enterococci are resistant to many
antimicrobials but in general sensitive to glycopeptides (vancomycin, teicopl-
anin) and linezolid. A VRE is defined as an Enterococcus with a minimum
inhibitory concentration (MIC) of C16 mg of vancomycin per litre [109].
(a) Linezolid sensitive VRE. Carriage of VRE even sensitive to linezolid is
abnormal [110]. As far as we are aware, there are no trials attempting to
eradicate VRE carriage. Theoretically, it is not impossible to clear abnor-
mal VRE carriage using vancomycin enterally: 2 g is associated with faecal
vancomycin levels varying between 3,000 and 24,000 mg per litre, with an
MIC around 16 mg/L [111].
28.8 No Resistance During SDD (Meta-Analysis)
Five RCTs of SDD involving 5,229 patients (2,631 SDD, 2,598 controls) reported
data on resistance [39, 43, 45, 50, 61]. There were 74 (2.8%) patients with resistant
microorganisms in the SDD group and 124 (4.8%) in controls, indicating a 44%
reduction in the odds of resistance by SDD (OR 0.56, 95% CI 0.410.76;
p \ 0.001) (Fig. 28.1). Heterogeneity was not shown (v2 = 2.58888, p = 0.63;
I2 = 0).
28.8.1 Subgroup Analysis
We investigated the impact of SDD on resistance among subgroups with different

types of regimens (Fig. 28.1). Three studies using the full SDD protocol of par-
enterally and enterally administered antimicrobials [43, 45, 50] involved 5,076
patients (2,612 SDD, 2,566 control). Resistance was demonstrated in 76 (2.9%)
patients in the SDD group and 118 (4.6%) in the control group, resulting in a
significant 44% reduction in the odds of resistance. In contrast, in studies using
only enterally administered antimicrobials [39, 61] and involving 153 patients
(69 SDD, 84 controls), the number of patients with resistant microorganisms was
two (2.9%) in the SDD and six (7.14%) in the control group. This reduction was
not significant.
Fig. 28.1 Meta-analysis of five randomised controlled trials (RCTs) of selective decontamina-
tion of the digestive tract (SDD) with relevant data on resistance Odds ratio\1 favours SDD; [1
favours controls. Results presented with the fixed-effect model, as heterogeneity was not
demonstrated T+, test; T, control Brun-Buisson [39], de Smet [45], Flaherty [50], Laggner [61],
de Jonge [43]
However, the sensitivity analysis showed that after excluding the de Jonge trial
[43], the reduction in resistance was not significant (OR 0.75, 95% CI 0.351.61;
p = 0.46; heterogeneity not significant).
28.9 SDD: Only EBM Manoeuvre with Grade 1A

Recommendation
In recent years, RCTs have shown that five different manoeuvres reduce mortality
rates:
ventilation with low tidal volumes for acute lung injury and respiratory distress
syndrome [112];
recombinant human activated protein C for severe sepsis [113];
intensive insulin therapy [114];
low doses of steroids in patients with septic shock [115];
SDD [43, 45, 60].
Table 28.5 reports the levels of evidence obtained using the Grading of
Recommendations Assessment, Development and Evaluation (GRADE) system
[116], which classifies the quality of evidence as high (A), moderate (B), low (C)
or very low (D). RCTs may be downgraded due to limitations in implementation,
inconsistency or imprecision of results, indirectness of the evidence and possible
reporting bias [116]. An example of this is tight glucose control (A down to C): the
Table 28.5 Intensive care unit (ICU) interventions that reduce mortality rates
Intervention Relative risk AMR (%) Number needed Grade of

(95% CI) (95% CI) to treat recommendation
1. Low tidal volume 0.78 (0.650.93) 8.8 (2.415.3) 11 1B
[112] 0.80 (0.690.94) 6.1 (1.910.4) 16 2B
2. Activated protein 0.44 (0.360.81) 3.7 (1.36.1) 27 2C
C [113] 0.90 (0.741.09) 6.4 (-4.817.6) 16 2C
3. Intensive insulin 0.65 (0.490.85) 8.1 (3.113.0) 12 1A
[114]
4. Steroids [115]
5. SDD [43, 45, 60]
CI confidence interval, AMR absolute mortality reduction, 1 strong, 2 weak, A high, B moderate,
C low
success of the original Belgian RCT [114] in reducing mortality rates has not to
date been reproduced [117121]. The GRADE system classifies recommendations
as strong (1) or weak (2). The grade of strong or weak is considered of greater
clinical importance than a difference in letter level of quality of evidence. A strong
recommendation in favour of an intervention reflects that the desirable effects of
adherence to a recommendation (beneficial health outcomes, less burden on staff
and patients and cost savings) will clearly outweigh the undesirable effects (harms,
more burdens, greater costs).
All RCTs and meta-analyses of SDD that assessed the full four-component
protocol consistently demonstrated a significant survival benefit, providing the
sample size was large enough. Mortality data show an intriguing observation that
trial design determines the magnitude of the survival benefit [122]. The relative
reduction in the OR for mortality was 41% when all patients received the full
SDD protocol [43], 29% when half the patients received SDD [122] and 17%
when one third of the population was treated with SDD [45, 123]. In the trial of
the unit-wide application of SDD [43], SDD virtually eliminated transmission of
potential pathogens via the hands of carers and hence exogenous infection in
decontaminated patients. Survival benefit is diluted by mixing decontaminated
and nondecontaminated patients in the same unit. This is the case in the RCT
design, wherein patients receiving and not receiving SDD are treated within the
same unit [45, 122]. Decontaminated patients protect control patients from
transmission, acquisition, carriage, and subsequent infection, whereas patients
receiving SDD remain at risk of acquiring potential pathogens and subsequent
exogenous infections, resulting in a reduction in the true effect of SDD. The
most recent multicentre RCT with a 17% relative reductionalbeit statistically
significantclearly emphasises the issue of diluting the SDD effect by
increasing the number of nondecontaminated patients treated in the same unit
with patients receiving SDD [45, 122].
28.10 Costs Implication for Using SDD
Although the cost-effectiveness of SDD has not yet been formally calculated, the
daily costs of 612 euros [45, 70] can hardly be an issue for an ICU intervention
that reduces pneumonia, septicaemia and mortality rates by 72, 37 and 29%,
respectively (Table 28.4).
References
1. WFSICCM congress Firenze 2009 Gullo A (2009) Intensive and Critical Care Medicine.
Springer, Milan, pp 273283
hospitalized patients. Emergence of Gram-negative bacilli. N Engl J Med 281:11371140
3. Viviani M, van Saene HK, Pisa F et al (2010) The role of admission surveillance cultures in
patients requiring prolonged mechanical ventilation in the intensive care unit. Anaesth
4. Mobbs KJ, van Saene HK, Sunderland D, Davies PD (1999) Oropharyngeal Gram-negative
bacillary carriage: a survey of 120 healthy individuals. Chest 115:15701575
5. van Saene HKF, Damjanovic V, Murray AE, de la Cal MA (1996) How to classify infections in
intensive care unitsthe carrier state, a criterion whose time has come? J Hosp Infect 33:112
7. Stoutenbeek CP, van Saene HKF, Liberati A (1994) Prevention of respiratory tract infections
in intensive care by selective decontamination of the digestive tract. In: Niederman MS,
Sarosi GA, Glassroth J (eds) Respiratory infections, 1st edn. A scientific basis for
management, Philidelphia, pp 579594
8. Silvestri L, van Saene HK, de la Cal MA, De Gaudio AR (2009) Carriage classification of
pneumonia rather than time improves survival. Chest 136:11881189
9. Vollaard EJ (1991) The concept of colonization resistance. PhD thesis, Benda BV,
Nijmegen, The Netherlands
237:419427
11. Van Uffelen R, van Saene HK, Fidler V, Lwenberg A (1984) Oropharyngeal flora as a
source of bacteria colonizing the lower airways in patients on artificial ventilation. Intensive
Care Med 10:233237
13. Baue AE (1993) The role of the gut in the development of multiple organ dysfunction in
cardiothoracic patients. Ann Thorac Surg 55:822829
14. van Saene HKF, Taylor N, Damjanovic V, Sarginson RE (2008) Microbial gut overgrowth
guarantees increased spontaneous mutation leading to polyclonality and antibiotic resistance
in the critically ill. Curr Drug Targets 9:419421
15. Horton JW, Maass DL, White J, Minei JP (2007) Reducing susceptibility to bacteremia after
experimental burn injury: a role for selective decontamination of the digestive tract. J Appl
Physiol 102:22072216
16. Conraads VM, Jorens PG, De Clerck LS (2004) Selective intestinal decontamination in
17. van Saene HKF, Stoutenbeek CP, Zandstra DF (1988) Cefotaxime combined with selective
decontamination in long term intensive care patients. Virtual absence of emergence of
rsistance. Drugs 35(Suppl 2):2934

Lancet 338:859862
19. DAmico R, Pifferi S, Leonetti C et al on behalf of the Study Investigators (1998)
Effectiveness of antibiotic prophylaxis in critically ill adult patients: systematic review of
randomised controlled trials. BMJ 316:12751285
infection in patients undergoing liver transplantation: a systematic review and meta-
tract infections and mortality in adults receiving intensive care (Cochrane Review). In: The
Cochrane Library Issue 1. Wiley, Chichester
of randomized controlled trials. Intensive Care Med 31:898910
23. Silvestri L, van Saene HKF, Milanese M (2007) Selective decontamination of the digestive
tract reduces bloodstream infections and mortality in critically ill patients: a systematic
review of randomized controlled trials. J Hosp Infect 65:187203
24. Silvestri L, van Saene HKF, Casarin AL et al (2008) Impact of selective decontamination of the
digestive tract on carriage and infection due to Gram-negative and Gram-positive bacteria.
Systematic review of randomized controlled trials. Anaesth Intensive Care 36:324338
digestive decontamination regimen. J Crit Care 24(474):e7e14
26. Liberati A, DAmico R, Pifferi S et al. (2009) Antibiotic prophylaxis to reduce respiratory
tract infections and mortality in adults receiving intensive care. Cochrane Database Syst
Rev CD000022
28. Silvestri L, van Saene HKF, Zandstra DF (2010) Selective digestive decontamination
reduces ventilator-associated tracheobronchitis. Respir Med 104(12):19531955
29. Abdel-Razek SM, Abdel-Khalek AH et al (2000) Impact of selective gastrointestinal
decontamination on mortality and morbidity in severely burned patients. Ann Burns Fire
Disasters 13:213216
30. Abele-Horn M, Dauber A, Bauernfeind A et al (1997) Decrease in nosocomial pneumonia
in ventilated patients by selective oropharyngeal decontamination (SOD). Intensive Care
Med 23:187195
31. Aerdts SJ, van Dalen R, Clasener HA et al (1991) Antibiotic prophylaxis of respiratory tract
infection in mechanically ventilated patients. A prospective, blinded, randomized trial of the
effect of a novel regimen. Chest 100:783791
32. Arnow PM, Carandang GC, Zabner R, Irwin ME (1996) Randomized controlled trial of
selective bowel decontamination for prevention of infections following liver transplantation.
in severely burned pediatric patients. Burns 27:439445
pneumonia by oral decontamination: a prospective, randomized, double-blind, placebo-
reduces Gram-negative pulmonary colonization but not systemic endotoxemia in patients
36. Blair P, Rowlands BJ, Lowry K et al (1991) Selective decontamination of the digestive tract: a
stratified, randomized, prospective study in a mixed intensive care unit. Surgery 110:303310
37. Boland JP, Sadler DL, Stewart WA et al. (1991)Reduction of nosocomial respiratory tract
infections in multiple trauma patient requiring mechanical ventilation by selective
parenteral and enteral antisepsis regimen (SPEAR) in the intensive care unit. 17th
International congress of chemotherapy, Berlin, Abstract 0465
randomized, placebo-controlled study. Crit Care Med 30:3843
nosocomial multiresistant Gram-negative bacilli. Study of an outbreak in an intensive care
unit. Ann Intern Med 110:873881
40. Camus C, Bellissant E, Sebille V et al (2005) Prevention of acquired infections in
intubated patients with the combination of two decontamination regimens. Crit Care Med
33:307314
41. Cerra FB, Maddaus MA, Dunn DL et al (1992) Selective gut decontamination reduces
nosocomial infections and length of stay but not mortality or organ failure in surgical
intensive care unit patients. Arch Surg 127:163169
42. Cockerill FR 3rd, Muller SR, Anhalt JP et al (1992) Prevention of infection in critically ill
patients by selective decontamination of the digestive tract. Ann Intern Med 117:545553
43. de Jonge E, Schultz MJ, Spanjaard L et al (2003) Effects of selective decontamination of
44. de la Cal MA, Cerd E, Garca-Hierro P et al (2005) Survival benefit in critically ill burned
patients receiving selective decontamination of the digestive tract: a randomized, placebo-
controlled, double-blind trial. Ann Surg 241:424430
46. Diepenhorst GM, van Ruter O, Besselink MG, van Santvoort HC, Wijnandts PR, Renooij
W, Gouma DJ, Goosen HG, Boermeester MA (2011) Influence of prophylactic probiotics
and selective decontamination on bacterial translocation in patients undergoing pancreatic
surgery: a randomized controlled trial. Shock 35: 916
47. Farran L, Llop J, Sans M et al (2008) Efficacy of enteral decontamination in the prevention
of anastomotic dehiscence and pulmonary infection in esophagogastric surgery. Dis
Esophagus 21:159164
48. Ferrer M, Torres A, Gonzlez J et al (1994) Utility of selective digestive decontamination in
mechanically ventilated patients. Ann Intern Med 120:389395
49. Finch RG, Tomlinson P, Holiday M, Sole K, Stack C, Rocker G. (1991) Selective
decontamination of the digestive tract (SDD) in the prevention of secondary sepsis in a
medical/surgical intensive care unit. 17th International congress of chemotherapy, Berlin,
Abstract 0474
50. Flaherty J, Nathan C, Kabins SA, Weinstein RA (1990) Pilot trial of selective
decontamination for prevention of bacterial infection in an intensive care unit. J Infect
Dis 162:13931397
51. Gastinne H, Wolff M, Delatour F, Faurisson F, Chevret S (1992) A controlled trial in
intensive care units of selective decontamination of the digestive tract with nonabsorbable
antibiotics. The French Study Group on Selective Decontamination of the Digestive Tract.
N Eng J Med 326:594599
52. Gaussorgues PH, Salord F, Sirodot M, Tigaud S, Cagin S, Gerard M, Robert D (1991)
Efficacite de la decontamination digestive sur la survenue des bacteriemies nosocomiales
chez les patients sous ventilation mecanique et recevant des betamimetiques. Rean Soins
Intens Med Urg 7:169174
53. Georges B, Mazerolles M, Decun JF, Rouge P, Pomies S, Cougot P, Andrieu P, Virenque
CH (1994) Dcontamination Digestive Slective Rsultats Dune tude Chez Le
Polytraumatis. Rean Urg 3:621627
54. Gosney M, Martin MV, Wright AE (2006) The role of selective decontamination of the
digestive tract in acute stroke. Age Ageing 35:4247
55. Hammond JM, Potgieter PD, Saunders GL, Forder AA (1992) Double-blind study of
selective decontamination of the digestive tract in intensive care. Lancet 340:59
56. Hellinger WC, Yao JD, Alvarez S, Blair JE, Cawley JJ, Paya CV et al (2007) A randomized,
prospective, double-blinded evaluation of selective bowel decontamination in liver
transplantation. Transplantation 73:19041909
57. Jacobs S, Foweraker JE, Roberts SE (1992) Effectiveness of selective decontamination of
the digestive tract (SDD) in an ICU with a policy encouraging a low gastric pH. Clin
58. Kerver AJ, Rommes JH, Mevissen-Verhage EA, Hulstaert PF, Vos A, Verhoef J et al (1988)
Prevention of colonization and infection in critically ill patients: a prospective randomized
study. Crit Care Med 16:10871093
59. Korinek AM, Laisne MJ, Nicolas MH, Raskine L, Deroin V, Sanson-Lepors MJ et al (1993)
Selective decontamination of the digestive tract in neurosurgical intensive care unit patients:
a double-blind, randomized, placebo-controlled study. Crit Care Med 21:14661473
60. Krueger WA, Lenhart FP, Neeser G, Ruckdeschel G, Schreckhase H, Eissner HJ et al (2002)
Influence of combined intravenous and topical antibiotic prophylaxis on the incidence of
infections, organ dysfunctions, and mortality in critically ill surgical patients: a prospective,
stratified, randomized, double-blind, placebo-controlled clinical trial. Am J Respir Crit Care
Med 166:10291037
61. Laggner AN, Tryba M, Georgopoulos A et al (1994) Oropharyngeal decontamination with
gentamicin for long-term ventilated patients on stress ulcer prophylaxis with sucralfate?
Wien Klin Wochenschr 106:1519
62. Lingnau W, Berger J, Javorsky F, Lejeune P, Mutz N, Benzer H et al (1997) Selective
intestinal decontamination in multiple trauma patients: prospective, controlled trial.
J Trauma 42:687694
63. Luiten EJ, Hop WC, Lange JF, Bruining HA (1995) Controlled clinical trial of selective
64. Martinez-Pells AE, Merino P, Bru M, Conejero R, Seller G, Muoz C et al (1992) Can
selective digestive decontamination avoid the endotoxemia and cytokine activation
promoted by cardiopulmonary bypass? Crit Care Med 21:16841691
65. Martinez-Pells AE, Merino P, Bru M, Canovas J, Seller G, Sapia J et al (1997)
Endogenous endotoxemia of intestinal origin during cardiopulmonary bypass. Role of type
of flow and protective effect of selective digestive decontamination. Intensive Care Med
23:12511257
66. Oudhuis GJ, Bergmans DC, Dormans T, Zwaveling JH, Kessels A, Prins MH, Stobberingh
EE, Verbon A (2011) Probiotics versus antibiotic decontamination of the digestive tract:
infection and mortality. Intensive Care Med 37:110-117
67. Palomar M, Alvarez-Lerma F, Jorda R, Bermejo B (1997) Catalan study group of
nosocomial pneumonia prevention. Prevention of nosocomial infection in mechanically
ventilated patients: selective digestive decontamination versus sucralfate. Clin Intensive
Care 8:228235
68. Pneumatikos I, Koulouras V, Nathanail C, Geo D, Nakos G (2002) Selective
decontamination of subglottic area in mechanically ventilated patients with multiple
trauma. Intensive Care Med 28:432437
decreases incidence of ventilator-associated pneumonia. A randomized, placebo-controlled,
double-blind clinical trial. JAMA 265:27042710
70. Quinio B, Albanse J, Bues-Charbit M, Viviand X, Martin C (1996) Selective
decontamination of the digestive tract in multiple trauma patients. A prospective double-
blind, randomized, placebo-controlled study. Chest 109:765772
72. Rios F, Maskin B, Saenex Valiente A, Galante A, Cazes Camarero P, Aguilar L et al (2005)
Prevention of ventilator associated pneumonia (VAP) by oral decontamination (OD).
Prospective, randomized, double-blind, placebo-controlled study. American Thoracic
Society International Conference, San Diego, USA, C95; poster 608
73. Rocha LA, Martin MJ, Pita S, Paz J, Seco C, Margusina L et al (1992) Prevention of
nosocomial infection in critically ill patients by selective decontamination of the digestive
tract. A randomized, double blind, placebo-controlled study. Intensive Care Med
18:398404
74. Rodrguez-Roldn JM, Altuna-Cuesta A, Lpez A, Carrillo A, Garcia J, Len J et al (1990)
Prevention of nosocomial lung infection in ventilated patients: use of an antimicrobial
pharyngeal nonabsorbable paste. Crit Care Med 18:12391242
75. Rolando N, Gimson A, Wade J, Philpott-Howard J, Casewell M, Williams R (1993)
Prospective controlled trial of selective parenteral and enteral antimicrobial regimen in
fulminant liver failure. Hepatology 17:196201
76. Rolando N, Wade JJ, Stangou A, Gimson AE, Wendon J, Philpott-Howard J et al (1996)
Prospective study comparing the efficacy of prophylactic parenteral antimicrobials, with or
without enteral decontamination, in patients with acute liver failure. Liver Transpl Surg
2:813
77. Roos D, Dijksman LM, Oudemans-van Straaten HM, de Wit LT, Gouma DJ, Gerhards MF
(2011) Randomized clinical trial of perioperative selective decontamination of the digestive
tract versus placebo in elective gastrointestinal surgery. Br J Surg [Epub ahead of print]
78. Ruza F, Alvarado F, Herruzo R, Delagado MA, Garcia S, Dorao P et al (1998) Prevention of
nosocomial infection in a pediatric intensive care unit (PICU) through the use of selective
digestive decontamination. Eur J Epidemiol 14:719727
79. Snchez Garca M, Cambronero Galache JA, Lpez Diaz J, Cerd Cerd E, Rubio Blasco J,
Gmez Aguinaga MA et al (1998) Effectiveness and cost of selective decontamination of
the digestive tract in critically ill intubated patients. A randomized, double-blind, placebo-
controlled, multicenter trial. Am J Respir Crit Care Med 158:908916
80. Schardey HM, Joosten U, Finke U Staubach KH, Schauer R, Heiss A et al (1997) The
prevention of anastomotic leakage after total gastrectomy with local decontamination. A
prospective, randomized, double-blind, placebo-controlled multicenter trial. Ann Surg
225:172180
81. Smith SD, Jackson RJ, Hannakan CJ, Wadowsky RM, Tzakis AG, Rowe MI (1993)
Selective decontamination in pediatric liver transplants. A randomized prospective study.
82. Stoutenbeek CP, van Saene HKF, Zandstra DF (1996) Prevention of multiple organ
system failure by selective decontamination of the digestive tract in multiple trauma
patients. Faist E, Baue AE, Schildberg FW (Eds.) In: The immune Consequence of
Trauma, Shock and SepsisMechanisms and Therapeutic Approaches. Pabst Science,
Lengerich, pp 10551066
83. Stoutenbeek CP, van Saene HK, Little RA, Whitehead A for the Working Group on
Selective Decontamination of the Digestive Tract (2007) The effect of selective
decontamination of the digestive tract on mortality in multiple trauma patients: a
multicenter randomized controlled trial. Intensive Care Med 33:261270
84. Tetteroo GW, Wagenvoort JH, Castelein A, Tilanus HW, Ince C, Bruining HA (1990)
Selective decontamination to reduce Gram-negative colonisation and infections after
oesophageal resection. Lancet 335:704707
85. Ulrich C, Harinck-de Weerd JE, Bakker NC, Jacz K, Doornbos L, de Ridder VA (1989)
Selective decontamination of the digestive tract with norfloxacin in the prevention of ICU-
acquired infections: a prospective randomized study. Intensive Care Med 15:424431
86. Unertl K, Ruckdeschel G, Selbmann HK, Jensen U, Forst H, Lenhart FP et al (1987)

