Review

Respiratory infections in patients undergoing mechanical

ventilation
Jordi Rello, Thiago Lisboa, Despoina Koulenti
Lancet Respir Med 2014;
2: 76474
Published Online
August 21, 2014
http://dx.doi.org/10.1016/
S2213-2600(14)70171-7
Critical Care Department,
Hospital Universitari Vall
dHebron, Barcelona, Spain,
Centro de Investigacin
Biomdica en Red
Enfermedades Respiratorias,
Barcelona, Spain, and
Universitat Autonoma de
Barcelona, Barcelona, Spain
(Prof J Rello PhD); Critical Care
Department and Infection
Control Committee, Programa
de Ps-Graduao
Pneumologia, Hospital de
Clinicas de Porto Alegre,
Universidade Federal do Rio
Grande do Sul, Porto Alegre,
Brazil, and Rede Institucional
de Pesquisa e Inovao em
Medicina Intensiva, Complexo
Hospitalar Santa Casa, Porto
Alegre, Brazil (T Lisboa MD);
and 2nd Critical Care
Department, Attikon
University Hospital, Athens,
Greece, and Burns Trauma and
Critical Care Research Centre,
The University of Queensland,
Brisbane, QLD, Australia
(D Koulenti PhD)
Correspondence to:
Dr Jordi Rello, Critical Care
Department, Hospital
Universitari Vall dHebron, PS
Vall dHebron 119, Anexe A - 5a
planta, 08035 Barcelona, Spain
jrello@crips.es

764

Lower respiratory tract infections in mechanically ventilated patients are a frequent cause of antibiotic treatment in
intensive-care units. These infections present as severe sepsis or septic shock with respiratory dysfunction in
intubated patients. Purulent respiratory secretions are needed for diagnosis, but distinguishing between pneumonia
and tracheobronchitis is not easy. Both presentations are associated with longlasting mechanical ventilation and
extended intensive-care unit stay, providing a rationale for antibiotic treatment initiation. Dierentiation of
colonisers from true pathogens is dicult, and microbiological data show Staphylococcus aureus and Pseudomonas
aeruginosa to be of great concern because of clinical outcomes and therapeutic challenges. Key management issues
include identication of the pathogen, choice of initial empirical antibiotic, and decisions with regard to the
resolution pattern.

Introduction
In the Extended Prevalence of Infection in Intensive
Care II (EPIC II) study,1 which looked at 1265 intensivecare units (ICUs) in 75 countries, 51% of adults
admitted to ICUs were infected, and the respiratory
tract was the focus of infection in 64% of cases. Airway
infection in intubated patients is the main reason for
antibiotic prescription in medical ICUs. Because no
gold standard exists for the diagnosis of respiratory
tract infections in intubated patients,2 prescription of
antibiotics for patients with purulent respiratory
secretions is common clinical practice in the ICU
setting. The panel shows the clinical challenges in
management of respiratory tract infections in ventilated
patients with a hypothetical clinical scenario. In this
Review we discuss ventilator-associated respiratory
infections (VARIs) in adult patients, with an emphasis
on diagnosis, microbiological causes, and management.

showing the limitations of present denitions and

diagnostic criteria.
The clinical pulmonary infection score was created to
predict the pretest probability of pneumonia.7 It
combines information on body temperature, volume
and appearance of tracheal secretions, chest radiograph
ndings, white blood cell count, oxygenation, and
tracheal aspirate culture.7 Many studies have used the
clinical pulmonary infection score to identify patients
with pneumonia because it allows an objective
assessment of clinical variables for pneumonia
diagnosis.8,9 Unfortunately, despite the use of objective
data, such as white blood cell count and oxygenation,
the clinical pulmonary infection score also includes
variables that are either subjective or retrospective,
such as chest radiograph ndings, secretion
appearance, and microbiological data, which might

Clinical challenge

Key messages

Patients with lower respiratory tract infection usually

present with progressive hypoxaemia and fever, which
is unlike the sudden onset of rigor and temperature
rise associated with bloodstream infections. In our
view, the low sensitivity and specicity of present
diagnostic criteria is the most important diculty in
the assessment and diagnosis of mechanically
ventilated patients with suspected lower respiratory
tract infections.3 Denition of a clinical syndrome on
the basis of the clinical presentation of VARI is thus a
challenge for clinicians. Because the diagnostic criteria
include many subjective components (eg, chest
radiographs, assessment of respiratory secretions, or
auscultation), interobserver variability in the identication of ventilator-associated pneumonia (VAP) is
high.4,5 In a prospective survey6 of a nationally
representative group of US hospitals, participants were
asked to classify standardised vignettes of possible
cases of VAP as having pneumonia or not having
pneumonia. Agreement at the hospital level on
classication of cases as VAP or not was nearly random,

Dierentiation of ventilator-associated pneumonia and

tracheobronchitis on the sole basis of clinical signs can be
a dicult task at the bedside in view of chest radiograph
interpretation variability and other subjective clinical
variables
High rates of pathogen resistance have emerged in patients
with ventilator-associated pneumonia, making therapeutic
choice a challenge, especially for Gram-negative pathogens
Improvements in pathogen detection, use of biomarkers
to guide treatment, and new antibiotic delivery devices
are promising strategies to optimise ventilator-associated
respiratory infection treatment
Clinical response assessed 3 days after ventilatorassociated respiratory infection onset is a key issue in
optimisation of therapy
Severity of illness and underlying conditions are the most
important variables inuencing survival.
Immunocompromise and delay in adequate antibiotic
therapy, such as management of hypoxaemia and shock,
results in increased use of health-care resources

www.thelancet.com/respiratory Vol 2 September 2014

Review

Panel: Hypothetical clinical scenario of a mechanically

ventilated patient developing pneumonia
A 52-year-old man with underlying chronic obstructive
pulmonary disease underwent colectomy for colon cancer.
He presented a complex postoperative course, receiving
ceftriaxone and metronidazole. After 1 week on mechanical
ventilation, the patient was not successfully weaned and a
tracheostomy was done. Respiratory secretions became
purulent 2 days after tracheostomy and the chest radiograph
was consistent with basal atelectasis. The patient developed
worsening hypoxaemia the following day, with a temperature
increase to 385C. The white blood cell count was 14 000 cells
per L and the C-reactive protein concentration was
18163 mol/L (normal values <6054 mol/L). The PaO2:FiO2
ratio remained greater than 250, but the need for vasopressors
was increased. Tracheal secretion specimens were collected for
direct stains and quantitative culture. The patient had a VAP
PIRO score of 3, suggestive of high risk of intensive care unit
death. Empirical treatment was started with a carbapenem
plus amikacin. After 3 days, the patient had defervescence, so
positive end-expiratory pressure was decreased, and C reactive
protein decreased by 15%. Quantitative cultures of respiratory
specimens and blood cultures were both positive for
Pseudomonas aeruginosa. The strain was susceptible to
piperacillin and tazobactam, carbapenems, uoroquinolones,
colistin, and aminoglycosides; antibiotic treatment was then
de-escalated to piperacillintazobactam. 10 days later, the
patient was successfully weaned.

compromise its use in some subgroups (eg, patients

with acute respiratory distress syndrome) because of
high interobserver variability in the application of the
score. We think that individual diagnostic decisions
should not be made on the basis of scores. Diagnostic
scores are helpful for provision of probabilities for
comparisons between groups, but do not seem appropriate for assessment of the probability of pneumonia
in individual patients.
The dierentiation of VAP and ventilator-associated
tracheobronchitis (VAT) on the sole basis of clinical
signs can be dicult at the bedside. The cuto points of
bacterial load for colonisation, tracheobronchitis, and
pneumonia in mechanically ventilated patients have
not been conclusively de ned, and vasopressor requirements and complications (eg, eect of respiratory
infection on oxygenation) need to be assessed. The US
Centers for Disease Control and Preventions diagnosis
of VAT is based on absence of chest radiograph
in ltrates and the presence of signs consistent with
respiratory inammation, and at least one microbiological criterion.4 However, the absence of objectivity
and inherent variability in interpretation of chest
radiographs in mechanically ventilated patients makes
decisions on their basis hard to make. Dallas and
colleagues10 reported a median onset of VAT 75 days
www.thelancet.com/respiratory Vol 2 September 2014