Prevention of colonization and respiratory infections in long-term ventilated patients by
local antimicrobial prophylaxis. Intensive Care Med 13:106113
87. Verwaest C, Verhaegen J, Ferdinande P, Schetz M, Van den Berghe G, Verbist L et al
(1997) Randomized, controlled trial of selective digestive decontamination in 600
mechanically ventilated patients in a multidisciplinary intensive care unit. Crit Care Med
25:6371
88. Wiener J, Itokazu G, Nathan C, Kabins SA, Weinstein RA (1995) A randomized, double-
blind, placebo-controlled trial of selective digestive decontamination in a medical-surgical
intensive care unit. Clin Infect Dis 20:861867
89. Winter R, Humphreys H, Pick A, MacGowan AP, Willatts SM, Speller DCE (1992) A
controlled trial of selective decontamination of the digestive tract in intensive care and its
effect on nosocomial infection. J Antimicrob Chemother 30:7387
90. Yilamazlar A, Ozyurt G, Kahveci F, Goral G (2009) Selective digestive decontamination
can be an infection-prevention regimen for the intoxicated patients. J Pharmaco Toxicol
4:36-40
91. Yu J, Xiao YB, Wang XY (2007) Effects of preoperatively selected gut decontamination on
cardiopulmonary bypass-induced endotoxemia. Chin Traumatol 10:131-137
92. Zobel G, Kuttnig M, Grubbauer HM, Semmelrock HJ, Thiel W (1991) Reduction of
colonization and infection rate during pediatric intensive care by selective decontamination
of the digestive tract. Crit Care Med 19:12421246
93. Zwaveling JH, Maring JK, Klompmaker IJ, Haagsma EB, Bottema JT, Laseur M et al
(2002) Selective decontamination of the digestive tract to prevent postoperative infection: a
randomized placebo-controlled trial in liver transplant patients. Crit Care Med
30:12041209
94. Baxby D, van Saene HKF, Stoutenbeek CP et al (1996) Selective decontamination of the
digestive tract: 13 years on, what it is and what it is not. Intensive Care Med 22:699706
95. Laupland KB, Fisman DN (2009) Selective digestive tract decontamination: A tough pill to
swallow. Can J Infect Dis Med Microbiol 20:911
96. Garrouste-Orgeas M, Marie O, Rouveau M et al (1996) Secondary carriage with multi-
resistant Acinetobacter baumannii and Klebsiella pneumoniae in an adult ICU population:
relationship with nosocomial infections and mortality. J Hosp Infect 34:279289
97. Verma N, Clarke RW, Bolton-Maggs PHB et al (2007) Gut overgrowth of vancomycin-
resistant enterococci (VRE) results in linezolid-resistant mutation in a child with severe
congenital neutropenia. J Pediatr Hematol Oncol 29:557560
98. Abecasis F, Kerr S, Sarginson RE et al (2007) Comment on: emergence of multidrug-
resistant Gram-negative bacteria during selective decontaminiation of the digestive tract on
an intensive care unit. J Antimicrob Chemother 60:445
99. Bodey GP (1981) Antibiotic prophylaxis in cancer patients: regimens of oral non-
absorbable antibiotics for prevention of infection during induction of admission. Rev Infect
Dis 3(Suppl):S259S268
100. Brun-Buisson C, Legrand P, Philippon A et al (1987) Transferable enzymatic resistance to
third-generation cephalosporins during nosocomial outbreak of multi-resistant Klebsiella
pneumoniae. Lancet ii:302306
101. Viviani M, van Saene HKF, Dezzoni R et al (2005) Control of imported and acquired
methicillin-resistant Staphylococcus aureus (MRSA) in mechanically ventilated patients: a
dose-response study of enteral vancomycin to reduce absolute carriage and infection.
Anaesth Intensive Care 33:361372
102. Cerda E, Abella A, de la Cal MA et al (2007) Enteral vancomycin controls methicillin-
resistant Staphylococcus aureus endemicity in an intensive care burn unit: a 9-year
prospective study. Ann Surg 245:397407
103. Guerin F, Buu-Hoi A, Mainardi JL (2000) Outbreak of methicillin-resistant Staphylococcus

aureus with reduced susceptibility to glycopeptides in a Parisian hospital. J Clin Microbiol
38:29852988
104. Pina P, Marliene C, Vandenesch F et al (2000) An outbreak of Staphylococcus aureus
strains with reduced susceptibility to glycopeptides in a French general hospital. Clin Infect
Dis 31:13061308
105. van Saene HKF, Damjanovic V, Pizer B et al (1999) Fungal infections in ICU. J Hosp Infect
41:337340
106. Bozzette SA, Gordon RL, Yen A et al (1992) Biliary concentrations of fluconazole in a
patient with candidal cholecytitis: case report. Clin Infect Dis 15:701703
107. Martins MD, Rex JH (1996) Resistance to antifungal agents in the critical care setting:
problems and perspectives. New Horiz 4:338344
108. Rex JH, Rinaldi MG, Pfaller MA (1995) Resistance of Candida species to fluconazole.
Antimicrob Agents Chemother 39:18
109. Malathum K, Murray BE (1999) Vancomycin-resistant enterococci: recent advances in
genetics, epidemiology and therapeutic options. Drug Resiste Updat 2:224243
110. Coque TM, Tomayko JF, Ricke SC et al (1996) Vancomycin-resistant enterococci from
nosocomial, community, and animal sources in the United States. Antimicrob Ag Chemother
40:26052609
111. Currie BP, Lemos-Filho L (2004) Evidence for biliary excretion of vancomycin into stool
during intravenous therapy: potential implications for rectal colonization with vancomycin-
resistant enterococci. Antimicrob Ag Chemother 48:44274429
112. The Acute Respiratory Distress Syndrome Network (2000) Ventilation with lower tidal
volumes as compared with traditional tidal volumes for acute lung injury and the acute
respiratory distress syndrome. N Engl J Med 342:13011308
activated protein C for severe sepsis. N Engl J Med 344:699709
114. Van den Berghe G, Wouters P, Weekers F et al (2001) Intensive insulin therapy in the
115. Annane D, Bellissant E, Bollaert P et al (2009) Corticosteroids in the treatment of severe
sepsis and septic shock in adults: a systematic review. JAMA 301:23622375
116. Atkins D, Best D, Briss PA et al (2004) Grading quality of evidence and strength of
recommendations. BMJ 328:14901494
117. Van den Berghe G, Wilmer A et al (2006) Intensive insulin therapy in the medical ICU.
N Engl J Med 354:449461
118. Preiser J, Devos P, Ruiz-Santana S et al (2009) A prospective randomised multi-centre
controlled trial on tight glucose control by intensive insulin therapy in adult intensive care
units: the Glucontrol study. Intensive Care Med 35:17381748
119. Brunkhorst FM, Engel C, Bloos F et al (2008) Intensive insulin therapy and pentastarch
resuscitation in severe sepsis. N Engl J Med 358:125139
120. The NICE-SUGAR Study Investigators (2009) Intensive versus conventional glucose
control in critically ill patients. N Engl J Med 360(13):12831297
121. Wiener RC, Wiener RC, Larson RJ (2008) Benefits and risks of tight glucose control in
critically ill adults: a meta-analysis. JAMA 300(8):933944
122. Silvestri L, van Saene HKF, Weir I et al (2009) Survival benefit of the full selective
digestive decontamination regimen. J Crit Care 24:474.e7474.e14
123. Oostdijk EA, De Smet AM, Blok HE et al (2010) Ecological effects of selective
Med 181(5):452457
ICU-Acquired Infection: Mortality,
Morbidity, and Costs 29
J. C. Marshall and K. A. M. Marshall
29.1 Introduction
Nosocomial infection is a common complication of critical illness [1]. Ambiguities

in diagnostic criteria and variation in patient demographics render reliable esti-
mates of its prevalence imprecise, but infection is generally reported to complicate
the course of between a quarter and a half of all intensive care unit (ICU)
admissions [24] and to be associated with increased morbidity and mortality rates
and costs. ICU-acquired infection, however, is not a random complication but,
rather, one that affects the sickest and most vulnerable of patients, and it has been
considered by many to be not so much a cause of morbidity as a reflection of the
state of being critically ill [5, 6].
Decisions about optimal strategies for preventing and managing ICU-acquired
infection hinge on assumptions about the attributable morbidity and costs of such
infections, assumptions that are only tenuously grounded in robust data. The
adoption of strategies to prevent nosocomial infection presupposes that these
infections produce excess morbidity or generate increased costs of care; the
decision to treat nosocomial infections assumes that eradicating infection will
improve the patients clinical status and so reduce morbidity. These assumptions,
self-evident on initial inspection, require closer scrutiny. This chapter reviews
the concepts of attributable morbidity, mortality, and costs of nosocomial
ICU-acquired infection, with the objective of providing an informed basis for
clinical decision making.
J. C. Marshall (&)
Department of Surgery and Interdepartmental Division of Critical Care,
General Hospital and University of Toronto, Toronto, Canada
470 J. C. Marshall and K. A. M. Marshall
29.2 Outcome Analyses: Methodological Considerations
Estimating the impact of nosocomial infection on an outcomewhether mortality,

morbidity, or costis a complex process, the conclusions of which are critically
dependent on assumptions made by the investigator, on the variables considered
in constructing the analytic model, and, in the case of cost analyses, whose
perspective is being consideredthe patient, the hospital, or society.
29.2.1 Crude and Attributable Mortality
ICUs concentrate a variety of life-sustaining technologies with the explicit

objective of intervening to avert patient death. The implication, therefore,
appears to be that mortality is an unequivocal measure of the success or failure
of ICU care. However, a moments reflection reveals the limitations inherent in
such an assumption. First, mortality as an outcome measure is time dependent:
whereas 1 h survival might be an appropriate measure of the effectiveness of an
intervention during cardiac arrest, successful treatment for cancer is generally
measured as 5-year survival. The duration of survival that best defines success or
failure in the ICU is unclear. Regulatory agencies have adopted the philosophy
that 28-day mortality is the best measure of the effectiveness of novel mediator-
directed therapies [7]. However, a patient who dies in the ICU after 47 days can
hardly be considered a treatment success, whereas another patient who survives
an acute illness to be discharged from hospital on the 13th day but who dies
when struck by a car while heading home does not really represent a treatment
failure.
Second, mortality is of variable importance to ICU patients and their families
[8]. Critical illness is often a complication of other diseases developing in an
already compromised host. For the elderly patient whose health has been failing,
death may be less a concern than the prospect of institutionalization and loss of
independence. Finally, for the patient with multiple comorbid conditions, it may be
very difficult to ascribe mortality to a specific condition. Is pneumonia that
develops in the patient admitted with congestive heart failure, diabetes, and
advanced chronic obstructive pulmonary disease the cause of death, or simply a
reflection of the further deterioration of a tenuous state of health? The concept of
attributable mortality attempts to define the incremental increase in mortality that
results from the disease process of interest or the decrement in mortality that
occurs as the consequence of an intervention.
The most reliable method of defining attributable mortality of nosocomial ICU-
acquired infection is by means of a randomized control trial (RCT) of an inter-
vention that reduces the risk of infection or that attenuates its severity. The
technique of randomization ensures that patient characteristics and health care
interventions that might independently impact on outcomeboth those that
are known and those that are unknownare roughly equally distributed between
the two populations. As the only variable that differs systematically between the
29 ICU-Acquired Infection: Mortality, Morbidity, and Costs 471
two populations is the experimental intervention, documentation of differential

survival between the study populations implicates the intervention as the cause of
that difference.
Clinical trials are costly and time consuming, and alternate approaches have been
used to derive estimates of attributable morbidity from observational data. The
technique of logistic regression analysis relates an outcome (survival or death) to a
number of potential explanatory factors (infection, acute severity of illness, pre-
morbid health status, etc.). A univariate analysis is generally performed first to
identify those variables associated with the outcome of interest. Those that are found
to be predictive of outcome (a p value of 0.10 is typically chosen for this analysis) are
then entered into a stepwise analysis that identifies the independent determinants of
outcome by adjusting for the strongest prognostic variable to determine the inde-
pendent contribution of the next most potent predictor variable. The process is then
repeated until none of the remaining variables is significantly related to the outcome
of interest. The conclusions drawn from such an analysis depend on the potential
explanatory variables chosen, which in turn reflect the assumptions and biases of the
investigator. For example, although the adequacy (or lack of) of antibiotic therapy is
commonly considered as a potential explanatory variable for adverse outcome in
nosocomial infection, the corollarythe effect of administering unnecessary anti-
bioticsis not, thus potentially biasing the analysis toward the conclusion that use of
broad-spectrum antibiotics is an important determinant of outcome.
Another approach to quantifying attributable mortality entails the use of a
matched cohort study. Each patient with the risk factor of interest is matched to a
contemporaneous control with similar baseline characteristics but lacking the risk
factor. If we are interested in the attributable mortality of bacteremia, for example,
we might match each patient with bacteremia with a control patient whose age,
gender, and admission acute physiology and chronic health evaluation (APACHE)
II score was the same as our case. Such an approach can only control for the
variables used in matching and may overestimate the attributable outcome because
of the unrecognized influence of variables that are not used in the matching pro-
cess. In our example, we would not be matching for events that occurred following
ICU admission that might modify mortality risk, such as the development of renal
failure or the need for prolonged mechanical ventilation. The use of a sensitivity
analysisaltering the assumptions associated with the use and weighting of dis-
crete variablescan increase ones confidence in the conclusions reached [9].
A variation on this technique was employed by Connors et al. [10] to estimate
the mortality attributable to the use of a Swan Ganz catheter. Recognizing that
there was considerable variability in the use of the catheter from one unit to the
next, the authors developed a propensity score that reflected the factors that pre-
dicted use of the catheter in those units where it was used frequently. Patients in
low-use units were then matched by propensity score, and the mortality rates of the
two groups were compared. The authors conclusion was that use of the right heart
catheter was associated with an attributable increase in mortality of 24%. Once
again, however, the conclusion is potentially biased by those variables that are not
included in the matching process.
The most reliable method of quantifying outcomes attributable to a particular risk

factor, however, is by means of an RCT. In an adequately powered study, random-
ization controls for both known and unknown confounding factors by distributing
them equally between the two groups. Other methods of estimating the impact of a
factor such as infection on outcome must be interpreted with caution [11].
29.3 Cost Analyses
Some general comments on economic analysis are in order prior to a consideration

of the differing approaches to estimating therapy costs. Costs are inherently dif-
ficult to calculate. The cost of an intervention includes both explicit coststhe
costs expended to achieve the outcome desiredand implicit coststhe potential
revenues that are not realized as a result of the decision that is made. For example,
an intervention that increases survival for an individual patient may increase ICU
costs by limiting the admission of a less seriously ill patient who might have
consumed fewer resources and brought in greater revenues from a payer. Usually,
such implicit costs are not incorporated into analyses of the costs of medical
therapy, for their perspective is not that of the patient but of the payer. Implicit
costs, however, have a potent impact on decisions regarding whether a particular
patient or population of patients will be cared for at a given institution and so
shape patterns of delivery of care within the health system. Explicit costs include
fixed coststhose such as staffing or building maintenancethat are incurred
regardless of the specific intervention, and variable coststhose such as the costs
of drugs, operative procedures, and radiological investigations that depend on the
specific decisions made for a particular patient.
Four types of economic analysis are typically employed to compare treatments
(Table 29.1) [12].
1. Cost-minimization analysis: proceeds from the assumption that two interven-
tions result in comparable outcomes, and seeksby enumerating the spectrum
of costs associated with each alternativeto identify the one that results in the
lowest costs. The units of analysis are monetary units: dollars, euros, etc.
2. Costbenefit analysis: considers both the costs and the outcome in monetary
units and so must assign a specific dollar value to the outcome to be achieved
the cost of a life or other clinical outcometo create meaningful comparisons.
3. Cost-effectiveness analysis: considers the costs generated per unit of clinical
benefit, for example, the cost per life saved or episode of pneumonia prevented.
4. Cost-utility analysis: uses a weighted measure of outcomequality adjusted
life yearsto express costs incurred for a comparable degree of clinical benefit.
This strategy permits comparison of the costs of very divergent therapies; for
example, liver transplantation with ventilator-associated pneumonia (VAP)
prevention. Utilities reflect the value placed on an outcome by patients and are
developed using techniques such as a standard gamble, time tradeoff, or through
administration of a questionnaire.
Table 29.1 Types of economic analyses (adapted from Second American Thoracic Society
Workshop on Outcomes Research [12])
Type of study Numerator Denominator (outcome) Comments

(costs)
Cost Dollars None Compares costs of two or more
minimization therapies without consideration of
consequences
Cost benefit Dollars Dollars Converts clinical effects into
associated costs
Cost- Dollars Measure of clinical Expresses costs in terms of
effectiveness efficacy outcome of intervention (e.g., cost
per case of pneumonia prevented)
Cost utility Dollars Quality-adjusted life Expresses costs relative to a
years standardized estimate of benefit
and so allows comparisons of
differing therapies
Each approach recognizes that patient-centered values represent a compromise

between the outcome achieved and the adverse consequences of the interventions
required to achieve that outcome. Whereas a cost-utility analysis provides the most
relevant measure of costs related to patient-centered outcomes, utilities are difficult
to measure and are dependent on a number of highly subjective assumptions.
29.4 Attributable Mortality of ICU-Acquired Infection
The most compelling evidence that nosocomial ICU-acquired infection is

responsible for increased morbidity and mortality rates derives from RCTs of
selective decontamination of the digestive tract (SDD). The most recent meta-
analysis from 2009, that of 6,914 patients enrolled in randomized trials of SDD,
documented a significant reduction in mortality rates from 30 to 24% in patients
receiving the prophylactic regimen, in association with a reduced odds ratio (OR)
for developing pneumonia to 0.28 [95% confidence interval (CI) 0.200.38] [13].
Whereas it is possible that the benefits of the intervention arose for reasons other
than preventing subsequent nosocomial infection (for example, treating occult
community-acquired infection or reducing the absorption of endotoxin from the
gastrointestinal tract), the most plausible interpretation of these data is that nos-
ocomial infection is responsible for a relative increase in the risk of ICU mortality
rates of 22% (an absolute 5% increase from a baseline risk of 24%). For surgical
patients, a population in which comorbid diseases are less prevalent, the relative
risk reduction is even greater [14], consistent with the hypothesis that the attrib-
utable morbidity and mortality of infection is greatest in patients with minimal
concomitant risk factors for an adverse outcome.
Table 29.2 Attributable

Type of infection Attributable mortality (%)
mortality of nosocomial
infection in the intensive care Bloodstream infection 050
unit (ICU)
Primary bacteremia 20
Catheter-related bacteremia 12
Secondary to nosocomial infection 55
Ventilator-associated pneumonia 2030
Urinary tract infection 5
Attributable mortality is highly population dependent: for patient populations

with minimal comorbidity; the signal attributable to an acute event is relatively
greater than it is for patient populations with significant comorbidities. Thus,
prophylactic strategies such as SDD show their greatest benefit in patients with no
significant premorbid illness and acute physiological derangement that is only
mild to moderate. Available estimates of attributable mortality of common
ICU-acquired infections are summarized in Table 29.2.
29.4.1 Bloodstream Infections
A cohort study of primary bloodstream infections in ICU patients found an overall

incidence of 1%, increasing to 4%or 4.0 infections per 1,000 catheter-daysin
patients with central venous catheters [15]. Similar figures have been reported for
pediatric patients. Gray et al. [16] found an incidence of nosocomial bloodstream
infection of 25/1,000 ICU admissions, or 6.8/1,000 bed-days. Crude mortality rates
for patients with bloodstream infection were higher (26.5 vs. 8.1%); however, 87%
of patients with these infections had significant underlying disease [16]. The
mortality associated with ICU-acquired bacteremia, however, is higher in adults,
with rates ranging from 43 to 53% [15, 1719]. Premorbid illness, reflected, for
example, in a do not resuscitate order, is an important independent predictor of
mortality. However, bacteremia was also an independent risk factor for death in a
study of 1,052 patients with severe sepsis or septic shock [20].
A French study of [2,000 critically ill patients reported that bacteremia occurs
in 5% of patients who remain in the ICU for [48 h. When the episode of bac-
teremia was judged to be primary (no obvious focus), the excess mortality
attributable to the event was 20%, whereas catheter-related bacteremia led to an
excess mortality of 11.5%. When it was judged as secondary nosocomial blood-
stream infection, the excess mortality attributable to the bacteremia was estimated
to be 55%, and patients remained in the ICU for a median of 9.5 days longer [21].
In contrast, a study of 2,076 episodes of infection in surgical patients, employing
logistic regression analysis to control for the potentially confounding influences of
comorbidities, concluded that although associated with critical illness and death,
bacteremia is not independently predictive of outcome but is, rather, a surrogate

measure of underlying severity of illness [22]. Digiovine et al. [23], in a study of
primary bacteremia in ICU patients, reported that when stratified by illness
severity, bacteremic patients did not experience a higher mortality risk, although
the length of ICU stay was extended by a median of 5 days [23]. Our own work
suggests that when patients are stratified by the clinical severity of the inflam-
matory response, bacteremia per se does not contribute to adverse outcome but
may, paradoxically, be associated with a superior outcome [3].
Several studies have attempted to evaluate the impact of bacteremia with
particular pathogens on ICU outcome. Harbarth et al. [24] evaluated 1,835 epi-
sodes of Gram-negative bacteremia in Swiss hospitals and found that bacteremia
caused by Klebsiella spp. or by Pseudomonas aeruginosa, but not by multi-
drug-resistant organisms, was associated with an elevated risk of death. Blot et al.
[25] matched 53 patients with P. aeruginosa bacteremia to 106 controls based on
APACHE II score and diagnosis. Hospital mortality was higher for cases (62.3 vs.
47.2%), but APACHE proved to be the only predictor of survival in a multivariate
analysis. These investigators performed a similar study of 73 critically ill patients
with candidemia and found that although candidemia was associated with a pro-
longed ICU and hospital stay, it did not increase the mortality risk, which was
determined primarily by age, acute illness severity, and preexisting underlying
disease [26]. In contrast, the authors found that bacteremia with methicillin-
resistant Staphylococcus aureus (MRSA) carried a significantly elevated attrib-
utable mortality rate of 23.4% [27].
29.4.2 Ventilator-Associated Pneumonia
Fagon et al. [28] evaluated the impact of VAP on ICU outcome. They demonstrated
a crude mortality rate of 52.4% for 328 patients developing pneumonia, signifi-
cantly higher than the 22.4% ICU mortality experienced by the 1,650 patients who
did not develop pneumonia. The APACHE II score, number of failing organs,
presence of pneumonia, development of nosocomial bacteremia, presence of sig-
nificant underlying disease, and admission from another ICU were all indepen-
dently associated with adverse outcome by logistic regression analysis. Similarly, a
French casecontrol study suggested that VAP was responsible for a twofold
increase in mortality and a 5-day prolongation of ICU stay [29]. A Canadian study
found that pneumonia was associated with an increased length of stay and a trend
toward an increased mortality rate [9]. In contrast, a Spanish study of 1,000 con-
secutive ICU admissions found that the development of VAP increased the length
of stay, but not the mortality rate, of ventilated ICU patients [30]. This conclusion
was also reached by a French matched casecontrol study, which showed rates of
VAP to be comparable in patients who died while in the ICU compared with those
who survived, when matched on the basis of a panel of risk factors for adverse
outcome [31]. A study of pneumonia complicating the course of acute respiratory
distress syndrome reported that the development of VAP increases the duration of
mechanical ventilation but does not adversely impact on patient survival [32].
Pooled data from a systematic review suggest that the development of VAP results
in an increased ICU length of stay of 4 days and an attributable mortality rate of
2030% [33].
29.4.3 Other Nosocomial Infections
There is little available evidence regarding the attributable morbidity and mortality
rates of other nosocomial infections in critically ill patients. Urinary tract infec-
tions, although relatively common, are generally thought to be of only modest
clinical significance. For example, an Argentinian study reported that catheter-
related infections, the most common ICU-acquired infection (comprising 32% of
all infections), carry an attributable mortality of 25% and are associated with an
excess length of stay of 11 days. VAP (25% of all nosocomial infections) was
associated with a 35% attributable mortality and a prolongation of ICU stay of
10 days. In contrast, urinary tract infections (23% of infections) had a 5%
attributable mortality and an increased length of ICU stay of 5 days [34].
29.4.4 Summary: Attributable Mortality of ICU Infections
Estimates of the attributable mortality of the two most common nosocomial

ICU-acquired infectionsbacteremia and VAPare highly variable depending on
study methodology, country of origin, and criteria used to define an optimal
control population. The most commonly used techniques of analysismatched
cohort studies or population studies using multivariate techniques to control for
potential confoundersyield estimates that likely overestimate the true attribut-
able morbidity, for they are unable to completely eliminate the effects of
unmeasured confounders. Moreover, inherent uncertainty in the diagnostic criteria
used to define nosocomial infections introduces substantial uncertainty into the
estimate of attributable outcome. Estimates of attributable morbidity are generally
limited to estimates of prolongation of ICU stay. These, however, uniformly
indicate an increased burden of illness resulting from the development of infection.
Because of the uncertainty and bias inherent in retrospective studies, the most
reliable estimates of the attributable mortality are those deriving from prospective
randomized trials of interventions to prevent infection. Pooled data from trials of
SDD suggest a 5% absolute and a 22% relative attributable mortality associated with
ICU-acquired infection [13]. Even this estimate, however, must be interpreted with
caution, for it is not possible to determine whether the increased risk arises from the
infection or from the additional interventions undertaken to treat that infection. The
distinction is subtle but important. Two RCTs of antibiotic minimization strategies in
the ICU both demonstrated that antimicrobial resistance is minimized [35] and
mortality reduced [36] when explicit diagnostic and therapeutic approaches are used
to minimize antibiotic exposure in the patient with suspected VAP.
29.5 Costs of Nosocomial Infection
Reliable estimates of the costs of an ICU complication such as nosocomial

infection are difficult to determine for the same reasons that attributable outcomes
are difficult to estimate. However, if we adopt the plausible assumption that there
is at least some degree of mortality and morbidity attributable to the development
of a nosocomial infection, it follows that interventions that can prevent infection
can reduce that toll: whether they are worth the cost is the domain of cost-
effectiveness analysis.
Cost-effectiveness analysis is predicated on the assumption that an intervention
has an effect, and reflects the tradeoff that must be made between the cost of an
intervention and the extent to which clinical benefit is achieved. In Fig. 29.1, the
cost-effectiveness of a variety of ICU therapies is presented. Increasing cost is
represented on the y axis, increasing benefit on the x axis. Obviously, the ideal
therapy would be one that reduced costs while bringing increased benefit to the
patient, the situation reflected in the lower right quadrant of the graph and
exemplified by the use of a lower transfusion threshold in the anemic critically ill
patient [37]. An intervention such as the use of activated protein C brings clinical
benefit but at a substantial cost [38], whereas the use of growth hormone increases
costs and morbidity [39] and so is clearly undesirable. It will be appreciated that
ascertainment of the net costs and benefits is challenging and that the results are
both qualitative and potentially controversial.
The perspective of the analysis also differs depending upon whether the primary
effect of the intervention is to treat a disease or to prevent its occurrence. If we assume
that measures taken to prevent a disease will have few or no consequences on long-
term health-related quality of life but that the treatment of a disease will not nec-
essarily restore a patient to his or her full premorbid state of health, then the tradeoff
in prevention is the cost of the intervention against the percentage of patients who
might develop the complication. The tradeoff for a therapeutic measure is the cost of
the intervention against the predicted improvement in health-related quality of life.
The former is a cost-effectiveness analysis, the latter a cost-utility analysis.
29.5.1 Costs of ICU-Acquired Infection
Several authors have attempted to generate estimates of the costs associated with
infection in critically ill patients. Angus et al. [40] used administrative data to
generate an estimate that there are approximately 750,000 new cases of severe
sepsis (sepsis in association with organ dysfunction) in the United States each year
and that the total costs attributable to these are US $16.7 billion annually. However,
these estimates include cases of community-acquired sepsis and sepsis developing in
patients who are not in an ICU, and they fail to include the costs of episodes of
nosocomial ICU-acquired infection that do not meet the criteria for severe
sepsis. Brun-Buisson et al. [41] found that the costs of sepsis in association with
Fig. 29.1 The relationship between cost and clinical effectiveness. Increasing costs are
represented on the y axis, with increasing clinical benefit on the x axis. Examples shown are
approximations of the incremental cost associated with increased (or reduced) clinical benefit
ICU-acquired infection were three times higher than those incurred when sepsis was
present at the time of ICU admission. The total costs were approximately 40,000
for each patient developing sepsis in the ICU. Nosocomial infection complicating
community-acquired sepsis increased costs by 2.5 times. These observations mirror
those of a British study that found a fivefold increase in costs when patients devel-
oping sepsis after the second day of their ICU stay were compared with patients
admitted with a diagnosis of sepsis [42]. Thus, preventing nosocomial infection in the
ICU has the potential to significantly impact on the costs of ICU care.
29.5.2 Nosocomial Bacteremia
Pittet et al. [43] reported that nosocomial bloodstream infections complicate the
course of 3% of patients admitted to an ICU and prolong both the ICU and hospital
stay, generating costs of approximately US $40,000 per survivor. Similar estimates
have been derived by others. Digiovine et al. [23], for example, found that
although nosocomial bloodstream infection in the ICU did not increase mortality
rates, it was associated with increased direct costs of US $34,508 per episode,
whereas Dimick et al. (unpublished data) suggested that catheter-related blood-
stream infection in the ICU results in increased total hospital costs of US $56,167
per case. Two analyses concluded that the use of antibiotic-coated catheters is cost
effective in preventing nosocomial bloodstream infections [44, 45]. However, the
conclusion is highly dependent on the estimate of the efficacy of such catheters,
and intrinsic limitations in the design of the studies evaluating them limit the
estimate of their benefits [46].
29.5.3 Ventilator-Associated Pneumonia
Warren et al. [47] estimated the attributable cost of an episode of VAP in the United
States to be approximately US $12,000. Because a variety of prophylactic strategies
have been shown to be effective in preventing VAP [48], and because these are
generally relatively inexpensive to institute, VAP prevention is readily demonstrable
to be cost effective. Zack et al. [49], for example, showed that instituting a com-
prehensive preventive program resulted in a 57.6% reduction in VAP rate and in cost
savings of as much as US $4 million/year. Other strategies, such as minimizing
intubation through the use of noninvasive positive-pressure ventilation [50] and
reducing the frequency of ventilator circuit changes [51] are also cost effective.
29.5.4 Infection with Antibiotic-Resistant Organisms
Independent of infection site, nosocomial infection with resistant organisms is

associated with increased costs [52]. Chaix et al. [53] undertook a casecontrol study
of patients with MRSA and reported that the attributable costs of such infections were
US $9,275 per episode. An infection control program that could lower transmission
by 14% or more was calculated to be cost effective. Infection with Enterobacter spp.
resistant to third-generation cephalosporins was associated with an attributable cost
of US $29,379 per case [54]. Divergent conclusions have been drawn with respect to
infection with vancomycin-resistant enterococci (VRE). Whereas one matched
cohort study suggested an 11% increase in mortality rates and increased hospital
costs of more than US $20,000 per case [55], a second study using stepwise logistic
regression analysis failed to demonstrate that such infections were associated with
either attributable morbidity or increased costs [56].
Because indiscriminate antibiotic use is a risk factor for the emergence of
resistance, implementation of restrictive antimicrobial prescribing practices cannot
only reduce costs but also limit the development of resistance in an ICU envi-
ronment [35, 57].
29.6 Costs of Effective Treatment
Our focus has been on infectious complications that are amenable to prevention, in
no small part because there are few proven effective therapies for infection in the
ICU. Thus, although it is widely believed that specific antimicrobial therapy,
adequate surgical source control, and the spectrum of supportive measures that
comprise ICU care will improve clinical outcome, the attributable effect of any of
Table 29.3 Proven and