A
Colonisation

VAT

VAP

B
VAT
Colonisation
VAP

Early VAP

Colonisation

VAP

VAT

VARI

Figure 1: Pathogenetic correlations between VAT and VAP

Colonisation leads to (A) VAT that leads to VAP; (B) to either VAT or VAP; or (C)
to VARI with some degree of overlap between VAT and VAP.
VAP=ventilator-associated pneumonia. VARI=ventilator-associated respiratory
infection. VAT=ventilator-associated tracheobronchitis.

after intubation and initiation of mechanical ventilation

compared with 5 days for VAP, suggesting that VAT is
not necessarily a precursor of VAP, although a high
proportion of patients VAT evolved to VAP. Antibiotic
use might be an important factor aecting whether
VAT progresses to VAP. Figure 1 proposes pathophysiological aspects of the correlation between these
two diseases.
A pilot translational study11 comparing gene
expression proles in VAP and VAT identied
5595 genes dierentially expressed in the preinfection
period. A substantial depression of the complement
system, cyclic adenosine monophosphate, and calcium
signalling pathways during the preinfection phase were
noted in the VAP group (when compared with VAT).

Epidemiology
The epidemiology of respiratory infections in the ICU
varies, dependent on whether the patient is mechanically
ventilated with a tracheostomy or an endotracheal tube.
The role of the biolm is important in patients who are
tracheostomised.12,13 Aspiration constitutes the main
pathophysiological event. Avoidance of an articial airway
is the best method of prevention of respiratory infection.
By contrast with community-acquired pneumonia,
respiratory infections in mechanically ventilated patients
are heterogeneous. The dierence between a patient who
has undergone intra-abdominal or cardiac surgery, or
trauma, and a medical patient, is hard to establish.
Although the attributable mortality of VAP is
controversial, this condition prolongs mechanical
ventilation and length of stay in the ICU.14 The overall
765

Review

attributable mortality of VAP was estimated to be 13%

in a meta-analysis.15 Admission diagnosis, age,
causative pathogen, and adequacy of treatment all
aect mortality. High attributable mortality rates were
reported in patients undergoing surgery and those
with intermediate (Acute Physiology and Chronic
Health Evaluation II score 2030) illness severity at
admission, whereas almost no attributable mortality
was reported in trauma or medical patients, or those
with low or high illness severity scores. The ndings
from this meta-analysis therefore show substantial
variability in incidence and attributable mortality
between studies.1622 Data from low income and
developing countries suggest that incidence and
attributable mortality might be higher in these
countries than that reported in the literature.1622
A particular challenge is the development of pneumonia in the postoperative period of lung transplantation
because its presentation can overlap with the signs of
acute rejection, but require the opposite therapeutic
approach (an increase rather than decrease in
immunosuppressors).23 Riera and colleagues23 reported
that episodes of tracheobronchitis doubled the risk of
episodes of pneumonia in this subset of patients.
Pneumonia was associated with increased in-hospital
death rates (429% for patients with pneumonia vs 115%
for those without, p=001), whereas tracheobronchitis
alone was not associated with increased mortality (140%
with tracheobronchitis vs 147% without, p=09).23 In a
prospective observational study24 of 2436 patients from
27 ICUs in nine European countries, mortality of VAP
was 426% in patients without trauma versus 172% in
patients with trauma. Additionally, Melsen and
colleagues15 reported that VAP development decreased
the daily probability of discharge from the ICU by 26%,
suggesting that the disorder lasts throughout the stay.
The overall daily risk of discharge after ventilatorassociated pneumonia was 074 (068080).
Additionally, variability in VAP incidence might be
due to the absence of a diagnostic gold standard. The
standard denition used to measure VAP incidence is
based on several non-specic clinical signs, with
microbiological criteria included to improve specicity,
but could be severely restricted by the absence of
sensitivity and specicity of present criteria.14
Dierences between surveillance strategies and the
clinical denition of VAP are crucial to understand
such variability. These dierences aect appropriate
assessment of prevention studies, because low rates
might be associated with dierent criteria in subjective
aspects of diagnosis. An attempt to design a
straightforward, objective surveillance denition for
ventilator-associated complications shifted the focus of
surveillance from pneumonia alone to complications
of mechanical ventilation.25 The eect on clinical
practice of adoption of such new criteria is not yet
available, however.
766

We believe that prevention trials and recommendations should no longer focus on VAP rates only.
Measures associated with improved outcomes
(especially in patients mechanically ventilated for a
short time) and reduced costs should also be
implemented. Investigators of a Spanish multicentre
cohort study26 reported that full VAP prevention care
bundle compliance was associated with an incidence
risk ratio of 078 (95% CI 015099) and a reduction
of both median ICU length of stay from 10 days to
6 days and duration of ventilation from 8 days to 4 days.
Key interventions were oral care, cu pressure
maintenance, hand hygiene before articial airway
manipulation, and strategies to avoid hypersedation.
Prolongation of ICU stay is associated with increased
(preventable) health-care costs27 and emphasises the
interest in the prioritisation of prevention measures,
which have shown potential cost reductions (rather
than rate reductions only).

Pathogens
Fewer than ten organisms are implicated in most VARI
cases and a substantial proportion are polymicrobial
infections.28 Increasingly, respiratory pathogens are
Gram-negative bacteria. The EU-VAP study28 identied
Staphylococcus aureus and Pseudomonas aeruginosa as
the pathogens isolated most often in patients with VAP.
Organisms such as methicillin-susceptible S aureus,
Haemophilus inuenza, and Streptococcus pneumoniae
are common causes of early-onset VAP in trauma
patients, but the infection improves quickly (within
3 days) when adequate treatment is promptly started.
Authors of a secondary analysis of the EU-VAP study
reported that elderly ICU patients with VAP had
increased rates of Enterobacteriaceae compared with
younger age groups.29 The table details the top three
causative pathogens of VAP reported in seven
studies28,3035 published during the past decade. Figure 2
shows the median onset of VAP for each pathogen.

Antibiotic resistance
ESKAPE pathogens
Nosocomial infections are often caused by ESKAPE
pathogens (Enterococcus faecium, S aureus, Klebsiella
pneumoniae, Acinetobacter baumannii, P aeruginosa, and
Enterobacter species).36 Sandiumenge and colleagues37
reported S aureus, P aeruginosa, and A baumannii to be
the three main causes of infection in patients in ICUs.
Enteroccus spp and Candida spp, on the other hand,
should be interpreted as oral contaminants. The risk of
multidrug-resistant pathogens causing VAP is mainly
determined by comorbidity and previous exposure to
more than two antibiotics.37 The increased mortality of
VAP due to multidrug resistance as compared with
drug susceptible pathogens is attributable to more
severe comorbidity and the presence of organ failure.38
Resistant ESKAPE VAP mortality was twice that of
www.thelancet.com/respiratory Vol 2 September 2014

Review

patients with VAP from other causes in one study

(relative risk 225, 95% CI 167948).37 We will focus
our discussion on methicillin-resistant S aureus
(MRSA) and P aeruginosa because they represent a
major eect on clinical outcomes and cost, and a
therapeutic
challenge.
For
Enterobacteriaceae,
emergence of extended-spectrum -lactamases or
K pneumoniae carbapenemases is concerning, but these
resistant pathogens are more frequently implicated in
extrapulmonary infections. Finally, we will briey
discuss A baumannii, which is endemic in some ICUs.

appropriateness of initial antibiotic treatment. A

prospective, double-blind, controlled, multicentre trial
of adults admitted to hospital with nosocomial MRSA
pneumonia reported that although 60 day mortality of
linezolid was similar to that of dose-optimised
vancomycin, linezolid clinical response was
Study type