Strategy Examples
promising strategies to reduce
the morbidity and cost of Prevent abnormal Selective digestive tract
ICU-acquired infection colonization decontamination
Minimize antibiotic exposure
Enteral feeding
Reduce device-related Antibiotic-coated catheters
infection Noninvasive ventilation
Coated endotracheal tubes or Foley
catheters
Reduced frequency of ventilator
circuit changes
Prevent aspiration Semi-recumbent positioning
these in the patient with infection is unknown, and therefore cost-utility analyses
are impossible.
The approval of activated protein C for treating patients with severe sepsis has
provided the first opportunity for cost-utility analyses in critically ill infected
patients. Treatment with activated protein C has been shown in a cohort of patients
with severe sepsis to reduce mortality rates by 6.1% but at a cost of approximately
US $7,000 per course of therapy. Two independent analyses of the cost-
effectiveness of activated protein C show a favorable profile when it is used in the
sickest patients. Manns et al. [38] calculated the cost per life-year gained to be US
$27,936 for all patients in the cohort, whereas Angus et al. [58] suggested that the
cost-utility of activated protein C is US $48,800 per quality-adjusted life year.
29.7 Conclusions
The development of nosocomial infection in the critically ill patient results in

increased morbidity and significantly increases the costs of care. Although the
magnitude of both effects is difficult to know with certainty, it is clear that mea-
sures that can prevent nosocomial infection in the critically ill patient can both
reduce costs and improve clinical outcomes (Table 29.3). More rigorous evalua-
tion of the economic impact of these is needed.
References
1. National Nosocomial Infections Surveillance System (NNIS) (2002) System report, data
summary from January 1992 to June 1992, issued August 2002. Am J Infect Control 30:458475
intensive care units in the United States. National nosocomial infections surveillance system.
3. Marshall JC, Sweeney D (1990) Microbial infection and the septic response in critical
surgical illness. Sepsis, not infection, determines outcome. Arch Surg 125:1723
4. Ponce de Leon-Rosales SP, Molinar-Ramos F, Dominguez-Cherit G et al (2000) Prevalence of

infections in intensive care units in Mexico: a multicenter study. Crit Care Med 28:13161321
5. Fagon JY, Novara A, Stephan F et al (1994) Mortality attributable to nosocomial infections in
the ICU. Infect Control Hosp Epidemiol 15:428434
6. Nathens AB, Chu PTY, Marshall JC (1992) Nosocomial infection in the surgical intensive
care unit. Infect Dis Clin North Am 6:657
7. Schwieterman W, Roberts R (1997) FDA perspective on study design for therapies for severe
sepsis. Sepsis 1:69
8. Patrick DL, Pearlman RA, Starks HE et al (1997) Validation of preferences for life-sustaining
treatment: implications for advance care planning. Ann Intern Med 127:509517
9. Heyland DK, Cook DJ, Griffith L, Keenan SP, Brun-Buisson C (1999) The attributable
morbidity and mortality of ventilator-associated pneumonia in the critically ill patient. Am J
10. Connors AF Jr, Speroff T, Dawson NV et al (1996) The effectiveness of right heart
catheterization in the initial care of critically ill patients. JAMA 276:889918
11. Asensio A, Torres J (1999) Quantifying excess length of postoperative stay attributable to
infections: a comparison of methods. J Clin Epidemiol 52:12491256
12. Second American Thoracic Society Workshop on Outcomes Research (2002) Understanding
costs and cost-effectiveness in critical care. Am J Respir Crit Care Med 165:540550
tract infections and mortality in adults receiving intensive care (Cochrane Review). In: The
Cochrane library, issue 1. Wiley, Chichester
patients. Arch Surg 134:170176
15. Warren DK, Zack JE, Elward AM et al (2001) Nosocomial primary bloodstream infections in
intensive care unit patients in a non-teaching community medical center: a 21-month
prospective study. Clin Infect Dis 33:13291335
16. Gray J, Gossain S, Morris K (2001) Three-year survey of bacteremia and fungemia in a
pediatric intensive care unit. Pediatr Infect Dis J 20:416421
17. Jamal WY, El-Din K, Rotimi VO, Chugh TD (1999) An analysis of hospital-acquired bacteraemia
in intensive care unit patients in a university hospital in Kuwait. J Hosp Infect 43:4956
18. Crowe M, Ispahani P, Humphreys H et al (1998) Bacteraemia in the adult intensive care unit
of a teaching hospital in Nottingham, UK, 19851996. Eur J Clin Microbiol Infect Dis
17:377384
19. Pittet D, Thievent B, Wenzel RP et al (1996) Bedside prediction of mortality from bacteremic
sepsis. A dynamic analysis of ICU patients. Am J Respir Crit Care Med 153:684693
20. Brun-Buisson C, Doyon F, Carlet J et al (1995) Incidence, risk factors, and outcomes of
severe sepsis and septic shock in adults. A multicenter prospective study in intensive care
units. JAMA 274:968974
21. Renaud B, Brun-Buisson C (2001) Outcomes of primary and catheter-related bacteremia. A
cohort and case-control study in critically ill patients. Am J Respir Crit Care Med 163:15841590
22. Raymond DP, Pelletier SJ, Crabtree TD, Gleason TG, Pruett TL, Sawyer RG (2001) Impact of
bloodstream infection on outcomes among infected surgical inpatients. Ann Surg 233:549555
23. Digiovine B, Chenoweth C, Watts C, Higgins M (1999) The attributable mortality and costs
of primary nosocomial bloodstream infections in the intensive care unit. Am J Respir Crit
Care Med 160:976981
24. Harbarth S, Rohner P, Auckenthaler R et al (1999) Impact and pattern of gram-negative
bacteraemia during 6 y at a large university hospital. Scand J Infect Dis 31:163168
25. Blot S, Vandewoude K, Hoste E, Colardyn F (2003) Reappraisal of attributable mortality in
critically ill patients with nosocomial bacteraemia involving Pseudomonas aeruginosa.
26. Blot SI, Vandewoude KH, Hoste EA, Colardyn FA (2002) Effects of nosocomial candidemia
on outcomes of critically ill patients. Am J Med 113:480485
27. Blot SI, Vandewoude KH, Hoste EA, Colardyn FA (2002) Outcome and attributable
mortality in critically ill patients with bacteremia involving methicillin-susceptible and
methicillin-resistant Staphylococcus aureus. Arch Intern Med 162:22292235
28. Fagon J-Y, Chastre J, Vuagnat A et al (1996) Nosocomial pneumonia and mortality among
patients in intensive care units. JAMA 275:866869
29. Bercault N, Boulain T (2001) Mortality rate attributable to ventilator-associated nosocomial
pneumonia in an adult intensive care unit: a prospective case-control study. Crit Care Med
29:23032309
30. Rello J, Quintana E, Ausina V et al (1991) Incidence, etiology, and outcome of nosocomial
pneumonia in mechanically ventilated patients. Chest 100:439444
31. Bregeon F, Ciais V, Carret V et al (2001) Is ventilator-associated pneumonia an independent
risk factor for death? Anesthesiology 94:554560
32. Markowicz P, Wolff M, Djedaini K et al (2000) Multicenter prospective study of ventilator-
associated pneumonia during acute respiratory distress syndrome. Am J Respir Crit Care Med
161:19421948
33. Cook D (2000) Ventilator associated pneumonia: perspectives on the burden of illness.
Intensive Care Med 26(1):S31S37
34. Rosenthal VD, Guzman S, Orellano PW (2003) Nosocomial infections in medical-surgical
intensive care units in Argentina: attributable mortality and length of stay. Am J Infect
Control 31:291295
patients with pulmonary infiltrates in the intensive care unit. A proposed solution for
indiscriminate antibiotic prescription. Am J Respir Crit Care Med 162:505511
36. Fagon J-Y, Chastre J, Wolff M et al (2000) Invasive and noninvasive strategies for
management of suspected ventilator-associated pneumonia. A randomized trial. Ann Intern
Med 132:621630
37. Hebert PC, Wells G, Blajchman MA et al (1999) A multicentre randomized controlled
clinical trial of transfusion requirements in critical care. N Engl J Med 340:409417
38. Manns BJ, Lee H, Doig CJ et al (2002) An economic evaluation of activated protein C
treatment for severe sepsis. N Engl J Med 347:9931000
39. Takala J, Ruokonen E, Webster NR et al (1999) Increased mortality associated with growth
hormone treatment in critically ill adults. N Engl J Med 341:785792
40. Angus DC, Linde-Zwirble WT, Lidicker J et al (2001) Epidemiology of severe sepsis in the
United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med
29:13031310
41. Brun-Buisson C, Roudot-Thoaval F, Girou E et al (2003) The costs of septic syndromes
in the intensive care unit and influence of hospital-acquired sepsis. Intensive Care Med
29:14641471
42. Edbrooke DL, Hibbert CL, Kingsley JM et al (1999) The patient-related costs of care for
sepsis patients in a United Kingdom adult general intensive care unit. Crit Care Med
27:17601767
43. Pittet D, Tarara D, Wenzel RP (1994) Nosocomial bloodstream infection in critically ill
patients. Excess length of stay, extra costs, and attributable mortality. JAMA 271:15981601
44. Shorr AF, Humphreys CW, Helman DL (2003) New choices for central venous catheters:
potential financial implications. Chest 124:275284
45. Marciante KD, Veenstra DL, Lipsky BA, Sainst S (2003) Which antimicrobial impregnated
central venous catheter should we use? modeling the costs and outcomes of antimicrobial
catheter use. Am J Infect Control 31:18
46. McConnell SA, Gubbins PO, Anaissie EJ (2003) Do antimicrobial-impregnated central
venous catheters prevent catheter-related bloodstream infection? Clin Infect Dis 37:
6572
47. Warren DK, Shukla SJ, Olsen MA et al (2003) Outcome and attributable cost of ventilator-
associated pneumonia among intensive care unit patients in a suburban medical center. Crit
Care Med 31:13121317
48. Kollef MH (1999) The prevention of ventilator-associated pneumonia. N Engl J Med
340:627634
49. Zack JE, Garrison T, Trovillion E et al (2002) Effect of an education program aimed at
reducing the occurrence of ventilator-associated pneumonia. Crit Care Med 30:24072412
50. Sinuff T, Cook DJ (2003) Health technology assessment in the ICU: noninvasive positive
pressure ventilation for acute respiratory failure. J Crit Care 18:5967
51. Kotilainen HR, Keroack MA (1997) Cost analysis and clinical impact of weekly ventilator
circuit changes in patients in intensive care unit. Am J Infect Control 25:117120
52. Niederman MS (2001) Impact of antibiotic resistance on clinical outcomes and the cost of
care. Crit Care Med 29(Suppl):N114N120
53. Chaix C, Durand-Zaleski I, Alberti C, Brun-Buisson C (1999) Control of endemic
methicillin-resistant Staphylococcus aureus: a cost-benefit analysis in an intensive care
unit. JAMA 282:17451751
54. Cosgrove SE, Kaye KS, Eliopoulos GM, Carmeli Y (2002) Health and economic outcomes of
the emergence of third-generation cephalosporin resistance in Enterobacter species. Arch
55. Carmeli Y, Eliopoulos G, Mozaffari E, Samore M (2002) Health and economic outcomes of
vancomycin-resistant enterococci. Arch Intern Med 162:22232228
56. Pelz RK, Lipsett PA, Swoboda SM et al (2002) Vancomycin-sensitive and vancomycin-
resistant enterococcal infections in the ICU: attributable costs and outcomes. Intensive Care
Med 28:692697
57. Bantar C, Sartori B, Vesco E et al (2003) A hospital-wide intervention program to optimize
the quality of antibiotic use: impact on prescribing practice, antibitoic consumption, cost
savings, and bacterial resistance. Clin Infect Dis 37:180186
58. Angus DC, Linde-Zwirble WT, Clermont G et al (2003) Cost-effectiveness of drotrecogin
alfa (activated) in the treatment of severe sepsis. Crit Care Med 31:111
Evidence-Based Medicine in ICU
30
A. J. Petros, K. G. Lowry, H. K. F. van Saene and J. C. Marshall
30.1 Introduction
Evidence-based medicine (EBM) was extolled by David Sackett, who described two
processes: one for assessing the quality of a therapy on a scale of 14, and a second
for making recommendations for using that therapy on a scale of AD. However,
more recently, a newer method of grading the quality of evidence and strength of
recommendation for a new therapy has been developed. The Grading of Recom-
mendations Assessment, Development and Evaluation (GRADE) Working Group
reported its suggestions in 2004, with further refinement in 2008 [1, 2]. The use of a
structured approach to collect, analyse and summarise all the relevant evidence
allows the production of grades of recommendations. GRADE is being increasingly
used as the structure on which to develop guidelines [3] and is used widely by the
World Health Organization, the American College of Physicians, the American
Thoracic Society, the Cochrane Collaboration and many other organisations, with
up to 25 groups demonstrating GRADEs success as a methodology [4].
GRADE guides assessment of the quality of evidence for a particular treatment or
therapy in one of four levelshigh (A), moderate (B), low (C) and very low (D).
Study design, quality, consistency and directness are all assessed. The factors
influencing the decision on quality are described in Table 30.1. Evidence from
randomised controlled trials (RCTs) contributes to high-quality evidence, but
confidence in that evidence may be decreased for several reasons: study limita-
tions; inconsistent results; indirectness of evidence; imprecision; reporting bias [1]
(Table 30.1). Observational studies, such as cohort and case-control studies, start
with a low-quality rating; grading upwards may be possible if, for example,
A. J. Petros (&)
PICU, Great Ormond Street Hospital, London, UK
Table 30.1 Descriptions of levels and quality used in the GRADE system
Quality, and confidence in the quality, of GRADE evidence

Quality Level Description
High quality A Further research is very unlikely to change our confidence in the estimate
of effect. Randomized control trials
Moderate B Further research is likely to have an important impact on our confidence in
quality the estimate of effect and may change the estimate
Low quality C Further research is very likely to have an important impact on our
confidence in the estimate of effect and is likely to change the estimate.
Observational studies
Very low D Any estimate of effect is very uncertain. Any other evidence
quality
Confidence in the quality of evidence used in GRADE
Decreases if
Serious (-1) or very serious (-2) limitation to study quality
Important inconsistency (-1)
Some (-1) or major (-2) uncertainty about directness
Imprecise or sparse data (-1)
High probability of reporting bias (-1)
Increases grade if
Strong evidence of association: significant relative risk of[2 (\0.5) based on consistent evidence
from two or more observational studies, with no plausible confounders (+1)
Very strong evidence of association: significant relative risk of [5 (\0.2) based on direct
evidence, with no major threats to validity (+2)
Evidence of a dose-response gradient (+1)
All plausible confounders would have reduced the effect (+1)
the size of the treatment effect is very large, or if there is a strong causal rela-
tionship. GRADE also makes recommendations from strong (1) to weak (2).
The former is where the intervention clearly outweighs its undesirable effects and
the latter where the trade-off between desirable and undesirable effects is less
clear. Making recommendations for a specific therapy involves a balance between
benefits and harms and inevitably involves placing a relative value on each
outcome, though it is difficult to judge how much weight to give to different
outcomes [1]. In making a recommendation, four main factors should be consid-
ered [1] (Table 30.2):
trade-offsthese should consider the estimated size of the effect for the main
outcomes, confidence limits around those estimates, and relative values placed
on each outcome;
quality of evidence;
30 Evidence-Based Medicine in ICU 487
Table 30.2 Recommendations by the GRADE system
Recommendations
Net benefits The intervention clearly does more good than harm
Trade-offs There are important trade-offs between benefits and harms
Uncertain trade-offs It is not clear whether the intervention does more good than harm
No net benefits The intervention clearly does not do more good than harm
translation of evidence into specific practice, allowing for factors that could
qualify the expected effect, such as proximity to a hospital or availability of
necessary expertise;
uncertainty about the baseline risk for the population of interest.
The strong or weak grading is felt to be of greater clinical important than
classifying the quality of the intervention.
Using GRADE provides a framework for structured assessment and can help
ensure that appropriate judgments are made about a new therapy or manoeuvre.
We screened intensive care unit (ICU) literature using these GRADE rules for
manoeuvres that may impact on infectious morbidity and mortality rates and
classified the most common manoeuvres according to levels of evidence and
grades of recommendations (Table 30.3).
30.2 Infection-Control Manoeuvres: Nonantibiotic

Interventions
30.2.1 Hand Washing, Isolation, Protective Clothing,

Equipment Care and Environment
It has never been shown in a RCT that hand hygiene prevents pneumonia and
reduces mortality rates in ventilated patients. The efficacy of hand hygiene on the
incidence of infection was studied in eight nonrandomised studies [512]
(Table 30.4); however, the incidence of pneumonia was not presented. The only
study demonstrating an impact on mortality due to hand hygiene was the cohort
study of Semmelweis in 1861 in postpartum women for reducing mortality due to
puerperal sepsis, with a decrease from 11 to 3% [13]. There are no data available
on the effect of isolation, protective clothing, equipment care and environment on
pneumonia and mortality rates in ventilated patients.
These five traditional infection control measures target microorganism trans-
mission via carriers hands. Although they are important, the impact should not be
overestimated. An optimal infection-control policy can only reduce infections
due to microorganisms acquired in the ICU, i.e. secondary endogenous and
exogenous infections. They fail to influence primary endogenous infections due to
Table 30.3 Analysis of the literature, grading of evidence and recommendations for controlling
morbidity and mortality rates due to infection in ventilated patients in the intensive care unit
Reduced infection Reduced mortality
Level of Grade of Level of Grade of

evidence recommendation evidence recommendation
Nonantibiotic interventions
Handwashing/isolation/protective 2 D 2 D
clothing/care of equipment and
environment
Positioning
Rotation therapy None None None None
Semi-recumbent position None None None None
Subglottic secretion drainage None None None None
Oral antiseptic decontamination None None None None
Immunomodulation
Steroids 1 A None None
Immunoglobulins None None 1 B
Activated protein-C None None 2 C
Anti-inflammatory modulators None None 1 B
None None 1 A
Antibiotic interventions
Selective Decontamination of the 1 A 1 A
Digestive Tract (SDD) (four-
component)
Level: 14; Grade: A high, B moderate, C low, D very low
microorganism present in admission flora. This type of infection is the major

infection problem in the ICU, varying between 60 and 85%.
30.2.2 Positional Therapy
30.2.2.1 Rotational Therapy

Severely ill patients who require ventilation are traditionally treated in the supine
position. This leads to collapse of the lower parts of the lung and reduced clearance
of lower-airway secretions. These two factors increase the risk of pneumonia.
Theoretically, treating a patient in a specialised rotating bed in which the patient is
continuously rotated from -40 to +40 around their longitudinal axis could help
prevent pneumonia. One meta-analysis of six RCTs and two RCTs [1416]
indicated a significant reduction in pneumonia in patients who received rotational
therapy. Of the six studies, five were performed in surgical or neurological
patients. The sixth trial, in which no reduction in pneumonia was found, involved
nonsurgical ICU patients; a more recent RCT in a mixed ICU population did not
30
Table 30.4 Studies into the effect of hand hygiene on the incidence of nosocomial infections, including pneumonia
Author Year Study design No. Outcome: infectious morbidity Evidence

Casewell [5] 1977 Sequential Not mentioned Significant reduction in nosocomial infections during Klebsiella outbreak 2D
Effect on pneumonia not mentioned
Massanari [6] 1984 Crossover 5,859 Significant reduction of nosocomial infection on some ICUs 2A

Maki [7] 1989 Crossover Not mentioned Significant reduction of nosocomial infection on some ICUs 2B
Simmons [8] 1990 Historically controlled Not mentioned No effect 2D
Doebbeling [9] 1992 Crossover 1,894 Significant reduction of nosocomial infection on some ICUs 2A
No effect on pneumonia
Webster [10] 1994 Sequential Not mentioned Control of MRSA outbreak. Significant reduction of nosocomial infections 2D
Koss [11] 2001 Prospective, randomised 153 No effect on pneumonia 2A
Slota [12] 2001 Prospective, randomised 98 No effect on pneumonia 2A
MRSA methicillin-resistant Staphylococcus aureus
489
Table 30.5 Randomised controlled trials and meta-analyses into the effect of nonantibiotic interventions on pneumonia and mortality rates in ventilated
490
patients
Manoeuvre Author Year Study Design No. Pneumonia Mortality Evidence

Rotation therapy Choi [14] 1992 MA 419 RR 0.50 No difference 2A
6 studies P = 0.002
Traver [16] 1995 RCT 103 RR 0.62 RR 0.62 2A
P = 0.21 P = 0.21
Semirecumbent position Silvestri [21] 2010 MA 337 OR, 0.59; 95% OR 0.86 1A
3 studies (0.152.35) (0.541.37)
(P = 0.45) (P = 0.53)
Van Saene [54] 2009 MA 311 OR 0.56 OR 0.81 1A
2 studies (0.065.54) (0.471.41)
Subglottic suction drainage Dezfoulian [32] 2005 MA 896 RR 0.5 Not significant 1A
5 studies (0.35 0.71)
(P \ 0.001)
Silvestri L [33] 2008 MA 1953 OR 0.43 OR 0.93 1A
9 studies (0.320.58) (0.711.21)
(P \ 0.001) (P = 0.57)
Van Saene [54] 2009 MA 1178 OR 0.40 OR 0.99 1A
7 studies (0.280.56) (0.711.38)
(continued)
A. J. Petros et al.
30
Manoeuvre Author Year Study Design No. Pneumonia Mortality Evidence

Oropharyngeal decontamination using antiseptics Pineda [50] 2006 MA 1202 OR 0.42 OR 0.77 1A
4 studies (0.161.06) (0.282.11)
Chlebicki [51] 2007 MA 1650 RR 0.70 RR 1.07 1A
7 studies (0.471.04) (0.761.51)
(P = 0.83) (P = 0.69)
Chan [52] 2007 MA 2144 RR 0.56 RR 0.96 1A
7 studies (0.390.81) (0.691.33)
Kola [53] 2007 MA RR 0.58 Not significnat 1A
7 studies (0.450.74)
van Saene [54] 2009 MA 2752 OR 0.49 OR 0.98 1A

11 studies (0.350.67) (0.641.50)
Carvajal [55] 2010 MA OR 0.56 Not significant 1A
10 studies (0.440.73)
RCT randomised controlled trial, MA meta-analysis, SR systematic review, RR relative risk (95% confidence intervals)
491
support the meta-analysis. Rotational therapy requires special beds, is associated

with considerable costs and is unpleasant for the patient; a cost-effective analysis is
not available.
30.2.2.2 Semirecumbent Position

Although in general the throat has been considered internal source of potential
pathogenic microorganisms (PPMs) causing pneumonia, some believe that aspi-
ration of PPMs carried in the stomach may play a role, the so called stomachlung
route [17]. Based on this concept, ventilating patients in a semirecumbent position
is thought to have a beneficial effect on reducing the incidence of reflux and
aspiration from the stomach, whereby pneumonia in ventilated patients could be
prevented. This manoeuvre has been investigated in three RCTs [1820]
(Table 30.5). The first shows a significant reduction in pneumonia. Mortality rates
however, were identical in both test and control group. Patients who underwent
abdominal or neurosurgery, patients with refractory shock and patients who were
readmitted to ICU within 1 month were excluded. The second RCT, published in
abstract form, failed to confirm these results. There was no difference in pneu-
monia or mortality rates. Treating ventilated ICU patients in a semirecumbent
position at an angle of 45 is difficult in practice and is often associated with
frequent changes in patient position. Keeley [20] was unable to demonstrate any
reduction in ventilator-associated pneumonia (VAP) in patients nursed at 45
(P \ 0.176). Meta-analysis of these studies reveals no significant impact on VAP
by this manoeuvre; cost was not reported [21].
30.2.3 Continuous Aspiration of Subglottic Secretions
Stasis of saliva contaminated with potential pathogens above the cuff on the
endotracheal tube increase the risk of aspiration pneumonia. Removing and pre-
venting this salivary stasis using continuous aspiration via a specially designed
endotracheal tube is thought to prevent pneumonia. The intervention of subglottic
secretion drainage (SSD) was evaluated in ten RCTs [2231] performed in a mixed
ICU population requiring ventilation for [72 h, and the fourth study in cardiac
surgery patients. Results were not consistent. Two studies showed a significant
reduction in pneumonia; the other two failed to show any impact on pneumonia
during ventilation. There was no difference in mortality rates between test and
control groups in any of these studies. Although the specially designed tubes and
suction equipment are expensive, this technique has been suggested to be cost
effective on theoretical grounds. There were no harmful side effects associated
with this manoeuvre in any of the studies. Bo et al. found that the presence of
subglottic secretion may be an origin of the pathogenetic organisms of VAP [27].
The morbidity rate of VAP in mechanically ventilated patients can be reduced by
SSD. Liu et al. confirmed that migration of the dominant bacteria of the subglottic
secretion was one of the important factors for ventilator-associated lower airway
infection [28]. Concentration of bacteria in subglottic secretion was significantly

reduced by subglottic secretion drainage when SSD was used, and SSD reduced
the incidence of ventilator-associated airway infection and pneumonia in patients
ventilated for \5 days. Zheng et al. supported these conclusions [30]. However,
Lorente et al. found that the use of an endotracheal tube with polyurethane cuff and
subglottic secretion drainage helps prevent early- and late-onset VAP [29]. Bouza
et al. demonstrated that continuous aspiration of subglottic secretions reduces the
incidence of VAP in at-risk patients [31].
There are two meta-analyses [32, 33]: Dezfulian et al. reported on five RCTs
involving 896 patients [32]. They found that SSD reduced the incidence of VAP by
nearly 50% (relative risk (RR) 0.51; 0.370.71) by reducing early-onset pneu-
monia (occurring within 57 days after intubation). SSD also shortened the
duration of mechanical ventilation by 2 days and the length of ICU stay by 3 days;
it also delayed the onset of pneumonia by 6.8 days. The authors concluded that
SSD appears effective in preventing early-onset VAP among patients expected to
require [72 h of mechanical ventilation. Silvestri et al. analysed ten RCTs of SSD
[33] (Table 30.5), nine of which reported results on pneumonia rates. In 1,953
patients studied, there was a 57% reduction in VAP [odds ratio (OR) 0.43, 95%
confidence interval (CI) 0.320.58; P \ 0.001] in 1,846 patients in seven RCTs
but no effect on mortality rates. Subglottic drainage seems to be effective in
preventing VAP, though subgroup analysis revealed it was not effective in cardiac
surgery patients.
30.2.4 Oropharyngeal Decontamination Using Antiseptics
There are 16 RCTs that report varying degrees of success with oropharyngeal
decontamination using antiseptics [3449]. However, the outcome of six meta-
analyses revealed that antiseptic usage has no benefit in reducing pneumonia or
mortality rates [5055] (Table 30.5). In 1,202 patients, Pineda et al. [50] reported
that use of oral decontamination with chlorhexidine did not result in significant
reduction in the incidence of nosocomial pneumonia in patients who received
mechanical ventilation, and it did not alter the mortality rate. Chlebicki and Safdar
[51] demonstrated no mortality benefit with chlorhexidine, though in seven small
RCTs there was a reduction in VAP, which was most marked in cardiac surgery
patients. Neither antiseptic nor antibiotic oral decontamination reduced mortality,
duration of mechanical ventilation or ICU stay in a meta-analysis of 11 studies by
Chan et al. [52]. Kola and Gastmeier [53] found in seven RCTs a reduction in RR
of lower respiratory tract infections in patients receiving chlorhexidine [RR
(random): 0.58]. However, these results only applied to patients ventilated for up
to 48 h. From 10 studiesbut not all RCTsCarvajal et al. [55] report a reduction
in the risk of VAP with chlorhexidine (OR 0.56, 95% CI 0.440.73). However, no
reduction in mortality rates, length of mechanical ventilation or ICU length of stay
was seen.
494
Table 30.6 Randomised controlled trials into the effect of nonantibiotic interventions on the general infection and mortality rates in ventilated patients
Manoeuvre Author Year Study Design No. Infection Rate Mortality Evidence
Immunonutrition Beale [56] 1999 Meta-analysis of 12 studies 1,482 RR 0.67 RR 0.05 2A
(0.500.89) (0.781.41)
P = 0.006 P = 0.76
Heyland [57] 2001 Meta-analysis of 22 studies 2,419 RR 0.66 RR 1.1 2A
(0.540.80) (0.931.31)
Steroids Cronin [58] 1995 Metaanalysis of 9 RCTs 1,232 No difference RR 1.13 2A
(0.991.29)
Lefering [59] 1995 Meta-analysis of 10 RCTs 1,329 No difference Difference in mortality 2A
0.2% (9.2 to 8.8)
Bollaert [60] 1998 RCT 41 No difference Difference in mortality 2A
31% (161)
Briegel [61] 1999 RCT 40 No difference No difference 2A
Annane [62] 2002 RCT 300 No difference Significant reduction 2A
RCT randomised controlled trial, RR relative risk (95% confidence intervals)
A. J. Petros et al.
30.2.5 Immunomodulation
30.2.5.1 Immunonutrition (Enteral Feeding)

Total parenteral nutrition (TPN) has been shown to be harmful in terms of
infection rates and liver impairment. This prompted enteral feeding the ICU
patient as soon as possible, because it is thought to be essential for the gut anatomy
and physiology into prevent loss of mucosal integrity and subsequent transloca-
tion. In addition, several nutrients added to the enteral feed have been shown to
influence immunologic and inflammatory responses in humans. There are two
meta-analyses available on immunonutrition in the critically ill [56, 57]
(Table 30.6). Both demonstrate a significant reduction in overall infection rate,
although they do not specifically state pneumonia. There was no reduction in
mortality rate in either meta-analysis. Surgical patients seemed to benefit more
than medical patients. In two large RCTs, mortality rate was significantly higher in
the subgroup who received immunonutrition. Some have speculated that adding
arginine may have been detrimental to the immune system.
30.2.5.2 Steroids
High doses of steroids given to septic patients are thought to be beneficial for three
reasons [5862] (Table 30.6): steroids effectively suppress generalised inflam-
mation due to microorganisms and their toxins; they have been shown to signif-
icantly reduce septic shock and early mortality within 72 h; they have been shown
to significantly reduce mortality rates due to particularly severe invasive infection,
including meningitis, typhoid and Pneumocystis carinii pneumonia (PCP). The
major perceived side effects of high-dose steroids are the associated immune
suppression and subsequent risk of superinfections. Indeed, the two meta-analyses
show a trend towards increased mortality rates from secondary infection in patients
receiving steroids. A systematic review by Annane et al. [63] examining the
benefits and risks of steroids in sepsis reviewed 17 RCTs encompassing 2,138
patients, and three quasi-RCTs of 246 patients. Sub group analysis of prolonged
low-dose corticosteroid therapy suggests a beneficial effect on short-term mortality
rates [63]. The Corticosteroid Treatment and Intensive Insulin Therapy for Septic
Shock in Adults (COIITSS) study [64] demonstrated that intensive insulin therapy
together with hydrocortisone for septic shock did not improve in-hospital mortality
rates. The addition of oral fludrocortisone did not result in a statistically significant
improvement in in-hospital mortality [64].
The next logical step would be to combine steroids with SDD, whereby the
perceived harmful effects of steroids could be abolished. In that way, the early
survival benefit from steroids can be preserved, while keeping the patient free from
secondary infections using SDD. The time has come to perform a randomised trial
of SDD and steroids versus SDD only, with the endpoint as mortality rate.
30.2.5.3 Immunoglobulins
Polyclonal intravenously administered immunoglobulins significantly reduce
mortality rates and can be used as an extra treatment option for sepsis and septic
shock [65]. Overall mortality rates were reduced in patients who received poly-
clonal immunoglobulin i.v. (492; RR 0.64; 95% CI 0.510.80). For the two high-
quality trials on polyclonal immunoglobulin i.v., the RR for overall mortality was
0.30, but the CI was wide (0.090.99; n = 91). However, all trials were small, and
the totality of the evidence is insufficient to support a robust conclusion of benefit.
Adjunctive therapy with monoclonal immunoglobulins i.v. remains experimental.
30.2.5.4 Activated Protein-C

Drotrecogin alfa (activated), or recombinant human activated protein C, is thought
to have anti-inflammatory, antithrombotic and profibrinolytic properties. In a
randomised trial of 1,690 patients, the mortally rate was 30.8% in the placebo
group and 24.7% in the drotrecogin alfa group, which translates into an absolute
reduction in risk of death of 6.2% (P = 0.005). The incidence of serious bleeding
was higher in the drotrecogin alfa (activated) group than in the placebo group [66].
This is level 1 evidence and a grade B recommendation.
30.2.5.5 Anti-Inflammatory Mediators

Almost 60 randomised controlled clinical trials have tested the hypothesis that
modulating the endogenous host inflammatory response can improve survival for
patients with a clinical diagnosis of sepsis. The results have been frustrating, and
no new agent has been introduced into clinical practice for this purpose [67].
Pooled data from studies using a monoclonal antibody to neutralise tumour
necrosis factor (TNF) demonstrates a statistically significant 3.5% reduction in
mortality rates. In aggregate, the three completed studies using recombinant
interleukin-1 (IL-1) receptor antagonists to neutralise IL-1 also showed an absolute
mortality rate reduction of 5%. Zeni et al. [68] showed that he combined results of
all completed trials, independent of the therapeutic agents employed, demonstrate
a statistically significant 3% overall reduction in 28-day all-cause mortality. It is
debateable whether this small clinical benefit is sufficiently important to justify
clinical use of these therapies, given the costs and potential toxicity of the agent
involved.
30.2.5.6 Tight Glucose Control

Van den Berghe et al. [69] demonstrated that intensive insulin therapy reduces
morbidity and mortality rates in cardiac surgical ICUs. However, intensive insulin
therapy significantly reduced morbidity but not mortality rates among all patients
in the medical ICU [70].
The American Diabetes Association and Surviving Sepsis Campaign recom-
mend tight glucose control in critically ill patients based largely on one trial that
shows decreased mortality rates in a surgical ICU. Because similar studies report
Table 30.7 Classification of microorganisms based upon their intrinsic pathogenicity
Intrinsic Type Site Microorganism Flora

pathogenicity
Low level Indigenous Throat Peptostreptococcus, Veillonella spp., Normal
IPI = 0.01 flora Gut Streptococcus viridans
Vagina Bacteroides spp., Clostridium spp.,
Skin enterococci, E. coli
Peptostreptococcus, Bacteroides spp.,
Lactobacillus spp.
Propionibacterium acnes, coagulase-negative
staphylococci
Potential Normal PPM Throat Streptococcus pneumoniae, Haemophilus Normal
IPI = 0.3 Abnormal Gut influenzae, Moraxella catarrhalis, Abnormal
0.6 PPM Throat Staphylococcus aureus, Candida spp.
and Escherichia coli, S. aureus, Candida spp.
Gut Klebsiella, Enterobacter, Citrobacter,
Proteus, Morganella, Serratia, Pseudomonas,
Acinetobacter spp., MRSA
High level Epidemic Throat Neisseria meningitidis Abnormal
IPI = 0.9 Micro- Gut Salmonella spp.
1.0 organisms
The intrinsic pathogenicity index (IPI) is the ratio between the number of ICU patients with an
infection due to a particular microorganism and the number of ICU patients who carry the same
particular microorganism. Normal PPMs are carried by healthy individuals in throat and gut.
Individuals with an underlying condition carry both normal and abnormal PPMs in throat and gut
IPI International Prognostic Index; MRSA methicillin-resistant Staphylococcus aureus
conflicting results and tight glucose control can cause dangerous hypoglycaemia,
data underlying this recommendation should be critically evaluated [71].
An RCT by Vlasselaers et al. [72] of intensive insulin therapy to achieve age-
adjusted normal fasting concentrations showed improved short-term outcome of
patients in a paediatric ICU (PICU). However, the Neonatal Insulin Replacement
Therapy in Europe (NIRTURE) study of tight glucose control in neonates and
infants did not conclusively demonstrate the value of insulin therapy in preterm
infants [73].
The practice of tight glucose control is accompanied by an increased incidence
of hypoglycaemia. Hermanides et al. [74] demonstrated that hypoglycaemia
increased the rate of death to 40:1,000 in those who experienced hypoglycaemia
and 17:1,000 for those who were not hypoglycaemic.
The four Types of Definitions of infection according

components of SDD infection to the criterion of carriage
\\ prevented
by SDD
Caused by community and

hospital PPMs carried by the
Parenteral Primary endogenous patients in throat and/or gut on
antibiotic admission to the ICU
Caused by hospitalPPMs not

carried by the patients in throat
Secondary and/or gut on admission to the
PTA endogenous ICU. The PPM is acquired during
ICU stay causing secondary
carriage
To control the
efficacy of PTA
The causative hospital PPM is
not carried in the patients
Hygiene Exogenous digestive tract and is introduced
directly into the sterile organ
Surveillance
To classify infections according
cultures
to the carrier state
To identify a resistance
problem
Fig. 30.1 The full four component protocol of SDD, that aims to control the three different types
of infection that occur on ICU
30.3 Infection-Control Manoeuvres: Antibiotic Interventions
SDD is based on the observation that critical illness changes body flora, promoting
a shift: (1) from normal (Streptococcus pneumoniae in the throat and Escherichia
coli in the gut) towards abnormal [aerobic Gram-negative bacilli (AGNB) and
methicillin-resistant Staphylococcus aureus (MRSA) in throat and gut] carriage
(Table 30.7); (2) from low- to high-grade carriage (gut overgrowth) of both normal
and abnormal flora. Parenterally administered cefotaxime controls gut overgrowth
due to normal bacteria; enterally administered polyenes control gut overgrowth
Table 30.8 Carriage classification of severe infections of lower airways and blood
Infection PPM Timing Frequency Manoeuvre

(%)
1 6 normal; [1 week 55 Parenteral antimicrobials
9 abnormal
2 9 abnormal [1 week 30 Hygiene and enteral
antimicrobials
3 9 abnormal Anytime during ICU 15 Hygiene and topical
1 primary endogenous, 2 secondary endogenous, 3 exogenous PPM potentially pathogenic
microorganism, 6 normal: Streptococcus pneumoniae, Haemophilus influenzae, Moraxella
catarrhalis, Candida albicans, Staphylococcus aureus, Escherichia coli, 9 abnormal: Klebsiella,
Enterobacter, Citrobacter, Proteus, Morganella, Serratia, Acinetobacter, Pseudomonas spp.,
methicillin-resistant Staphylococcus aureus (MRSA), ICU intensive care unit
due to normal Candida spp. Enterally administered polymyxin/tobramycin