Findings

Koulenti et al (2009)28*

Prospective multicentre study; 27 ICUs in

nine European countries; 465 cases of
VAP

Overall: Staphylococcus aureus

326% (MSSA 180%, MRSA 146%)
Pseudomonas aeruginosa 228%
Acinetobacter baumannii 202%
Early onset (<5 days):
MSSA 276%
P aeruginosa 179%
MRSA 124%
Late onset (5 days):
A baumannii 265%
P aeruginosa 261%
MRSA 161 %

Esperatti et al (2010)30

Prospective single-centre Spanish study;

164 VAP cases

P aeruginosa 240%
MSSA 140%
MRSA 90%

Kollef et al (2005)31

Retrospective multicentre study; 59 US

hospitals; 499 culture-positive VAP cases

MSSA 285%
P aeruginosa 212%
MRSA 190%

Lee MS et al (2013)32

Prospective multicentre study; 31 US

community hospitals; 247 VAP cases

MRSA 245%
Pseudomonas spp 140%
Klebsiella spp 119%

Canadian Critical Care Trials

Group (2006)33

Prospective, multicentre study; 28 ICUs

in Canada and the USA; 739 VAP cases

S aureus 172%
Haemophilus inuenzae 134%
Enterobacter spp 93%

Rosenthal V et al (2012)34

Prospective, multicentre study; 44 ICUs

from 31 cities of 14 developing countries
of four continents

Pseudomonas spp 260%

Acinetobacter spp 250%
S aureus 190%

Bekaert M et al (2011)35

Longitudinal prospective French

multicentre OUTCOMEREA database;
16 ICUs; 685 patients with
microbiologically conrmed VAP

P aeruginosa 262%
MSSA 97%
A baumannii 82%

MRSA

www.thelancet.com/respiratory Vol 2 September 2014

ICU=intensive care unit. MRSA=methicillin-resistant Staphylococcus aureus. MSSA=methicillin-sensitive Staphylococcus

aureus. VAP=ventilator-associated pneumonia. *Percentages refer to microbiologically conrmed VAP cases (n=356).
Percentages refer to all VAP cases. Percentages refer to the total number of isolated microorganisms (n=868).

hil
op

ot

ro

ph

to
en
St

ine

ia

nii

ba

om

as

cte

rb

au

alt

gin

an

os

RS
A
M

Ac

Ps

eu
d

om

En

te

on

ro

as

ba

ae
ru

cte

ria

ce

SS

ae

14
12
10
8
6
4
2
0

St
pn rept
or em eum oco
c
M op
or hi onia cus
ax lu
ell s in e
a c fl
at uen
ar z
rh ae
ali
s

Median onset (days)

Table: Top three pathogens of VAP reported in seven studies published during the past decade

Ha

Illness severity does not seem to aect the cause of

VAP, therefore severity per se should not be an
indication for broad spectrum antibiotics; initial
empirical antibiotic treatment should be guided by
multidrug resistance risk factors.39 However, patients
with high severity scores and septic shock at onset of
pneumonia have signicantly reduced survival and an
increased systemic inammatory response.39 In relation
to VAP caused by MRSA, factors aecting decisions on
anti-MRSA empirical prescription dier from risk
factors.40 Risk factors for MRSA VAP development
include previous antibiotic exposure, extended
admission to hospital, underlying chronic obstructive
pulmonary disease (COPD), and steroid use.41
Conversely, young age (<25 years) and neurological
impairment (eg, head trauma) have been associated
with
methicillin-susceptible
strains.
Baseline
prevalence of MRSA VAP in a specic ICU should be
considered carefully before the initial empirical
antibiotic treatment is chosen because of suspected
pneumonia. Data from the International Nosocomial
Infection Control Consortium42 showed a MRSA
prevalence of more than 70% in ICU patients with
S. aureus VAP in developing countries. Bacteraemic
VAP is independently associated with MRSA and
mortality.43 Moreover, mortality is higher for MRSA
than susceptible infections in the ICU; a secondary
analysis of one study44 reported that MRSA was
independently associated with an almost 50% higher
likelihood of hospital death compared with methicillinsusceptible infection.
Regarding MRSA VAP treatment, controversy
surrounds the use of glycopeptides (eg, vancomycin)
versus linezolid, fuelled by vancomycins minimum
inhibitory concentration increment in the past few
years, poor alveolar penetration, and potential adverse
events; linezolids cost; and the poor clinical resolution
of MRSA VAP.45 MRSA VAP traditionally has poor
resolution and half of the patients need mechanical
ventilation for more than 3 weeks after pneumonia
onset.45 Vidaur and colleagues45 reported that the
resolution of MRSA VAP was associated with a need for
increased respiratory support compared with VAP
caused by other pathogens, irrespective of the

Figure 2: Median onset (days after intubation) of ventilator-associated

pneumonia for the most common pathogens
Data for 465 episodes of ventilator-associated pneumonia from the EU-VAP/CAP
study28 database (only the rst episodes of pneumonia were included in the
analysis). MRSA=methicillin-resistant Staphylococcus aureus. MSSA=methicillinsensitive Staphylococcus aureus.

767

Review

signicantly higher.46 Moreover, acute kidney injury

has been associated with vancomycin use in patients
with a glomerular ltration rate of more than 50 mL
per min (188% of patients who recieved vancomycin vs
56% of those who received linezolid).46 This study,
however, has some potential biases, such as an unequal
distribution of comorbidity and bacteraemia between
groups,47 and no dierences between outcomes of
vancomycin and linezolid were noted in patients on
vasopressors. We suggest use of linezolid in patients
with immunosuppression, concomitant administration
of nephrotoxic drugs for severe sepsis or in elderly
patients, and provision of glycopeptides such as
vancomycin in the absence of these factors. Further
research is needed into adjunctive treatment, such as
neutralising virulence factors (alginate, pantovalentin
leukocidine, or alfatoxin) to improve outcomes and
minimise injury.

P aeruginosa
An increased frequency of multidrug-resistant
P aeruginosa strains has been reported in the past
decade.48 In the ICU, these infections are a key issue for
infection management, especially VAP.49 In developing
countries, resistance rates to several antibiotic classes are
high (eg, uoroquinolones 46%, piperacillintazobactam
40%, and carbapenems 43%).42 Patients at risk of
P aeruginosa infection should receive combination
treatment with two drugs at pneumonia onset because of
the probability of wrong initial treatment, which has
been associated with signicantly increased mortality.50
However, when susceptible, one drug has similar
outcomes to that of two, and simplication of the
treatment regimen can be implemented after
susceptibility is established.50 For empirical treatment
choice, prescribers should bear in mind factors associated
with isolation of multidrug-resistant P aeruginosa. These
factors include admission to hospital from chronic care
facilities,51 old age, diabetes, long-term treatment in
hospital,5153 use of invasive devices,51,54,55 recent surgery,53
extended ICU stay, extended ventilation periods, and
higher illness severity scores.55,56,38
Candida spp airway colonisation could promote
pneumonia development, especially when caused by
P aeruginosa, perhaps linked to the biolm environment
in the articial airway.5760 In episodes with clinical
suspicion of VAP, Candida spp airway colonisation has
been associated with increased mortality risk (odds ratio
[OR] 172).58 Yeasts have been reported to be an
independent risk factor for identication of multidrugresistant microorganisms (OR 179). Whether Candida
airway colonisation should be a variable aecting
selection of VAP empirical treatment is not clear.58
Crossinfection could also contribute to emergence of
multidrug-resistant P aeruginosa strains.55,61 Previous
antibiotic exposure plays a key part in acquisition of
multidrug-resistant strains.6163 Aminoglycoside ex768