(without or with vancomycin) eradicates (if already present) and prevents over-
growth with abnormal bacteria.
Gut overgrowth is the crucial event preceding two classes of infections: primary
and secondary endogenous infections (Table 30.8).
1. Primary endogenous infection is caused by normal or abnormal potential
pathogens present in the patients admission flora. This infection generally
develops within 1 week and is the most frequent type of infection (55%).
2. Secondary endogenous infection is invariably caused by abnormal bacteria not
present in the admission flora but acquired during treatment in the ICU. This
infection generally occurs after 1 week in the ICU (30%).
3. Exogenous infection is caused by abnormal bacteria never carried in the
patients oropharynx and/or gut and may occur anytime during ICU treatment
(15%).
Each of these three types of ICU infection requires different prophylaxis: primary
endogenous can only be controlled by parenterally administered antimicrobials;
secondary endogenous are prevented by enterally administered antimicrobials and
high hygiene standards; exogenous are controlled by topically applied antimicro-
bials and hygiene. These three classes of intervention were first combined by
Stoutenbeek et al., who expanded the prophylaxis to include surveillance cultures,
thus creating the full four-component SDD protocol, the main mechanism of action
being gut overgrowth control [75] (Fig. 30.1).
SDD has been assessed in 11 meta-analyses [7686] covering 60 RCTs
(Table 30.9), showing that SDD reduces pneumonia (72%), septicaemia (37%)
and mortality rates (29%) without resistance emerging. Of the 11 meta-analyses,
lower airway infection was the endpoint in six [76, 77, 79, 82, 84, 86].
All meta-analyses invariably demonstrate a significant reduction in lower airway
Table 30.9 Overview: efficacy of SDD: 60 RCTs and 11 meta-analyses
500
Author No. of Sample Lower airway Bloodstream Multiple organ dysfunction Mortality
RCTs size infection infection syndrome
OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI)
Vandenbroucke-Grauls 6 491 0.12, 0.080.19 NR 0.92, 0.45
[76] 1.84
DAmico [77] 33 5,727 0.35, 0.290.41 NR 0.80, 0.69
0.93
Safdar [78] 4 259 NR NR 0.82, 0.22
2.45
Liberati [79] 36 6,922 0.35, 0.290.41 NR 0.78, 0.68
0.89
Silvestri [80] yeasts 42 6,075 NR 0.89, 0.164.95 NR
Silvestri [81] 51 8,065 NR 0.63, 0.460.87 0.74, 0.61
0.91
Silvestri [82] 54 9,473 0.07, 0.040.13 0.36, 0.220.60 NR
G- 0.52, 0.340.78 1.03, 0.751.41 NR
G+
Silvestri [83] 21 4,902 NR NR 0.71, 0.61
0.82
Liberati [84] 36 6,914 0.28, 0.200.38 NR 0.75, 0.65
0.87
Silvestri [85] 7 1,270 NR NR 0.50, 0.340.74 0.82, 0.51
1.32
Silvestri [86] 12 2,252 0.54, 0.420.69 NR NR
A. J. Petros et al.
infection (OR 0.28, 95% CI 0.200.38). Bloodstream infection was the endpoint
in three meta-analyses [8082] and was significantly reduced (OR 0.63, 95% CI
0.460.87). When assessing bloodstream infection, AGNB septicaemias were
significantly reduced; Gram-positive ones were increased but not significantly
due to the low incidence in the control group (Table 30.9). Multiple organ
dysfunction syndrome (MODS) was the endpoint in one of the most recent
meta-analyses [85], in which the relative reduction of 50% was significant.
Mortality was the endpoint in eight meta-analyses [7679, 81, 8385]. SDD
consistently reduced mortality rates as long as the sample size was large
enough; the sample size was too small in three meta-analyses [76, 78, 85].
References
1. Atkins D, Best D, Briss PA et al, for the GRADE working group (2004) Grading quality of
evidence and strength of recommendations. BMJ 328:14901495
2. Schnemann HJ, Oxman AD, Brozek J et al, for the GRADE working group (2008) Grading
quality of evidence and strength of recommendations for diagnostic tests and strategies. BMJ
336:11061110
3. Jaeschke R, Guyatt GH, Dellinger P et al, for the GRADE working group (2008) Use of
GRADE grid to reach decisions on clinical practice guidelines when consensus is elusive.
BMJ 337:a744
4. Guyatt GH, Oxman AD, Vist GE et al, for the GRADE working group (2008) GRADE: an
emerging consensus on rating quality of evidence and strength of recommendations.
336:924926
5. Casewell M, Philips I (1977) Hands as route of transmission for Klebsiella species. BMJ
2:13151317
6. Massanari RM, Hierholzer J (1984) A cross-over comparison of antiseptic soaps on
nosocomial infection rates in the intensive care units. Am J Infect Control 12:247248
7. Maki DG (1989) The use of antiseptics for handwashing by medical personnel. J Chemother
1(Suppl 1):311
8. Simmons B, Bryant J, Neiman K et al (1990) The role of handwashing in prevention of
endemic intensive care unit infections. Infect Control Hosp Epidemiol 11:589594
9. Doebbeling RN, Stanley G, Sheetz CT et al (1992) Comparative efficacy of alternative
handwashing agents in reducing nosocomial infections in intensive care units. New Engl J
Med 327:8893
10. Webster J, Faogali JL, Cartwright D (1994) Elimination of methicillin-resistant Staphylococcus
aureus from a neonatal intensive care unit after handwashing with tricloson. J Paediatr Child
Health 30:5964
11. Koss WG, Khalili TM, Lemus JF et al (2001) Nosocomial pneumonia is not prevented by
protective contact isolation in the surgical intensive care unit. Am Surg 67:11401144
12. Slota M, Green M, Farley A et al (2001) The role of gown and glove isolation and strict
handwashing in the reduction of nosocomial infection in children with solid organ
transplantation. Crit Care Med 29:405412
big measure with modest effects. J Hosp Infect 59:172179
14. Choi SC, Nelson LD (1992) Kinetic therapy in critically ill patients: combined results based
on meta-analysis. J Crit Care 7:5762
15. Summer WR, Curry P, Haponik EF et al (1989) Continuous mechanical turning of intensive care
unit patients shortens length of stay in some diagnostic-related groups. J Crit Care 4:4553
16. Traver GA, Tyler ML, Hudson LD et al (1995) Continuous oscillation: outcome in critically
ill patients. J Crit Care 10:97103
17. Craven DE, Steger KA (1996) Nosocomial pneumonia in mechanically ventilated patients:
epidemiology and prevention in 1996. Semin Respir Infect 11:3253
18. Drakulovic MB, Torres A, Bauer TT et al (1999) Supine body position as a risk factor for
nosocomial pneumonia in mechanically ventilated patients: a randomised trial. Lancet
354:18511858
19. van Nieuwenhoven CA, van Tiel FH, Vandenbroucke-Grauls C et al (2002) The effect of
semi-recumbent position on development of ventilator-associated pneumonia (VAP).
Intensive Care Med 27(Suppl 2):S285, Abstract 585
20. Keeley L (2007) Reducing the risk of ventilator-acquired pneumonia through head of bed
elevation. Nurs Crit Care 12:287294
21. Silvestri L, Gregori D, van Saene HKF, Belli R (2010) Semirecumbent position to prevent
ventilator-associated pneumonia is not evidence based. J Critical Care 25:152153
22. Mahul P, Auboyer C, Jaspe R et al (1992) Prevention of nosocomial pneumonia in intubated
patients: respective role of mechanical subglottic secretions drainage and stress ulcer
prophylaxis. Intensive Care Med 18:2025
23. Valles J, Artigas A, Rello J et al (1995) Continuous aspiration of subglottic secretions in
preventing ventilator-associated pneumonia. Ann Intern Med 122:179186
24. Kollef MH, Skubas NJ, Sundt TM (1999) A randomised clinical trial of continuous aspiration
of subglottic secretions in cardiac surgery patients. Chest 116:13391346
25. Smulders K, van der Hoeven H, Weers-Pothoff I et al (2002) A randomised clinical trial of
intermittent subglottic secretion drainage in patients receiving mechanical ventilation. Chest
121:858862
26. Metz C, Linde HJ, Gobel L et al (1998) Influence of intermittent subglottic lavage on
subglottic colonization and ventilator associated pneumonia. Clin Intensive Care 9:2024
27. Bo H, He L, Qu J et al (2000) Influence of the subglottic secretion drainage on the morbidity
of ventilator associated pneumonia in mechanically ventilated patients. Zhonghua Jie He He
Hu Xi Za Zhi 23:472474
28. Liu SH, Yan XX, Cao SQ et al (2006) The effect of subglottic secretion drainage on
prevention of ventilator-associated lower airway infection. Zhonghua Jie He He Hu Xi Za Zhi
29:1922
29. Lorente L, Lecuona M, Jimenez A et al (2007) Influence of an endotracheal tube with
Med 176:10791083
30. Zheng RQ, Lin H, Shao J et al (2008) A clinical study of subglottic secretion drainage for
prevention of ventilator associated pneumonia. Zhongguo Wie Zhong Bing Ji Jiu Yi Xue
20:338340
31. Bouza E, Perez MJ, Munoz P et al (2008) Continuous aspiration of subglottic secretions
(CASS) in the prevention of ventilator-associated pneumonia in the postoperative period of
major heart surgery. Chest 134:938946
32. Dezfulian C, Shojania K, Collard HR et al (2005) Subglottic secretion drainage for
preventing ventilator associated pneumonia: a meta-analysis. Am J Med 118:1118
33. Silvestri L, Milanese M, van Saene HKF et al (2008) Impact of subglottic secretion drainage
on ventilator-associated pneumonia and mortality: systematic review of randomized
controlled trials. In: Proceedings of the 21st anesthesia and ICU symposium Alpe Adria.
Udine, 56 Sept 2008, pp 2629
34. De Riso AJII, Ladowski JS, Dillon TA et al (1996) Chlorhexidine gluconate 0.12% oral rinse
reduces the incidence of total nosocomial respiratory infection and nonprophylactic systemic
antibiotic use in patients undergoing heart surgery. Chest 109:15561561
35. Fourrier F, Cau-Pottier E, Boutigny H et al (2000) Effects of dental plaque on antiseptic
decontamination on bacterial colonisation and nosocomial infections in critically ill patients.
36. Houston S, Houghland P, Anderson JJ et al (2002) Effectiveness of 0.12% chlorhexidine

gluconate oral rinse in reducing prevalence of nosocomial pneumonia in patients undergoing
heart surgery. Am J Crit Care 11:567570
37. MacNaughton PD, Bailey J, Donlin N et al (2004) A randomised controlled trial assessing the
efficacy of oral chlorhexidine in ventilated patients. Intensive Care Med 30(Suppl 1):S12
Abstract 029
38. Grap MJ, Munro CL, Elswick RE et al (2004) Duration of action of a single early oral
application of chlorhexidine on oral microbial flora in mechanically ventilated patients: a
pilot study. Heart Lung 33:8391
39. Fourrier F, Dubois D, P Pronnier et al (2005) Effect of gingival and dental plaque antiseptic
decontamination on nosocomial infections acquired in the intensive care unit: a double-blind
placebo-controlled multicenter study. Crit Care Med 33:17281735
40. Segers P, Speckenbrink RGH, Ubbink DT et al (2006) Prevention of nosocomial infection in
cardiac surgery by decontamination of the nasopharynx and oropharynx with chlorhexidine
gluconate. JAMA 296:24602466
41. Koeman M, van der Ven AJAM, Hak E et al (2006) Oral decontamination with chlorhexidine
reduces the incidence of ventilator-associated pneumonia. Am J Respir Crit Care Med
173:13481355
42. Bopp M, Darby M, Loftin KC et al (2006) Effects of daily oral care with 0.12% chlorhexidine
gluconate and a standard care protocol on the development of nosocomial pneumonia in
intubated patients: a pilot study. J Dent Hyg 80:9
43. Tad YD (2007) Efficacy of Chlorhexidine oral decontamination in the prevention of VAP.
Chest 51:498S
44. Tantipong H, Morkchareonpong C, Jaiyindee S et al (2008) Randomised controlled trial and
meta-analysis of oral decontamination with 2% chlorhexidine solution for the prevention of
ventilator-associated pneumonia. Infect Control Hosp Epidemiol 29:131136
45. Panchabhai TS, Dangayach NS, Krishnan A et al (2009) Oropharyngeal cleansing with 0.2%
chlorhexidine for prevention of nosocomial pneumonia in critically ill patients. Chest
135:11501156
46. Scannapieco FA, Yu J, Raghavendran K et al (2009) A randomised trial of chlorhexidine
gluconate on oral bacterial pathogens in mechanically ventilated patients. Crit Care 13:R117
47. Munro CL, Grap MJ, Jones DJ et al (2009) Chlorhexidine, toothbrushing, and ventilator-
associated pneumonia in critically ill adults. Am J Crit Care 18:428438
48. Bellissimo-Rodrigues F, Bellissimo-Rodrigues WT et al (2009) Effectiveness of oral rinse
with chlorhexidine in preventing nosocomial respiratory tract infections among intensive care
unit patients. Infect Control Hosp Epidemiol 30:952958
49. Cabov T, Macan D, Husedzinovic I et al (2010) The impact of oral health and 0.2%
chlorhexidine oral gel on the prevalence of nosocomial infections in surgical intensive-care
patients: a randomized placebo-controlled study. Wien Klin Wochenschr 122:397404
50. Pineda LA, Saliba RG, El Solh AA (2006) Effects of oral decontamination with chlorhexidine
on the incidence of nosocomial pneumonia: a meta-analysis. Crit Care 10:R35
51. Chlebicki MP, Safdar N (2007) Topical chlorhexidine for prevention of ventilator-associated
pneumonia: a meta-analysis. Crit Care Med 35:595602
52. Chan EY, Ruest A, OMeade M et al (2007) Oral decontamination for prevention of
pneumonia in mechanically ventilated adults: systematic review and meta-analysis. BMJ
334:889893
53. Kola A, Gastmeier P (2007) Efficacy of oral chlorhexidine in preventing lower respiratory
tract infections: meta-analysis of randomised controlled trials. J Hosp Infect 66:207216
54. van Saene HKF, Silvestri L, de la Cal MA et al (2009) The emperors new clothes: the fairy
tale continues. J Crit Care 24:149152
55. Carvajal C, Pobo A, Diaz E et al (2010) Oral hygiene with chlorhexidine on the prevention of
ventilator-associated pneumonia in intubated patients: a systematic review of randomized
clinical trials. Med Clin (Barc) 135:491497
56. Beale RJ, Bryg DJ, Bihari DJ (1999) Immunonutrition in the critically ill: a systematic review
of clinical outcome. Crit Care Med 27:27992805
57. Heyland DK, Novak F, Drover JW et al (2001) Should immunonutrition become routine in
critically ill patients? a systematic review of the evidence. JAMA 286:944953
58. Cronin L, Cook DJ, Carlet J et al (1995) Corticosteroid treatment for sepsis: a critical
appraisal and meta-analysis of the literature. Crit Care Med 23:14301439
59. Lefering R, Neugebauer EAM (1995) Steroid controversy in sepsis and septic shock: a meta-
analysis. Crit Care Med 23:12941303
60. Bollaert PE, Charpentier C, Levy B et al (1998) Reversal of late septic shock with
supraphysiologic doses of hydrocortisone. Crit Care Med 26:645650
61. Briegel J, Forst H, Haller M et al (1999) Stress doses of hydrocortisone reverse hyperdynamic
septic shock: a prospective, randomised, double-blind, single-center study. Crit Care Med
27:723732
62. Annane D, Sebille V, Charpentier C et al (2002) Effect of treatment with low doses of
hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA
288:862871
63. Annane D, Bellissant E, Bollaert PE et al (2009) Corticosteroids in the treatment of severe
sepsis and septic shock in adults: a systematic review. JAMA 301:23622375
64. Annane D, Cariou A, Maxime V et al, for the COIITSS study investigators (2010) Corticosteroid
treatment and intensive insulin therapy for septic shock in adults: a randomized controlled trial.
JAMA 303:341348
65. Alejandria MM, Lansang MA, Dans LF, Mantaring JBV (2000) Intravenous immunoglobulin
for treating sepsis and septic shock (Cochrane review). In: The Cochrane library, issue 3.
Update Software, Oxford
activated protein C for severe sepsis. New Engl J Med 344:699709
67. Marshall J (2000) Clinical trials of mediator-directed therapy in sepsis: what have we
learned? Intensive Care Med 26:575583
68. Zeni F, Freeman B, Natanson C (1997) Anti-inflammatory therapies to treat sepsis and septic
shock: a reassessment. Crit Care Med 25:10951100
69. van den Berghe G, Wouters P, Weekers F et al (2001) Intensive insulin therapy in the
70. Van den Berghe G, Wilmer A, Hermans G (2006) Intensive insulin therapy in the medical
ICU. N Engl J Med 354:449461
71. Wiener RS, Wiener DC, Larson RJ (2008) Benefits and risks of tight glucose control in
critically ill adults: a meta-analysis. JAMA 300:933944
72. Vlasselaers D, Milants I, Desmet L et al (2009) Intensive insulin therapy for patients in
paediatric intensive care: a prospective, randomised controlled study. Lancet 373:547556
73. Beardsall K, Vanhaesebrouck S, Ogilvy-Stuart AL et al (2007) A randomised controlled trial
of early insulin therapy in very low birth weight infants, NIRTURE (neonatal insulin
replacement therapy in Europe). BMC Pediatr 7:29
74. Hermanides J, Bosman RJ, Vriesendorp TM (2010) Hypoglycemia is associated with
intensive care unit mortality. Crit Care Med 38:14301434
75. Stoutenbeek CP, van Saene HKF, Miranda DR et al (1984) The effect of selective
Lancet 338:859862
77. DAmico R, Pifferi S, Leonetti C et al, on behalf of the study investigators (1998)
Effectiveness of antibiotic prophylaxis in critically ill adult patients: systematic review of
randomised controlled trials. BMJ 316:12751285
infection in patients undergoing liver transplantation: a systematic review and meta-analysis.
Liver Transpl 10:817827
tract infections and mortality in adults receiving intensive care (Cochrane review). In: The
Cochrane library, issue 1. Wiley, Chichester
of randomized controlled trials. Intensive Care Med 31:898910
digestive tract reduces bloodstream infections and mortality in critically ill patients: a
systematic review of randomized controlled trials. J Hosp Infect 65:187203
82. Silvestri L, van Saene HKF, Casarin AL et al (2008) Impact of selective decontamination of
the digestive tract on carriage and infection due to Gram-negative and Gram-positive
bacteria: systematic review of randomized controlled trials. Anaesths Intens Care 36:324338
digestive decontamination regimen. J Crit Care 24:474.e7474.e14
84. Liberati A, DAmico R, Pifferi S (2009) Antibiotic prophylaxis to reduce respiratory tract
infections and mortality in adults receiving intensive care. Cochrane Database Syst Rev
4:CD000022
86. Silvestri L, van Saene HKF, Zandstra DF (2010) Selective digestive decontamination reduces
ventilator-associated tracheobronchitis. Respir Med 104:19531955
Index
A 134, 137, 155, 158, 163, 165, 166, 168,