posure has been identied as a risk factor;51 however, the

importance of antipseudomonal cephalosporins,51,64
uoroquinolones (levooxacin in preference to
ciprooxacin),53,65,66 and carbapenems54,67 has also been
reported. Imipenem potentially has the greatest potential
for multidrug-resistant strain selection,65 whereas
ertapenem does not induce carbapenem resistance to
Pseudomonas strains.62 Previous treatment with
antipseudomonal penicillin or -lactamase inhibitor
combinations does not seem to increase isolation of
multidrug-resistant organisms.66 However, antibiotic
treatment is not the only factor associated with the
acquisition of resistant organisms: ndings from a study
of Pseudomonas strains highly resistant to meropenem68
did not identify previous antibiotic exposure as a risk
factor for multidrug resistance.
The implications of multidrug-resistant P aeruginosa
infection, especially respiratory infection, remain controversial. In 2006, such infections were reported to be
associated with increased mortality (OR 44) and
hospital stay (hazard ratio 20) compared with controls.51
However, other reports suggest that multidrug
resistance does not directly aect outcomes. Whether
worse outcomes are due to resistant strains or not is
dicult to establish because of many confounding
factors.69,70 The presence of organ dysfunction (OR 104),56
multiple comorbidities, and inappropriate empirical
antibiotic treatment increased mortality (relative
risk 159), ICU and mechanical ventilation periods (to
4 days), and length of hospital stay (to 13 days).48
Piperacillin resistance has not been shown to aect
outcomes in episodes of VAP.69,70
Research has emphasised the contribution of
virulence factors in P aeruginosa pneumonia. Quorum
sensing and biolm formation have been studied.71
Type III secretion systems (TTSS) encoded by
P aeruginosa might play a key role. The needle-like
TTSS mechanism allows bacteria to inject toxins
directly into the cytoplasm of host cells. Toxins are
therefore not exposed extracellularly, evading direct
recognition by the host immune system.72 These
ndings have fuelled the hypothesis that the inability to
eradicate Pseudomonas spp in pneumonia might be due
to the TTSS. In a study of pneumonia caused by
P aeruginosa, despite appropriate antimicrobial
treatment, more than 50% of the strains expressing at
least one type of TTSS protein were recovered 1 week
after treatment initiation.73 By contrast, eradication was
documented in all episodes caused by strains that did
not express TTSS proteins. In a retrospective cohort74 of
143 patients with P aeruginosa VAP, O6 and O11 were
the most prevalent strains. Mortality tended to be worse
with O1 or O11 serotypes, and better with O2 or O6
serotypes. Clones exhibiting ExoU, one of the toxins
secreted by the TTSS, were frequently serotyped as O11,
by contrast with serotype O6 strains, which were often
associated with a negative ExoU serotype.75,76 These
www.thelancet.com/respiratory Vol 2 September 2014

Review

ndings show the importance of immunomodulatory

adjunctive treatment in the future management of
severe pneumonia.77 Elective P aeruginosa vaccination of
patients at high risk of late onset pneumonia is another
potential method of prevention that warrants further
research.

A baumannii
A baumannii is a non-fermentative Gram-negative
bacilli that has caused large outbreaks in contaminated
ICUs. The pathogen has dierent risk factors to
P aeruginosa and no virulence factors. Independent risk
factors for A baumannii pneumonia in intubated
patients include acute respiratory distress syndrome,
head trauma, large volume pulmonary aspiration,78
presence of tracheostomy,79 and extended ICU stay.80 An
extended antibiotic course is a frequent risk factor for
A baumannii infections. The presence of a resistant
phenotype that often implicates carbapenems is of great
concern; rates of carbapenem resistance of up to 66%
have been reported.42 For carbapenem-resistant strains,
high doses of nebulised colistin have been associated
with good resolution and short periods of hospital
admission.81 High doses of colistin can purportedly be
delivered by nebulisation without substantial systemic
exposure because, even in the presence of severe lung
infection, colistin does not easily cross the alveolar
capillary membrane.81

Management
VAP
Administration of antibiotics should not be delayed in
pneumonia management because inadequate treatment
increases mortality and, in survivors, increases healthcare costs. The use of broad-spectrum antibiotics as
initial empirical treatment has been advocated;
however, concerns remain about resistance. The most
eective strategy against development of resistance
should be prompt and unequivocal killing of the
microbes, thereby defeating resistance before it starts.
Additionally, the de-escalation strategy that allows the
use of broad-spectrum antibiotics as initial empirical
treatment, maximisation of the odds of appropriate
antibiotic treatment associated with early de-escalation,
use of a more strict spectrum coverage after pathogen
identication, and minimisation of exposure and risk
of resistance emergence have shown benet on clinical
outcomes in ventilated patients.82,83 The so-called right
rst time idea and short duration of treatment
whenever possible represents the two-step strategy for
VAP management.
The best diagnostic method for VAP (invasive vs noninvasive sampling techniques) is unknown. From their
meta-analysis, Shorr and colleagues84 concluded that the
use of invasive strategies did not aect mortality, but did
aect antibiotic use, leading to modications in the
antibiotic regimen in more than half of patients. By
www.thelancet.com/respiratory Vol 2 September 2014

contrast, ndings from the Canadian Critical Care Trials

Group study33 showed no dierence in clinical and
microbiological outcomes between an invasive and noninvasive diagnostic approach, suggesting endotracheal
aspirate might be as eective as bronchoalveolar lavage
for causal diagnosis in VAP. Nevertheless, use of
quantitative culture techniques might help to assess
probability of colonisation or infection, although no
unequivocal cuto has been identied. Authors of
another meta-analysis85 reported a non-signicant
association between an increased number of antibiotic
changes with invasive diagnostic testing, but no
dierences in duration of mechanical ventilation, length
of ICU stay, or mortality.
The optimum length of antibiotic treatment of VAP
has not been conclusively established, but in the EU-VAP
study,40 standard of care was an 8 day antibiotic regimen
according to present recommendations.86 This study
listed the causes of antibiotic prescription for intubated
patients in Europe, including reasons for anti-MRSA
prescription. Other studies are now needed to assess
recent treatment strategies and their eect on emerging
drug resistance. Careful antibiotic monitoring is
recommended in the ICU setting, but which monitoring
practices are benecial is not known. The eect of
antimicrobial stewardship on the emergence of
resistance and on patients outcomes is still to be
established, but individualisation of management, avoidance of homogeneous selective pressure, and use of the
entire range of antimicrobials seem to be useful strategies
to escape the adverse eects of emerging resistance.87

VAT
Use of antibiotics in VAT is more controversial than it
is in VAP. Only 24% of prescribers routinely gave
antibiotics for VAT in a survey and 24% stated a
preference for an antibiotic course of less than 7 days,
although only 26% thought that VAT should not be
treated with antibiotics.88 Nseir and colleagues89
reported reduced mortality rates and more days without
mechanical ventilation when VAT was treated with
intravenous antibiotics (45% of the cohort were patients
with COPD). However, several methodological
limitations should be considered as this study was
unblinded and underpowered, with high risk of crosscontamination, some unbalanced distribution in
severity of diseases, and an absence of adequate risk
analysis. Palmer and colleagues90 reported a decrease of
VAP development rates, faster weaning from the
ventilator, reduced use of systemic antibiotics, and
reduction of bacterial resistance when nebulised
antibiotics were given for VAT compared with no
antibiotics in a double-blind, randomised, placebocontrolled study. Dallas and colleagues10 reported that
patients diagnosed with VAT had much the same
outcomes to those with VAP, suggesting that
antimicrobial treatment might also be appropriate for
769

Review

Therapeutic management of VARIs

Empirical antibiotic therapy*

NO

Baseline multidrug resistance?

Mechanical ventilation >5 days?
COPD?
IC, vasopressors, or severe hypoxaemia?

Narrow spectrum
antibiotics

YES

Broad spectrum
antibiotics

Assessment of the delta value of a biomarker (expressed

as a dierence between or ratio of a biomarker value
during evolution and on the day of diagnosis) could
contribute to more objective decisions, but increases
costs.92 Resolution of chest radiographs, white blood cell
count, or clearance of respiratory secretions does not
help the decision.93 Therefore no present evidence
supports an objective decision to prolong treatment in
VARI.

Areas of uncertainty
Favourable

Clinical response

De-escalate
Combination therapy to monotherapy
Change to narrower spectrum
according to culture results
Stop antibiotic if culture negative

Delayed

Switch to broader spectrum

antibiotics
Add second agent
Increase dose
Rule out empyema or abscess

Figure 3: Approach to the management of VARIs

COPD=chronic obstructive pulmonary disease. IC=immunocompromised.
VARI=ventilator-associated respiratory infection. *Take into account
colonisation (if known) and risk factors for specic pathogens (if present).