Abnormal bacteria, 3, 18, 51, 55, 116118, 169, 171174, 176, 179, 183, 187, 191,
172, 412, 431, 475 193, 197, 200, 205, 207, 209, 213215,
Abnormal colonization, 457 220, 230, 231, 240, 241, 245, 251, 258,
Acute pancreatitis, 25, 200, 242244, 256, 259, 261, 265, 268, 270, 273, 277, 279,
268, 269, 379, 396, 442 281, 284, 285, 288293, 295, 300, 307,
AGNB, 3, 6, 1821, 24, 31, 40, 42, 44, 5057, 309, 342, 354358, 362, 364, 367, 368,
63, 67, 69, 70, 7274, 77, 78, 80, 81, 370, 375, 376, 386, 398, 404,
83, 86, 116125, 127, 128, 163, 165, 405, 417, 421, 427429, 434, 441, 444,
168, 170, 171, 178, 179, 181, 186, 187, 448
193, 197, 273, 274, 277, 279281, 284, Antifungal therapy, 97, 99, 111, 113, 115, 185,
305, 376, 412, 413, 416418, 422, 429, 251, 364
431, 432, 434, 435, 474, 477 Antimicrobial resistance, 8, 33, 35, 51, 61, 67,
AIDS, 22, 104, 321, 334340, 342344, 89, 90, 123, 124, 133, 134, 147,
346348, 350353 172174, 193, 197, 290, 428, 430,
Airway, 4, 5, 7, 9, 13, 1519, 2125, 28, 432436, 438, 440, 442, 444, 453
4042, 41, 42, 48, 52, 5659, 123, 145, Antimicrobials, 9, 24, 51, 56, 62, 116128,
153, 160, 168, 169, 171, 173, 176, 130, 132, 134, 136, 156, 162166,
193195, 204206, 208, 210212, 214, 168172, 186, 188, 191194, 197, 198,
216, 250, 258, 274277, 281, 285, 320, 201, 279, 282, 284, 287290, 292, 295,
356, 357, 358, 362365, 367, 381, 313, 355, 356, 358, 359, 361, 363, 364,
383386, 411, 412, 430, 432, 433, 439, 367, 404, 406, 409, 411, 412, 414, 416,
464, 468, 469, 475478 418, 419, 423, 427, 430432, 434, 436,
Albumin, 64, 76, 81, 253, 254, 261, 337, 392, 443, 475
393, 399 Arginine, 251, 391, 396398, 400, 471
Amphotericin b, 9, 24, 42, 9498, 100, 101, Azole-resistant candida spp, 93, 104, 435
102, 112115, 119, 120, 122, 123, 125,
132, 165167, 169, 171, 175, 185, 191,
193, 194, 198, 213, 251, 280, 283, 284, B
292, 296, 298300, 302, 310312, 347, Bacteremia, 3, 10, 13, 2527, 34, 36, 37, 47,
348, 356, 358360, 412, 413, 415, 48, 72, 77, 81, 129, 190, 196198, 202,
417420, 424, 428, 431, 432 206, 215, 216, 222, 225, 228230,
Antacids, 209, 212, 213, 372, 401, 407, 412 232234, 238, 250, 289, 304, 308, 340,
Antibiotic therapy, 34, 37, 41, 6971, 89, 342, 346, 352, 360, 369, 395, 439, 448,
99, 112, 115, 128, 145, 158, 171, 173, 451453, 455, 458, 459
176, 215, 234, 236, 240, 250, 259, 262, Bacterial colonization, 18, 24, 134, 148, 153,
265, 267, 287, 288, 289, 291, 293, 294, 203, 210214, 285, 383, 385, 445
295, 305, 341, 367, 448, 459 Bacterial pneumonia, 78, 89, 254, 340, 352
Antibiotics, 35, 18, 21, 25, 42, 43, 49, 59, Bacterial translocation, 13, 23, 26, 58, 199,
6173, 75, 77, 79, 81, 83, 8587, 259, 267, 268, 302, 307, 374376, 378,
8991, 98, 119, 123, 128130, 133, 391, 399, 439, 441
DOI: 10.1007/978-88-470-1601-9, Springer-Verlag Italia 2012
508 Index
B (cont.) 190, 201, 273, 277, 284, 312, 369, 412,

Bladder, 4, 11, 1517, 21, 22, 28, 42, 52, 101, 431, 432, 443, 478
157, 158, 166, 168, 262264, 266, 276, Combination therapy, 81, 97, 98, 295, 355,
277, 279, 360, 361, 363, 364 357359, 361, 363
Blood and body fluid, 315, 325, 328 Community-acquired infection, 129, 232, 234,
Blood stream, 193, 274, 278 297, 450
Bloodstream infection, 2, 3, 12, 14, 37, 99, Compliance, 121, 123, 136, 140, 141, 143,
115, 124, 147, 149, 151, 155, 160, 194, 155, 193, 383, 418420
195, 202, 218222, 224, 226, 228, 230, Costs, 113, 144, 147, 172, 173, 206, 208, 213,
232234, 274, 277, 281, 285, 289, 299, 224, 233, 261, 271, 290, 291, 293, 300,
300, 342, 343, 362, 363, 367, 368, 377, 334, 380, 393, 394, 398, 423, 424, 438,
392, 399, 400, 426, 430, 432, 433, 440, 439, 446, 448450, 452, 454460, 468,
451, 455, 458, 459, 476, 477, 481 472
Bone marrow transplant, 110, 296, 302, 307, Critical illness, 13, 18, 23, 26, 49, 50, 51, 53,
309, 313, 332 63, 64, 66, 89, 176, 285, 292, 368, 370,
Bundles, 209, 230, 231 376, 377, 389391, 396,
398400, 405, 409, 429, 446,
447, 451, 474
C Critically ill patient, 4, 8, 9, 12, 13, 15, 16, 19,
Carriage, 25, 7, 9, 1517, 1921, 2325, 28, 21, 23, 24, 39, 52, 55, 5759, 61, 64,
29, 31, 32, 35, 39, 41, 4347, 4959, 65, 72, 90, 111, 117, 124, 129, 132,
71, 116122, 125128, 132, 134, 135, 159, 160, 168, 172, 175, 186, 190,
145, 148, 163, 169173, 179, 182, 183, 193, 200, 202205, 212, 214, 217,
186, 188192, 194, 197, 198, 202, 273, 219, 221, 227, 230, 234, 237, 239,
275, 281, 285, 299, 354356, 360, 364, 250, 290, 291, 293295, 355, 356, 362,
365, 368, 370372, 388, 411, 412, 416, 364, 367, 368, 371, 374, 375, 378, 384,
417, 426, 429431, 434436, 438440, 387, 389, 390, 392, 394403, 405,
444, 474, 475, 481 407409, 426, 428, 432, 434, 435, 440,
Carrier state, 3, 4, 9, 13, 16, 18, 24, 28, 36, 39, 442, 443, 445, 451454, 457459, 472,
4144, 46, 47, 4953, 55, 5759, 477481
116118, 121, 134, 168, 169, 171, 172, Critically ill, 4, 8, 9, 12, 13, 15, 16, 19, 21,
176, 191193, 195, 275, 285, 412, 429, 23, 24, 26, 37, 39, 40, 51, 52, 55,
430, 432, 435, 439, 474 59, 61, 64, 65, 72, 89, 90, 111,
Catheter, 2, 3, 11, 25, 28, 72, 111, 112, 136, 115117, 119, 121, 123, 124,
142144, 149, 151, 155161, 168, 180, 129, 132134, 159161, 166,
185, 205, 207, 214, 219, 220, 222225, 168, 172, 175, 184, 186, 190,
233, 234, 239, 242, 246, 248, 264266, 193, 199205, 212, 214, 217,
269, 271, 274, 279, 289, 296, 299, 304, 219, 221, 224, 227, 230234,
306, 308, 342, 356, 360364, 366, 368, 237, 239, 249251, 266, 267,
382, 392, 395, 396, 398, 448, 451, 453, 270, 281, 285, 286, 290, 291, 293295,
455, 457459 333, 355, 356, 362,
Cholangitis, 356 364, 367, 368, 370379, 384,
Classification, 12, 27, 29, 3137, 3949, 56, 387, 389, 390, 400410, 426,
57, 59, 78, 82, 83, 90, 121, 137, 138, 428, 431, 432, 434, 435,
164, 166, 227, 232, 234236, 243, 259, 439443, 445, 446, 451454, 457459,
262, 268, 270, 275, 388, 430, 439, 473, 471, 472, 477481
475
Clinical experience, 9395, 97, 129, 403
Cohorting, 142, 189, 316, 317, 319, 323 D
Colistin, 66, 67, 72, 74, 86, 91, 130, 132, 150, De novo development, 428, 434
152, 165, 292, 358, 413415, 417424 De-escalation, 71, 72, 90, 290, 293, 295
colonic ileus, 368, 372, 377, 409 Definitions, 15, 7, 9, 1115, 24, 28, 33, 41,
Colonisation, 24, 25, 38, 56, 112, 115, 146, 47, 48, 52, 192, 204, 216, 232, 233,
147, 173, 174, 176, 180, 182184, 189, 259, 266, 474
509 Index
Diagnostic samples, 6, 16, 17, 43, 5153, 57, Exogenous, 3, 59, 11, 21, 23, 25, 29, 42, 43,
121, 186, 354, 357, 362, 366, 432 4547, 53, 5557, 59, 121, 123, 125,
Drainage, 11, 157159, 209, 220, 229, 236, 127, 170, 173, 176, 179, 183, 186, 188,
239, 241, 242, 246248, 251, 266, 191193, 205, 211, 275, 277, 356, 363,
268270, 310, 359361, 364, 382, 364, 367, 382, 385, 393, 394, 396, 411,
384387, 464, 466, 468, 469, 478 412, 429, 430, 438, 463, 474, 475
Extended infusion, 65
E
Echinocandins, 92, 94, 97, 99, 107111, F
114, 348 Faecaloral transmission
Efficacy, 9, 24, 64, 6668, 81, 89, 9496, Formulations, 113, 140, 302, 312, 397, 411,
111, 121, 123, 126, 134, 137, 415, 416, 418, 425
150, 151, 158, 162, 168170, Four-quadrant method, 53, 54
172, 173, 186, 191196, 199,
201, 209, 214, 215, 234, 245,
281, 284, 288, 292, 302, 312, G
320, 355, 357, 381, 383, 387, Ganciclovir, 296, 298, 303, 309, 312, 313,
396, 406, 411, 414, 419, 426, 343, 344
430, 432, 433, 435, 441, 443, Glutamine, 391, 396398, 400
445, 450, 455, 463, 474, 476, Grade, 50, 51, 55, 93, 94, 97, 111, 113, 180,
477, 479, 480 212, 236, 246, 312, 338, 367, 375,
Empirical antibiotic treatment, 230, 288 429431, 437, 438, 461464, 472, 474,
Empirical, 4, 5, 65, 72, 81, 92, 94, 99, 108, 477
112, 113, 115, 172174, 230, 240, 251, Gut overgrowth, 8, 13, 23, 26, 50, 51, 53, 55,
283, 284, 288291, 293, 294, 297, 306, 58, 117, 124, 126, 129, 184, 187, 190,
309, 354356, 359, 361, 367, 368 285, 413, 428431, 434, 435, 439, 444,
Endogenous, 3, 57, 9, 11, 17, 21, 29, 39, 474, 475
4247, 51, 53, 5557, 116, 118, 121,
123, 124, 127, 133, 168170, 173, 179,
186, 191193, 200, 205, 211, 275277, H
297, 300, 356, 362, 365, 370, 378, 385, Haart, 335, 336, 338, 340, 343,
406, 408, 411413, 429, 430, 435, 442, 346351
463, 472, 474, 475 Hand hygiene, 55, 126, 127, 137, 138, 140,
Endotracheal tube, 21, 23, 72, 136, 144, 141, 146, 155, 158, 191, 319, 323, 324,
150, 151, 153, 159, 160, 205, 332, 366, 380, 463, 465
210, 211, 214, 216, 277, 356, Hand washing, 6, 11, 111, 140, 225,
364, 366, 381, 384, 385, 387, 316, 317, 324, 325, 329,
388, 457, 468, 469, 478 365, 366, 463
Enteral feeding, 151, 208, 212, 270, 359, 386, Helicobacter pylori, 402, 404,
392, 394, 396, 399, 400, 402, 405407, 406, 409
457, 471 High-level pathogen, 6, 191, 275
Enteral nutrition, 18, 19, 23, 26, 129, 151, 156, HIV infection, 327, 334, 335, 338346, 348,
209, 210, 217, 245, 268270, 277, 285, 351353
375, 381, 382, 385, 395397, 399, 400, Hospital, 4, 6, 7, 13, 18, 19, 27, 29, 3134, 36,
409, 471 37, 3942, 47, 48, 57, 58, 69, 72, 83,
Eradication, 9, 24, 66, 117, 119, 121, 123, 90, 111, 115, 116, 125, 128, 129, 134,
171, 182, 184, 190, 197, 300, 136, 137, 139, 140, 143, 144, 146, 147,
324, 326, 359, 364, 365, 404, 149, 152, 154, 157, 164, 168, 172, 174,
418, 430 176, 179185, 188190, 197, 198, 204,
Evidence-based practice, 137 206, 208, 215, 216, 218220, 222225,
Exogenous infection, 57, 9, 42, 4547, 53, 227229, 231234, 236, 237, 258260,
5557, 123, 127, 173, 191, 193, 275, 263, 265, 270272, 276, 280, 287, 288,
277, 364, 385, 412, 430, 438, 463, 475 290, 291, 294, 296, 297, 299, 303, 312,
510 Index
317324, 328, 331, 332, 335, 337, 342, 253, 266268, 270, 279, 355, 356, 359,
343, 352, 355, 361, 377, 380, 383, 387, 362, 368, 372, 393
389393, 395, 397, 398, 411, 414, 416, Intrinsic pathogenicity index, 32, 48, 275,
418, 423427, 434, 435, 439, 444447, 285, 473
452, 455, 456, 458461, 463, 471, 474 Isolation, 6, 11, 12, 17, 28, 30, 31, 54, 56, 57,
112, 137, 139, 141, 142, 146, 147, 184,
220, 221, 233, 279, 287, 316, 317, 319,
I 323325, 328330, 346, 348, 350, 362,
IA abscess, 235, 253, 258 365, 463, 464, 477
Immunonutrients, 390, 391
Immunonutrition, 391, 399, 406, 409, 470,
471, 480 K
Immunosuppression, 8, 18, 19, 33, 42, 44, 51, Kaposis sarcoma, 334, 335, 345,
55, 56, 129, 260, 289, 297, 299, 303, 348350, 353
306, 309, 310, 322, 338, 342, 343, 346,
364, 390, 432
Import, 1, 2, 4, 18, 2022, 24, 25, 33, 37, 39, L
4245, 53, 55, 58, 63, 6769, 71, 77, Leadership, 145
85, 92, 98, 111, 116119, 125, 126, Lowest resistance potential, 163
128, 135, 140, 155, 168, 172, 179, 181, Low-level pathogen, 44, 128, 191, 241, 275,
184, 186, 193, 197, 198, 204, 205, 208, 362, 436
212, 215, 219, 221, 224227, 231, 234,
238, 245, 251, 254, 258, 260, 261, 264,
266, 273, 274, 278, 285, 299, 310, 314, M
319, 321, 323, 325, 330, 340, 342, 344, Macconkey agar, 54
354, 355, 364, 366, 370, 375, 376, Mediastinitis, 248251, 269, 270
382384, 386, 389392, 395, 397, 398, Meningitis, 4, 63, 68, 101, 104, 278, 329, 347,
401407, 413, 431, 432, 434, 438, 444, 355, 356, 361, 471
447, 448, 451, 453, 462, 463, 468, 472 MIC, 2, 13, 15, 18, 2045, 4773, 7585, 87,
Indigenous flora, 7, 17, 20, 21, 31, 32, 34, 118, 89102, 104, 106136, 138142,
119, 163, 164, 275, 366, 413, 144147, 151158, 160, 162176, 179,
431, 473 180, 182195, 197, 203, 205, 207, 211,
Infection control, 1, 7, 15, 24, 27, 39, 47, 48, 213216, 218226, 228230, 232238,
50, 57, 59, 61, 92, 113, 116, 128, 130, 240, 241, 243245, 249, 251254,
135, 136, 138, 140, 142, 144, 146149, 256267, 270, 272295, 299301, 303,
159, 161, 162, 177, 179, 180, 183, 305, 307, 309, 311323, 328335,
187189, 191, 204, 216, 218, 232, 235, 337342, 346, 348, 351, 352, 354380,
270, 272, 277, 284, 287, 296, 314, 316, 383387, 389, 391, 392, 394396, 398,
319, 320, 323, 327, 329, 331334, 354, 401, 403409, 411419, 423, 426436,
356, 370, 380, 381, 383, 388, 389, 401, 438440, 442445, 449460, 463, 464,
405, 411, 426428, 430, 446, 456, 468, 471, 473, 475481
461, 463 Morbidity, 7, 28, 72, 115, 118, 123, 128,
Infection, 129, 31, 3353, 5563, 65, 66, 69, 133, 147, 154, 172, 173, 179,
71, 72, 7578, 8083, 86, 89, 9092, 198, 208, 209, 218, 224, 242,
94101, 104, 108, 112116, 118, 250, 252, 257, 258, 263, 270,
121152, 154163, 165, 166, 168481 274, 281, 285, 288, 289, 293,
Inflammation, 2, 5, 6, 8, 12, 1618, 24, 28, 42, 294, 314, 321, 330, 340, 342,
51, 52, 55, 129, 165, 166, 168, 169, 347, 367, 376, 380, 384, 389391,
238, 241, 244, 256, 259, 277, 279, 339, 394396, 398, 401, 405, 412,
340, 344, 359, 362364, 366, 367, 370, 425, 428, 432, 440, 446448,
371, 375, 392, 402, 403, 405, 413, 432 450, 454, 456458, 460, 463465, 468,
Inoculum effect, 62, 63 472, 478
Intra-abdominal, 6, 89, 96, 97, 220, 222, 223, Mortality, 7, 13, 18, 28, 3335, 37, 40, 47, 53,
225, 231, 235, 236, 238, 239, 241, 242, 58, 60, 65, 72, 96, 99, 111, 112, 115,
511 Index
118, 121, 123125, 128, 129, 132, 133, O