VAT. Consistent reports of increased length of ICU stay

for patients with VAT due to an extended need for
mechanical ventilation provide a strong rationale for
antibiotic administration. The duration of targeted VAT
treatment has not been established, but VAT might
respond to short treatment courses.91 Further research
with appropriate methods and design is warranted to
identify the subgroup of patients with VAT that would
benet from antimicrobial treatment and the group
that could safely have antimicrobial treatment withheld
or restricted.
Key issues for management of VAP and VAT
(gure 3) are: identication of when to start an
antibiotic; which microbiological test can be of help to
guide antibiotic prescriptions (quantitative cultures of
a high quality respiratory specimen are advisable);
which organism should be covered on the basis of
direct staining and presence of potential risk factors;
which initial drug should be prescribed and at which
dose on the basis of baseline susceptibilities, patients
disorders, and previous antibiotic exposure; and
duration of treatment.
The present standard duration of antibiotic treatment
for VAP is 1 week, although patients with core pathogens
(S pneumoniae, Haemophilus inuenzae, and methicillinsensitive S aureus) present quick resolution and might
benet from even shorter courses. Conversely, in
patients with P aeruginosa VAP who receive inappropriate
initial treatment or those with MRSA VAP, the resolution
is usually delayed and more than 10 days of antibiotics
are needed.45 Improvement in oxygenation and
defervescence occurs within 3 days in most patients.
770

A long-standing diculty in ICU care is the dierential

diagnosis between the inammatory response and
infection. In many cases, the challenge is to establish
whether bacteria are merely colonising the patient or
whether they are the cause of disease. Two main
questions remain. First, how can clinicians conclusively
establish that the bacterial growth in the respiratory tract
sample is the cause of the inammatory response in a
ventilated patient? Second, how can they establish that
the microorganism isolated from an upper airway sample
is the cause of the disease in the lower airway? The
biomarkers needed for pre-emptive treatment are often
insucient to resolve these issues.94 Procalcitonin is
most widely used,95,96 but research continues into other
biomarkers oering better sensitivity and specicity.96,97
In the future, genomics might provide a better answer to
these issues,97,98 but further research is needed into gene
expression before procalcitonin (or others) can be widely
used in clinical practice.
Another present focus of research is timely identication of the causative microorganism.95,99,100 Matrixassisted laser desorption ionisation time-of-ight mass
spectrometry is a promising technique. It can identify
either Gram-positive or Gram-negative bacteria (to
species level) within a few minutes from a low bacterial
load, providing a notable clinical benet.100 A system of
this kind, able to identify the pathogen and its sensitivity
quickly and accurately in a point-of-care test, would
signal a new era for management of respiratory
infections. However, the technique is currently costly
and some issues regarding validation still need to be
claried.
Unlike the scores used to stratify severity of illness in
community-acquired respiratory infections, stratication
is not common practice in mechanically ventilated
patients. However, because VARIs are heterogeneous
and need to be compared, Lisboa and colleagues101
designed the VAP PIRO score to stratify risk of death.
This score combines information on predisposition
(comorbidities), injury (bacteraemia), response (systolic
blood pressure <90 mm Hg or use of vasopressors), and
organ failure (acute respiratory distress syndrome). The
score classies patients into three categories (01, 2, and
34) dependent on the risk of ICU mortality: low (one of
every eight patients), intermediate (one of every two
patients), and high (four of every ve patients). A cohort
www.thelancet.com/respiratory Vol 2 September 2014

Review

Search strategy and selection criteria

We searched Medline for articles published in English
between January 1, 2004 and February 28, 2014 using the
following keywords: ventilator-associated pneumonia and
ventilator-associated tracheobronchitis. We restricted
inclusion of reports to those in the adult intensive care unit
population with use of the term adult [MeSH].

validation showed a good correlation with health-care

resource use.101 The method needs rening and perhaps
addition of biomarkers for intermediate severity to
improve its stratication capacity.
In a prospective randomised study102 of patients
undergoing major heart surgery, patients were
assigned to either standard care or 3 days of
prophylactic antibiotics (meropenem and linezolid) to
determine the possibility of avoiding VARI with
antibiotic prophylaxis. No patient-centred outcome
benets (ie, mortality, days on ventilation, or length of
ICU stay) were identied, but the authors noted a
signicantly lower incidence of VARI (VAP and VAT
combined) in the intervention group than the control
group and a 45 day delay in the onset of pneumonia.102
The authors also reported an association between
3 days of pre-emptive antibiotic treatment and an
increase in resistance to linezolid.102 Our group
established in a previous study that an antimicrobial
regimen of more than 2 days might be associated with
increased resistance and modication of the gut
ora.103 Our groups nding that VAP can be prevented
with one dose of antibiotic, as is the case in surgical
prophylaxis, underlines the importance of appropriate
antimicrobial stewardship in the ICU.104
Inhalation could become a new method of antibiotic
delivery because of its ability to achieve high lung tissue
concentrations, minimising systemic absorption.81,105
Use of nebulised ceftazidime (15 mg/kgevery 3 h) and
amikacin (25 mg/kgdaily) provided high lung tissue
concentrations and rapid bacterial killing in VAP caused
by P aeruginosa.106 Outcomes in patients with multidrugresistant non-fermentative Gram-negative bacilli when
high dose colistin (5 M international units every 8 h)
was nebulised, either in isolation or combined with
parenteral treatment, have been shown to be similar.107
This method of administration of high concentrations
of antibiotics in the distal airways increases bacterial
killing in the case of organisms with very high minimum
inhibitory concentrations and customised use according
to the pathogen and minimum inhibitory concentration,
and is an opportunity for further research. Potential
adverse events, such as blockage of the expiratory limb
of the ventilator or bronchospasm and contraindications
in severe hypoxaemia are potential limitations that need
further research.
www.thelancet.com/respiratory Vol 2 September 2014

Conclusions
Respiratory infections in mechanically ventilated
patients present in the form of severe sepsis or septic
shock in intubated patients. Purulent respiratory
secretions are needed for diagnosis, but dierentiation
between pneumonia and tracheobronchitis on the basis
of clinical ndings is a challenge. Both presentations
are associated with extended ventilation and ICU stay,
providing the rationale for treatment. Key VARI
management issues include identication of the
pathogen, initial antibiotic choice, and resolution pattern
and criteria. New opportunities for research include
roles for biomarkers, early causal diagnosis with
molecular diagnostic techniques, and optimisation and
customisation of treatment.
Contributors
JR reviewed the literature and conceived, organised, wrote, and gave
nal approval to the manuscript. DK and TL reviewed the literature
and organised and wrote the report.
Declaration of interests
JR has served on Speakers Bureaux and advisory boards for Pzer and
received grant research support from Sano Pasteur, Kenta, and
Cubist, and the Spanish Health Ministry (FISS11/1122). TL and DK
declare no competing interests.
References
1
Vincent JL, Rello J, Marshall J, et al, for the EPIC II Group of
Investigators. International study of the prevalence and outcomes
of infection in intensive care units. JAMA 2009; 302: 232329.
2
Klompas M. Does this patient have ventilator-associated
pneumonia? JAMA 2007; 297: 158393.
3
Klompas M. What can we learn from international ventilatorassociated pneumonia rates? Crit Care Med 2012; 40: 330304.
4
Niederman MS. Hospital-acquired pneumonia, health careassociated pneumonia, ventilator-associated pneumonia, and
ventilator-associated tracheobronchitis: denitions and challenges
in trial design. Clin Infect Dis 2010; 51 (suppl 1): S1217.
5
Improving surveillance for ventilator-associated events in adults.
http://www.cdc.gov/nhsn/PDFs/pscManual/10-VAE_FINAL.pdf
(accessed April 13, 2014).
6
Stevens JP, Kachniarz B, Wright SB, et al. When policy gets it
right: variability in U.S. hospitals diagnosis of ventilatorassociated pneumonia. Crit Care Med 2014; 42: 497503.
7
Pugin J, Auckenthaler R, Mili N, Janssens JP, Lew PD, Suter PM.
Diagnosis of ventilator-associated pneumonia by bacteriologic
analysis of bronchoscopic and nonbronchoscopic blind
bronchoalveolar lavage uid. Am Rev Respir Dis 1991;
143: 112129.
8
Luna CM, Blanzaco D, Niederman MS, et al. Resolution of
ventilator associated pneumonia: prospective evaluation of the
clinical pulmonary infection score as an early clinical predictor for
outcome. Crit Care Med 2003; 31: 67682.
9
Parks NA, Magnotti LJ, Weinberg JA, et al. Use of the clinical
pulmonary infection score to guide therapy for ventilatorassociated pneumonia risks antibiotic overexposure in patients
with trauma. J Trauma Acute Care Surg 2012; 73: 5258.
10 Dallas J, Skrupky L, Abebe N, Boyle WA 3rd, Kollef MH.
Ventilator-associated tracheobronchitis in a mixed surgical and
medical ICU population. Chest 2011; 139: 51318.
11 Martin-Loeches I, Papiol E, Almansa R, Lpez-Campos G,
Bermejo-Martin JF, Rello J. Intubated patients developing
tracheobronchitis or pneumonia have distinctive complement
system gene expression signatures in the pre-infection period:
a pilot study. Med Intensiva 2012; 36: 25763.
12 Solomon DH, Wobb J, Buttaro BA, Truant A, Soliman AM.
Characterization of bacterial biolms on tracheostomy tubes.
Laryngoscope 2009; 119: 163338.

771

Review

13
14
15

16

17

18

19

20

21

22

23

24

25

26

27

28

29

30

31

32

772

Meslemani D, Yaremchuk K, Rontal M. Presence of biolm on

adult tracheostomy tubes. Ear Nose Throat J 2010; 89: 496504.
Blot S, Lisboa T, Angles R, Rello J. Prevention of VAP: is zero rate
possible? Clin Chest Med 2011; 32: 59199.
Melsen WG, Rovers MM, Groenwold RH, et al. Attributable
mortality of ventilator-associated pneumonia: a meta-analysis of
individual patient data from randomised prevention studies.
Lancet Infect Dis 2013; 13: 66571.
Cuellar LE, Fernandez-Maldonado E, Rosenthal VD, et al.
Device-associated infection rates and mortality in intensive care
units of Peruvian hospitals: ndings of the International
Nosocomial Infection Control Consortium.
Rev Panam Salud Publica 2008; 24: 1624.
Guanche-Garcell H, Requejo-Pino O, Rosenthal VD, Morales-Perez C,
Delgado-Gonzalez O, Fernandez-Gonzalez D. Device-associated
infection rates in adult intensive care units of Cuban university
hospitals: International Nosocomial Infection Control Consortium
(INICC) ndings. Int J Infect Dis 2011; 15: e35762.
Hu B, Tao L, Rosenthal VD, et al. Device-associated infection
rates, device use, length of stay, and mortality in intensive care
units of 4 Chinese hospitals: International Nosocomial Control
Consortium ndings. Am J Infect Control 4: 30106.
Mehta A, Rosenthal VD, Mehta Y, et al. Device-associated
nosocomial infection rates in intensive care units of seven Indian
cities. Findings of the International Nosocomial Infection Control
Consortium (INICC). J Hosp Infect 2007; 2: 16874.
Moreno CA, Rosenthal VD, Olarte N, et al. Device-associated
infection rate and mortality in intensive care units of 9 Colombian
hospitals: ndings of the International Nosocomial Infection
Control Consortium. Infect Control Hosp Epidemiol 2006; 27: 34956.
Salomao R, Rosenthal VD, Grimberg G, et al. Device-associated
infection rates in intensive care units of Brazilian hospitals:
ndings of the International Nosocomial Infection Control
Consortium. Rev Panam Salud Publica 2008; 24: 195202.
Rosenthal VD, Guzman S, Crnich C. Device-associated
nosocomial infection rates in intensive care units of Argentina.
Infect Control Hosp Epidemiol 2004; 25: 25155.
Riera J, Caralt B, Augustin S, et al. Complications in the immediate
postoperatory of lung transplantation: three years of practice at a
high-experienced center. Chest 2014; 145 (suppl 3): 631A.
Magret M, Amaya-Villar R, Garnacho J, et al, for the EU-VAP/CAP
Study Group. Ventilator-associated pneumonia in trauma patients
is associated with lower mortality: results from EU-VAP study.
J Trauma 2010; 69: 84954.
Muscedere J, Sinu T, Heyland DK, et al, for the Canadian Critical
Care Trials Group. The clinical impact and preventability of
ventilator-associated conditions in critically ill patients who are
mechanically ventilated. Chest 2013; 144: 145360.
Rello J, Afonso E, Lisboa T, et al, for the FADO Project
Investigators. A care bundle approach for prevention of ventilatorassociated pneumonia. Clin Microbiol Infect 2013; 19: 36369.
Rello J, Ollendorf DA, Oster G, et al, for the VAP Outcomes Scientic
Advisory Group. Epidemiology and outcomes of ventilator-associated
pneumonia in a large US database. Chest 2002; 122: 211521.
Koulenti D, Lisboa T, Brun-Buisson C, et al, for the EU-VAP/CAP
Study Group. Spectrum of practice in the diagnosis of nosocomial
pneumonia in patients requiring mechanical ventilation in
European intensive care units. Crit Care Med 2009; 37: 236068.
Blot S, Koulenti D, Dimopoulos G, et al, for the EU-VAP Study
Investigators. Prevalence, risk factors, and mortality for ventilatorassociated pneumonia in middle-aged, old, and very old critically
ill patients. Crit Care Med 2014; 42: 60109.
Esperatti M, Ferrer M, Theessen A, et al. Nosocomial pneumonia
in the intensive care unit acquired by mechanically ventilated
versus nonventilated patients. Am J Respir Crit Care Med 2010;
182: 153339.
Kollef MH, Shorr A, Tabak YP, Gupta V, Liu LZ, Johannes RS.
Epidemiology and outcomes of health-care-associated pneumonia:
results from a large US database of culture-positive pneumonia.
Chest 2005; 128: 385462.
Lee MS, Walker V, Chen LF, Sexton DJ, Anderson DJ. The
epidemiology of ventilator-associated pneumonia in a network of
community hospitals: a prospective multicenter study.
Infect Control Hosp Epidemiol 2013; 34: 65762.