144, 147, 152, 154, 172, 173, 175, 179, Open treatment, 240
180, 184, 185, 187, 188, 195, 196, Oral chlorhexidine, 387, 479
198203, 206, 208, 209, 213, 215217, Oral decontamination, 132, 152, 159, 199,
224, 226, 229, 230, 232235, 238, 240, 201, 384, 387, 425, 440, 443, 469, 479
242252, 257, 258, 263, 270, 273, 274, Outbreaks, 7, 35, 116, 118, 127, 142, 162,
276, 281, 282, 285, 286, 288295, 297, 177190, 273, 284, 314, 315, 321325,
300303, 305, 306, 308, 312, 314, 318, 334, 344
320, 321, 329, 330, 334, 336348, 351, Overgrowth, 3, 8, 9, 13, 1517, 19, 21, 2326,
355, 357, 361, 366369, 372374, 44, 47, 5056, 58, 71, 116118,
376378, 380, 383392, 394399, 120122, 124129, 163, 169, 170, 184,
401, 402, 405, 407, 408, 412, 425, 186, 187, 190, 193, 285, 360, 365, 366,
426, 428, 432435, 437444, 446448, 370, 372, 374, 376, 392, 400, 413,
450460, 463, 464, 466472, 475, 478, 428432, 434436, 439, 444, 474, 475
480, 481
MRSA, 3, 6, 8, 18, 19, 21, 34, 4244, 5058,
65, 73, 74, 7679, 82, 84, 8688, 91, P
116122, 124127, 132, 134, 135, 137, paediatric intensive care unit, 273, 284
142, 144147, 163, 167171, 178180, Parenteral nutrition, 19, 23, 26, 129, 156, 245,
186188, 193, 198, 205208, 215, 222, 268, 269, 277, 285, 375, 396, 399, 400,
228230, 250, 276, 277, 280, 289, 290, 471
299, 306, 307, 358, 364366, 368, 412, Peritoneal lavage, 239, 240, 267
413, 429431, 435, 444, 452, 456, 465, Peritonitis, 6, 25, 81, 96, 108, 129, 235242,
473475 266, 267, 279, 304, 308, 358, 359, 367,
Multiorgan dysfunction, 389, 390 368, 395
Multiple organ dysfunction syndrome Pharmacist, 162, 411, 412, 414, 425
(MODS), 12, 131, 195, 196, 202, 215, Pharmacodynamics, 89, 114, 130, 288, 291, 295
370, 379, 390, 401, 410, 432, 477, 433, Pharmacokinetics, 89, 91, 114, 129, 133, 291,
476 295
Mycobacterial infections, 350, 353 Pharmacological properties, 75, 76, 79, 80, 85,
93, 95, 97
PK/PD, 61, 89, 292, 293
N Pneumocystis jiroveci pneumonia, 351
Neonatal intensive care unit, 36, 135, 175, 187, Pneumonia, 1, 3, 68, 12, 17, 18, 2426, 28,
190, 272, 283, 285, 286, 317, 332, 427, 30, 3234, 36, 37, 3942, 44, 45,
477 4749, 56, 59, 62, 69, 72, 7683, 85,
Neostigmine, 357, 359, 368, 372, 377, 406, 86, 89, 90, 116, 117, 124, 130132,
409 144, 147, 149154, 159, 160, 167, 168,
Neuroendocrine system, 390 171, 174, 176, 181, 185, 189, 194,
Nonabsorbable antimicrobials, 9, 24, 118, 123, 196199, 201218, 221, 225, 226, 228,
169, 364, 411, 412, 430 230, 231, 237, 252, 254, 256, 276278,
Nonantibiotic management, 381 281285, 289291, 294, 295, 298, 300,
Normal bacteria, 3, 18, 51, 55, 116118, 128, 301, 303309, 318, 320322, 334,
172, 412, 431, 475 336338, 340343, 346, 347, 350353,
Nosocomial infections, 36, 37, 40, 47, 59, 86, 355357, 361, 362, 364, 367, 368, 372,
135, 144, 147, 159, 174, 199, 206, 208, 373, 379384, 386388, 393, 403406,
216, 232, 259, 271, 285, 314, 324, 368, 408, 411, 412, 425, 426, 429, 430, 434,
390, 399, 441, 444, 446, 453, 457459, 439, 440, 442444, 447, 449, 452, 456,
465, 477479 458460, 463469, 471, 473475,
Nutrition, 18, 19, 23, 26, 119, 129, 147, 151, 477479
156, 205, 209, 210, 212, 217, 220, 241, Polymyxin e, 9, 42, 86, 119, 122, 125, 130,
245, 260, 261, 268270, 277, 285, 301, 165, 167, 169, 170, 191, 193, 213, 280,
340, 375, 381, 382, 385, 386, 389400, 356, 358, 360, 412, 413, 415, 427, 431,
406, 409, 470, 471, 480 432, 435
512 Index
P (cont.) 332, 335, 336, 343, 355, 356,

Posaconazole, 9294, 103, 104106, 113, 348 359, 360, 362, 366, 367, 368, 370372,
Postoperative complications, 261, 296, 389, 376378, 391, 393,
393, 399 397, 401403, 406, 408, 409,
Potential pathogen, 3, 7, 9, 11, 17, 23, 33, 50, 437, 441, 443, 445, 451, 454,
53, 55, 119, 126, 128, 165, 167273, 455, 457460, 463, 471, 472, 480
275277, 283, 284, 301, 355357, 364, Septic shock, 2, 9, 10, 12, 13, 64, 69, 89, 112,
429, 430, 434, 435, 438, 468 223, 226, 229232, 234, 250, 263, 267,
Pre-emptive, 92, 112, 320 287291, 294, 295, 337, 355, 359, 360,
Pregnancy, 68, 253, 256, 271, 322 362, 366, 367, 369, 371374, 377, 378,
Prevention, 12, 25, 36, 40, 47, 48, 58, 59, 104, 437, 445, 451, 458, 471, 480
113, 117, 122, 129, 130, 132, 136, 137, Severe infections, 56, 72, 145, 362, 430, 475
142, 145147, 153, 155, 157161, 175, Shock, 2, 9, 10, 12, 13, 26, 37, 64, 69, 70, 89,
176, 179, 186, 199201, 204, 206, 208, 112, 199, 201, 215, 223, 226, 229232,
209, 214, 219, 224, 233, 268, 270, 271, 234, 241, 243, 244, 249251, 263, 267,
279, 280, 284, 286, 296, 309, 310, 312, 287291, 294, 295, 322, 328, 337, 341,
331, 333, 350, 352, 354, 367, 368, 372, 355, 359, 360, 362, 366, 367, 369,
376, 380, 381, 383, 386388, 403408, 371374, 376378, 383, 392, 401, 402,
423, 425, 426, 439, 440444, 449, 454, 408, 413, 437, 441, 443, 445, 451, 458,
456, 460, 477, 478, 479 468, 471, 472, 480
Probiotics, 90, 199, 200, 214, 217, 301, 375, Silver-coated endotracheal tube, 150, 151,
376, 379, 384, 387, 398, 441, 442 160, 385
Prokinetics, 359, 376 Solid organ transplant, 296, 298, 300, 301,
Prolonged hospitalization, 389 303, 307, 312, 313
Prophylaxis, 9, 24, 59, 60, 92, 93, 95, 96, 104, Source control, 240, 267, 363, 456
108, 111114, 127, 130, 132134, 151, Spectrum of activity, 7577, 79, 80, 82, 83,
156, 158, 161, 166, 168171, 174, 175, 8588, 93, 95, 96, 100, 103, 107, 166,
185, 191, 199202, 208210, 212, 213, 167, 261
217, 259, 261, 262, 268, 270, 272, Stress-ulcer prophylaxis, 208, 212, 213, 381,
279281, 285, 286, 298303, 309, 311, 382
312, 314, 320, 322, 325327, 333, 334, Stress-ulcer-related bleeding (surb), 401, 406
338, 346, 349, 351, 352, 368, 370, 372, Subglottic secretion drainage, 159, 209, 382,
376, 377, 379, 381, 382, 387, 401, 402, 384387, 464, 469, 478
405410, 425, 426, 440, 442, 444, 458, Surgical site, 125, 258, 260, 270
475, 478, 480, 481 Surveillance cultures, 3, 9, 31, 43, 47, 52, 53,
5558, 121, 126, 145, 147, 179, 184,
186188, 190192, 274, 275, 354, 360,
S 362, 364, 365, 367, 411, 428, 430, 439,
Secondary endogenous infection, 5, 29, 474, 475
42, 43, 45, 169, 186, 193, 412, Surveillance samples, 3, 8, 9, 1517, 41, 5053,
429, 475 5557, 59, 121, 123, 169, 172, 173, 193,
Selective decontamination of the digestive 277, 300, 354, 365, 431
tract (sdd), 3, 134, 145, 168, 191, 280, Surveillance, 2, 3, 8, 9, 12, 1517, 29, 31, 33,
296, 403, 411, 415, 417419, 422, 423, 39, 41, 43, 47, 48, 5059, 92, 98, 99,
431, 433, 437, 450, 464 115, 121, 123, 126, 134, 143145, 147,
Semirecumbent position, 159, 160, 210, 212, 155, 169, 170, 172, 173, 176, 179, 180,
381, 385, 386, 388, 466, 468, 478 184188, 190, 191193, 208, 216, 219,
Sepsis, 1, 2, 9, 10, 1214, 23, 26, 37, 64, 69, 224, 232, 245, 274, 275, 277, 285, 297,
72, 89, 112, 115, 129, 138, 175, 179, 300, 328, 330, 333, 342, 354, 355,
184, 199, 201, 223, 226, 229234, 237, 360362, 364, 365, 367, 380, 393, 411,
238, 240, 241, 246248, 250, 251, 258, 413, 419, 428, 430, 431, 439, 457, 474,
266, 475
267, 270, 274, 278, 285, 287289, 291, Systemic inflammatory response, 2, 9, 10, 13,
294, 295, 307, 322, 324, 223, 237, 278, 366, 371, 389, 391
513 Index
Systemic, 2, 8, 9, 10, 13, 18, 20, 21, 23, 25, 33, U

41, 49, 51, 58, 59, 61, 63, 65, 67, 69, Urinary tract, 9, 11, 18, 22, 25, 28, 81, 82, 144,
7173, 75, 77, 79, 81, 83, 85, 87, 89, 149, 151, 157, 159, 161, 183, 219, 222,
91, 92, 94, 96, 98, 100102, 104, 106, 225, 226, 229, 235, 237, 239, 241, 243,
108, 110119, 119, 124, 129, 131, 133, 245, 247, 249, 251, 253, 255, 257, 259,
156158, 166, 168, 169, 175, 176, 193, 261, 262, 263, 265, 267, 269, 271, 279,
197, 199, 213, 223, 226, 229, 236238, 299, 304, 308, 355, 360, 393, 451, 453
241, 243245, 251253, 256258, 263, Urosepsis, 175, 356, 360
267, 278, 281, 284, 285, 299301, 307,
309, 311, 352, 357360, 362364,
366371, 373, 375, 389, 391, 392, 394, V
395, 405, 406, 413, 423, 432, 439, 440, Vaccination, 298, 299, 309, 319
478 Vancomycin, 8, 13, 35, 37, 5759, 62, 65, 74,
8487, 91, 117129, 131, 132, 134,
135, 144, 163, 165167, 169171, 173,
T 175, 178180, 187189, 193, 198, 203,
Tobramycin, 9, 24, 42, 67, 69, 70, 74, 80, 81, 222, 250, 251, 261, 280, 292, 293, 301,
85, 119124, 126, 130132, 164, 165, 358, 360, 361, 364366, 368, 413, 427,
167, 169, 171, 191, 193, 197, 198, 213, 431, 432, 435, 436, 439, 444456, 460,
261, 280, 300, 356, 475
358, 360, 412, 413, 415, 417419, Vasodilators, 372, 376, 403, 405407
421424, 428, 431, 432, 434, Ventilator circuit, 209211, 382, 384, 386,
435, 475 456, 457
Toxicity, 65, 66, 78, 81, 85, 86, 95, 101, 102, Ventilator-associated pneumonia, 1, 24, 33,
110, 130, 168, 169, 289, 292, 299, 310, 36, 37, 41, 48, 132, 144, 159, 160, 176,
318323, 355, 358, 472 201, 205, 208210, 214, 216, 217, 294,
Transmission, 8, 11, 35, 45, 47, 51, 53, 57, 59, 295, 298, 368, 380, 382, 384, 386388,
116, 125128, 139, 142, 146, 147, 151, 403, 405, 411, 442, 449, 451, 452, 456,
179, 180, 182, 185, 186, 191, 193, 209, 458460, 468, 478, 479
272, 273, 315, 316, 318, 323, 325327, Viral, 8, 173, 177, 187, 253, 257, 274, 303,
329331, 364, 366, 383, 404, 434, 438, 304, 307, 312, 314, 318, 320, 321, 323,
456, 477 324, 326, 328, 330337, 342, 343, 345,
Treatment, 3, 8, 9, 13, 25, 31, 3337, 41, 43, 348, 351353, 362, 363, 364
51, 53, 55, 56, 59, 71, 72, 78, 8991, VRE, 8, 35, 48, 56, 8688, 128, 135, 140, 143,
9397, 108, 112115, 122, 123, 126, 144, 163, 178180, 186, 187, 198, 202,
127, 129, 130, 132, 135, 149, 152, 154, 222, 272, 279, 425, 429, 430, 436, 437,
171, 172, 174, 175, 176, 182187, 191, 441, 444, 456, 463, 466468, 470, 474,
200, 206, 208, 215, 216, 226, 229, 230, 475
232, 234238, 240, 241, 246248, 250,
251, 257259, 261, 271, 282284,
287295, 297, 298, 301303, 306, 307, W
309, 310, 312, 314, 321, 322, 326329, Wound, 7, 9, 12, 16, 17, 28, 41, 52, 138, 168,
332, 340, 341, 343348, 351355, 170172, 215, 219, 222, 227, 235, 237,
357361, 363, 368, 370, 373, 376378, 239, 241, 243, 245, 247, 249, 251253,
380, 386, 391, 405, 408, 411, 412, 255, 257259, 261263, 265, 267,
414, 420, 430, 431, 434, 435, 442, 445, 269271, 276, 277, 279, 299, 300, 304,
447, 449, 454, 456462, 471, 472, 475, 308, 327, 330, 354, 356, 357, 360,
480 362364, 393, 394, 412

Infection Control in The Intensive Care

Uploaded by

Copyright:

Available Formats

Infection Control in The Intensive Care

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Infection Control in The Intensive Care

Uploaded by

Copyright:

Available Formats

Infection Control in the Intensive

Foreword by Julian Bion

ISBN 978-88-470-1600-2 e-ISBN 978-88-470-1601-9

Library of Congress Control Number: 2011929635

Springer-Verlag Italia 1998, 2005, 2012

Printed on acid-free paper

Springer is part of Springer Science+Business Media (www.springer.com)

In 1847, Ignatius Semmelweiss friend and colleague Jakob Kolletschka died of

November 2011 Julian Bion

A week-long postgraduate course was organised in Trieste, Italy, in 1994. This

November 2011 H. K. F. van Saene

Part I Essentials in Clinical Microbiology

1 Glossary of Terms and Definitions . . . . . . . . . . . . . . . . . . . . . . . 3

2 Carriage, Colonization and Infection. . . . . . . . . . . . . . . . . . . . . . 17

3 Classification of Microorganisms According

4 Classification of ICU Infections . . . . . . . . . . . . . . . . . . . . . . . . . . 41

5 Gut Microbiology: Surveillance Samples for Detecting

8 Enteral Antimicrobials . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123

Part III Infection Control

9 Evidence-Based Infection Control in the Intensive Care Unit . . . . 145

10 Device Policies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 159

11 Antibiotic Policies in the Intensive Care Unit . . . . . . . . . . . . . . . 173

12 Outbreaks of Infection in the ICU: Whats up at the Beginning

13 Preventing Infection Using Selective Decontamination

Part IV Infections on ICU

14 Lower Airway Infection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219

15 Bloodstream Infection in the ICU Patient . . . . . . . . . . . . . . . . . . 233

16 Infections of Peritoneum, Mediastinum, Pleura, Wounds,

17 Infection in the NICU and PICU. . . . . . . . . . . . . . . . . . . . . . . . . 289

18 Early Adequate Antibiotic Therapy. . . . . . . . . . . . . . . . . . . . . . . 305

19 ICU Patients Following Transplantation . . . . . . . . . . . . . . . . . . . 315

20 Clinical Virology in NICU, PICU and AICU . . . . . . . . . . . . . . . . 333

21 AIDS Patients in the ICU . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 353

22 Therapy of Infection in the ICU . . . . . . . . . . . . . . . . . . . . . . . . . 373

Part V Special Topics

23 The Gut in the Critically Ill: Central Organ in Abnormal

24 Nonantibiotic Measures to Control

25 Impact of Nutritional Route on Infections: Parenteral

26 Gut Mucosal Protection in the Critically Ill Patient:

27 Selective Decontamination of the Digestive Tract:

28 Antimicrobial Resistance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 451

29 ICU-Acquired Infection: Mortality, Morbidity, and Costs . . . . . . 469

30 Evidence-Based Medicine in ICU . . . . . . . . . . . . . . . . . . . . . . . . 485

A. Abella Department of Intensive Care Medicine, Hospital Universitario de

C. J. Collins Clinical Microbiology, Trinity College Dublin, Dublin, Leinster,

J. C. Marshall Department of Surgery and Intensive Care, St Michaels Hospital,

M. Snchez Garcia PhD, MD Servicio de Medicina Intensiva, Hospital Clnico

Defining terms is important to avoid ambiguity, particularly in the era of global

1.2 Terms and Definitions

A patient is considered to have acquired a microorganism when a single surveil-

1.2.2 Bloodstream Infection

1.2.2.1 Primary Bloodstream Infection

1.2.2.2 Secondary Bloodstream Infection

1.2.2.3 Catheter-Related Bloodstream Infection

1.2.3 Carriage/Carrier State

The same strain of microorganism is isolated from two or more surveillance

1.2.3.1 Normal Carrier State

1.2.3.2 Abnormal Carrier State

1.2.3.3 Primary Carriage