33

34

35

36
37

38

39

40

41

42

43

44

45

46

47

48

49

50

51

52

Canadian Critical Care Trials Group. A randomized trial of

diagnostic techniques for ventilator-associated pneumonia.
N Engl J Med 2006; 355: 261930.
Rosenthal VD, Rodrigues C, Alvarez-Moreno N, et al.
Eectiveness of a multidimensional approach for prevention of
ventilator-associated pneumonia in adult intensive care units
from 14 developing countries of four continents: ndings of the
International Nosocomial Infection Control Consortium.
Crit Care Med 2012; 40: 312128.
Bekaert M, Timsit JF, Vansteelandt S, et al, for the Outcomerea
Study Group. Attributable mortality of ventilator-associated
pneumonia: a reappraisal using causal analysis.
Am J Respir Crit Care Med 2011; 184: 113339.
Rice LB. Federal funding for the study of antimicrobial resistance in
nosocomial pathogens: no ESKAPE. J Infect Dis 2009; 197: 107981.
Sandiumenge A, Lisboa T, Gomez F, Hernandez P, Canadell L,
Rello J. Eect of antibiotic diversity on ventilator-associated
pneumonia caused by ESKAPE Organisms. Chest 2011; 140: 64351.
Depuydt PO, Vandjick DM, Bekaert MA, et al. Determinants and
impact of multidrug antibiotic resistance in pathogens causing
ventilator-associated pneumonia. Crit Care 2008; 12: R142.
Di Pasquale M, Ferrer M, Esperatti M, et al. Assessment of
severity of ICU-acquired pneumonia and association with
etiology. Crit Care Med 2014; 42: 30312.
Rello J, Ulldemolins M, Lisboa T, et al, for the EU-VAP/CAP Study
group. Determinants of prescription and choice of empirical
therapy in hospital-acquired pneumonia and ventilator-associated
pneumonia. Eur Respir J 2011; 37: 133239.
Rello J, Torres A, Ricart M, et al. Ventilator-associated pneumonia
by Staphylococcus aureus. Comparison of methicillin-resistant and
methicillin-sensitive episodes. Am J Respir Crit Care Med 1994;
150: 154549.
Rosenthal VD, Bijie H, Maki DG, et al. International Nosocomial
Infection Control Consortium (INICC) report, data summary of
36 countries, for 20042009. Am J Infect Control 2012; 40: 396407.
Agbaht K, Diaz E, Muoz E, et al. Bacteremia in patients with
ventilator-associated pneumonia is associated with increased
mortality: a study comparing bacteremic vs. nonbacteremic
ventilator-associated pneumonia. Crit Care Med 2007;
35: 206470.
Hanberger H, Walther S, Leone M, et al, for the EPIC II Group of
Investigators. Increased mortality associated with methicillinresistant Staphylococcus aureus (MRSA) infection in the intensive
care unit: results from the EPIC II study. Int J Antimicrob Agents
2011; 38: 33135.
Vidaur L, Planas K, Sierra R, et al. Ventilator-associated
pneumonia: impact of organisms on clinical resolution and
medical resources utilization. Chest 2008; 133: 62532.
Wunderink RG, Niederman MS, Kollef MH, et al. Linezolid in
methicillin-resistant Staphylococcus aureus nosocomial
pneumonia: a randomized, controlled study. Clin Infect Dis 2012;
54: 62129.
Lahey T. Questionable superiority of linezolid for methicillinresistant Staphylococcus aureus nosocomial pneumonia: watch
where you step. Clin Infect Dis 2012; 55: 15960.
Tumbarello M, De Pascale G, Trecarichi EM, et al. Clinical
outcomes of Pseudomonas aeruginosa pneumonia in intensive care
unit patients. Intensive Care Med 2013; 39: 68292.
Sandiumenge A, Rello J. Ventilator-associated pneumonia caused
by ESKAPE organisms: cause, clinical features, and management.
Curr Opin Pulm Med 2012; 18: 18793.
Garnacho-Montero J, Sa-Borges M, Sole-Violan J, et al. Optimal
management therapy for Pseudomonas aeruginosa ventilatorassociated pneumonia: an observational, multicenter study
comparing monotherapy with combination antibiotic therapy.
Crit Care Med 2007; 35: 188895.
Aloush V, Navon-Venezia S, Seigman-Igra Y, Cabili S, Carmeli Y.
Multidrug-resistant Pseudomonas aeruginosa: risk factors and
clinical impact. Antimicrob Agents Chemother 2006; 50: 4348.
Parker CM, Kutsogiannis J, Muscedere J, et al, for the Canadian
Critical Care Trials Group. Ventilator-associated pneumonia
caused by multidrug-resistant organisms or
Pseudomonas aeruginosa: prevalence, incidence, risk factors, and
outcomes. J Crit Care 2008; 23: 1826.

www.thelancet.com/respiratory Vol 2 September 2014

Review

53

54

55

56

57

58

59

60

61

62

63

64

65

66

67

68

69

70

71

Nakamura A, Miyake K, Misawa S, et al. Meropenem as predictive

risk factor for isolation of multidrug-resistant Pseudomonas
aeruginosa. J Hosp Infect 2013; 83: 15355.
Liew YX, Tan TT, Lee W, et al. Risk factors for extreme-drug
resistant Pseudomonas aeruginosa infections in patients with
hematologic malignancies. Am J Infect Control 2013; 41: 14044.
Park YS, Lee H, Chin BS, et al. Acquisition of extensive drugresistant Pseudomonas aeruginosa among hospitalized patients:
risk factors and resistance mechanisms to carbapenems.
J Hosp Infect 2011; 79: 5458.
Pea C, Gmez-Zorrilla S, Oriol I, et al. Impact of multidrug
resistance on Pseudomonas aeruginosa ventilator-associated
pneumonia outcome: predictors of early and crude mortality.
Eur J Clin Microbiol Infect Dis 2013; 32: 41320.
Azoulay E, Timsit JF, Taet M, et al. for the Outcomerea Study
Group. Candida colonization of the respiratory tract and
subsequent Pseudomonas ventilator-associated pneumonia. Chest
2006; 129: 11017.
Hamet M, Pavon A, Dalle F, et al. Candida spp. airway
colonization could promote antibiotic-resistant bacteria selection
in patients with suspected ventilator-associated pneumonia.
Intensive Care Med 2012; 38: 127279.
Ricard JD, Roux D. Candida colonization in ventilated ICU
patients: no longer a bystander! Intensive Care Med 2012;
38: 124345.
Nseir S, Jozefowicz E, Cavestri B, et al. Impact of antifungal
treatment on CandidaPseudomonas interaction: a preliminary
retrospective case-control study. Intensive Care Med 2007;
33: 137142.
Boyer A, Doussau A, Thibault R, et al. Pseudomonas aeruginosa
acquisition on an intensive care unit: relationship between
antibiotic selective pressure and patients environment. Crit Care
2011; 15: R55.
Cook PP, Gooch M, Rizzo S. Reduction in uoroquinolone use
following introduction of ertapenem into a hospital formulary is
associated with improvement in susceptibility of
Pseudomonas aeruginosa to group 2 carbapenems: a 10-year study.
Antimicrob Agents Chemother 2011; 55: 559701.
Augustin P, Kermarrec N, Muller-Serieys C, et al. Risk factors for
multidrug resistant bacteria and optimization of empirical antibiotic
therapy in postoperative peritonitis. Crit Care 2010; 14: R20.
Ginn AN, Wiklendt AM, Gidding HF, et al. The ecology of
antibiotic use in the ICU: homogeneous prescribing of cefepime
but not tazocin selects for antibiotic resistant infection. PLoS One
2012; 7: e38719.
Kanj SS, Kanafani ZA. Current concepts in antimicrobial
therapy against resistant Gram-negative organisms: extendedspectrum beta-lactamase-producing Enterobacteriaceae,
carbapenem-resistant Enterobacteriaceae, and multidrugresistant Pseudomonas aeruginosa. Mayo Clin Proc 2011;
86: 25059.
Yang K, Zhuo H, Guglielmo BJ, Wiener-Kronish J. Multidrugresistant Pseudomonas aeruginosa ventilator-associated
pneumonia: the role of endotracheal aspirate surveillance
cultures. Ann Pharmacother 2009; 43: 2835.
Plss-Suard C, Pannatier A, Kronenberg A, Mhlemann K,
Zanetti G. Impact of antibiotic use on carbapenem resistance in
Pseudomonas aeruginosa: is there a role for antibiotic diversity?
Antimicrob Agents Chemother 2013; 57: 170913.
Eagye KJ, Kuti JL, Nicolau DP. Risk factors and outcomes
associated with isolation of meropenem high-level-resistant
Pseudomonas aeruginosa. Infect Control Hosp Epidemiol 2009;
30: 74652.
Kaminski C, Timsit JF, Dubois Y, et al. for the OUTCOMEREA
study group. Impact of ureido/carboxypenicillin resistance on the
prognosis of ventilator-associated pneumonia due to
Pseudomonas aeruginosa. Crit Care 2011; 15: R112.
Combes A, Luyt CE, Fagon JY, Wol M, Trouillet JL, Chastre J.
Impact of piperacillin resistance on the outcome of Pseudomonas
ventilator-associated pneumonia. Intensive Care Med 2006;
32: 197078.
Hueck CL, Hueck CJ. Type III protein secretion systems in bacterial
pathogens of animals and plants. Microbiol Mol Biol Rev 1998;
62: 379433.

www.thelancet.com/respiratory Vol 2 September 2014

72

73

74

75

76

77

78

79

80

81

82
83

84

85

86

87
88

89

90

91

92

93

94

Veesenmeyer JL, Hauser AR, Lisboa T, Rello J.

Pseudomonas aeruginosa virulence and therapy: evolving
translational strategies. Crit Care Med 2009; 37: 177786.
El Solh AA, Akinnusi ME, Wiener-Kronish JP, Lynch SV, Pineda LA,
Szarpa K. Persistent infection with Pseudomonas aeruginosa in
ventilator-associated pneumonia. Am J Respir Crit Care Med 2008;
178: 51319.
Lu Q, Eggimann P, Luyt CE, et al. Pseudomonas aeruginosa serotypes
in nosocomial pneumonia: prevalence and clinical outcomes.
Crit Care 2014; 18: R17.
Le Berre R, Nguyen S, Nowak E, et al. for the Pyopneumagen
Group. Relative contribution of three main virulence factors in
Pseudomonas aeruginosa pneumonia. Crit Care Med 2011; 39: 211320.
Faure K, Shimabukuro D, Ajayi T, Allmond LR, Sawa T,
Wiener-Kronish JP. O-antigen serotypes and type III secretory
toxins in clinical isolates of Pseudomonas aeruginosa. J Clin Microbiol
2003; 41: 215860.
Lu Q, Rouby JJ, Laterre PF, et al. Pharmacokinetics and safety of
panobacumab: specic adjunctive immunotherapy in critical
patients with nosocomial Pseudomonas aeruginosa O11 pneumonia.
J Antimicrob Chemother 2011; 66: 111016.
Baraibar J, Correa H, Mariscal D, Gallego M, Valls J, Rello J. Risk
factors for infection by Acinetobacter baumannii in intubated
patients with nosocomial pneumonia. Chest 1997; 112: 105054.
Mah MW, Memish ZA, Cunningham G, Bannatyne RM. Outbreak
of Acinetobacter baumannii in an intensive care unit associated with
tracheostomy. Am J Infect Control 2001; 29: 28488.
Playford EG, Craig JC, Iredell JR. Carbapenem-resistant Acinetobacter
baumannii in intensive care unit patients: risk factors for acquisition,
infection and their consequences. J Hosp Infect 2007; 65: 20411.
Rouby JJ, Bouhemad B, Monsel A, Brisson H, Arbelot C, Lu Q, for
the Nebulized Antibiotics Study Group. Aerosolized antibiotics for
ventilator-associated pneumonia: lessons from experimental
studies. Anesthesiology 2012; 117: 136480.
Rello J, Vidaur L, Sandiumenge A, et al. De-escalation therapy in
ventilator associated pneumonia. Crit Care Med 2004; 32: 218390.
Raman K, Nailor MD, Nicolau DP, et al. Early antibiotic
discontinuation in patients with clinically suspected ventilatorassociated pneumonia and negative quantitative bronchoscopy
cultures. Crit Care Med 2013; 41: 165663.
Shorr AF, Sherner JH, Jackson WL, Kollef MH. Invasive approaches
to the diagnosis of ventilator-associated pneumonia: a metaanalysis. Crit Care Med 2005; 33: 4653.
Tejerina E, Esteban A, Fernandez-Segoviano P, et al. Accuracy of
clinical denitions of ventilator-associated pneumonia: comparison
with autopsy ndings. J Crit Care 2010; 25: 6268.
Chastre J, Wol M, Fagon JY, et al. Comparison of 8 vs 15 days of
antibiotic therapy for ventilator associated pneumonia in adults: a
randomized trial. JAMA 2003; 290: 258898.
Lisboa T, Nagel F. Infection with multidrug-resistant agents in the
ICU: how to escape? Rev Bras Ter Intensiva 2011; 23: 12024.
Rodrguez A, Pvoa P, Nseir S, et al, on behalf of The TAVeM Group
Investigators. Incidence and diagnosis of ventilator-associated
tracheobronchitis (VAT) in the intensive care unit: an
international online survey. Crit Care 2014; 18: R32.
Nseir S, Favory R, Jozefowicz E, et al, for the VAT Study Group.
Antimicrobial treatment for ventilator-associated tracheobronchitis: a
randomized, controlled, multicentre study. Crit Care 2008; 12: R62.
Palmer LB, Smaldone GC, Chen JJ, et al. Aerosolized antibiotics
and ventilator-associated tracheobronchitis in the intensive care
unit. Crit Care Med 2008; 36: 200813.
Craven DE, Chroneou A, Zias N, Hjalmarson KI. Ventilatorassociated tracheobronchitis: the impact of targeted antibiotic
therapy on patient outcomes. Chest 2009; 135: 52128.
Seligman R, Meisner M, Lisboa TC, et al. Decreases in procalcitonin
and C-reactive protein are strong predictors of survival in ventilatorassociated pneumonia. Crit Care 2006; 10: R125.
Vidaur L, Gualis B, Rodriguez A, et al. Clinical resolution in
patients with suspicion of ventilator-associated pneumonia: a cohort
study comparing patients with and without acute respiratory
distress syndrome. Crit Care Med 2005; 33: 124853.
Calfee CS, Pugin J. The search for diagnostic markers in sepsis:
many miles yet to go. Am J Respir Crit Care Med 2012; 186: 24.

773

Review

95

Pvoa P, Salluh JI. Biomarker-guided antibiotic therapy in adult

critically ill patients: a critical review. Ann Intensive Care 2012;
2: 32.
96 Mitsuma SF, Mansour MK, Dekker JP. Promising new assays and
technologies for the diagnosis and management of infectious
diseases. Clin Infect Dis 2013; 56: 99602.
97 Sutherland A, Thomas M, Brandon R, et al. Development and
validation of a novel molecular biomarker diagnostic test for the
early detection of sepsis. Crit Care 2011; 15: R149.
98 Johnson SB, Lissauer M, Bochicchio GV, Moore R, Cross AS,
Scalea TM. Gene expression proles dierentiate between sterile
SIRS and early sepsis. Ann Surg 2007; 245: 61121.
99 Prodhom G, Bizzini A, Durussel C, Bille J, Greub G.
Matrix-assisted laser desorption ionization-time of ight mass
spectrometry for direct bacterial identication from positive blood
culture pellets. J Clin Microbiol 2010; 48: 148183.
100 Clerc O, Prodhom G, Vogne C, Bizzini A, Calandra T, Greub G.
Impact of matrix-assisted laser desorption ionization time-of-ight
mass spectrometry on the clinical management of patients with
Gram-negative bacteremia: a prospective observational study.
Clin Infect Dis 2013; 56: 110107.
101 Lisboa T, Diaz E, Sa-Borges M, et al. The ventilator-associated
pneumonia PIRO score: a tool for predicting ICU mortality and
health-care resources use in ventilator-associated pneumonia.
Chest 2008; 134: 120816.

774

102 Bouza E, Perez-Granda MJ, Hortal J, Barrio JM, Cercenado E,

Munoz P. Pre-emptive broad-spectrum treatment for ventilatorassociated pneumonia in high-risk patients. Intensive Care Med
2013; 39: 154755.
103 Rello J, Ausina V, Ricart M, Castella J, Prats G. Impact of prior
antibiotic therapy on etiology and outcomes of ventilator-associated
pneumonia. Chest 1993; 104: 123035.
104 Rello J. Antibiotic stewardship in the ICU. Chest 2013; 143: 119596.
105 Luyt CE, Combes A, Nieszkowska A, et al. Aerosolized antibiotics to
treat ventilator associated pneumonia. Curr Opin Infect Dis 2009;
22: 15458.
106 Lu Q, Yang J, Liu Z, et al. Nebulized ceftazidime and amikacin in
ventilator-associated pneumonia caused by Pseudomonas aeruginosa.
Am J Respir Crit Care Med 2011; 184: 10615.
107 Lu Q, Luo R, Bodin L, et al. Ecacy of high-dose nebulized colistine
in ventilator-associated pneumonia caused by multidrug resistant
Pseudomonas aeruginosa and Acinetobacter baumannii.
Anesthesiology 2012; 117: 133547.

www.thelancet.com/respiratory Vol 2 September 2014