Ven t i l a t o r- A s s o c i a t e d

P n e u m o n i a i n C r i t i c a l l y Il l
C h i l d ren : A New Paradigm
a, b
Peter M. Mourani, MD *, Marci K. Sontag, PhD

KEYWORDS
 Ventilator-associated pneumonia  Mechanical ventilation  Microbiome
 Metagenomics  Pediatric intensive care

KEY POINTS
 Surveillance definitions for ventilator-associated pneumonia (VAP) are in the process of
being updated. Further evaluation is needed to assure that the patients who are most
amenable to targeted prevention strategies are identified.
 VAP is the most common indication for antibiotic use in the pediatric intensive care unit.
Present approaches to clinical cultures and antibiotic use may actually increase risk for
VAP by depleting protective commensal organisms and selecting for antibiotic resistant
pathogens.
 Recent metagenomic and proteomic technologies offer the potential to better interrogate
the relationships between an intubated individual’s respiratory microbiota, the host’s im-
mune response, and the underlying disease process to provide important insights into the
pathogenesis of VAP.
 The normal lung is colonized with diverse microbial communities. It seems likely that crit-
ical illness and its care contribute to a dysbiosis and the selection of a disease-promoting
microbiome or pathobiome that increases risk for nosocomial infection, including VAP.

INTRODUCTION

Mechanically ventilated children are at high risk for nosocomial infections, including
ventilator-associated pneumonia (VAP). Children who develop VAP have an increased
risk of mortality1 and morbidities such as prolonged intubation and intensive care unit
(ICU) stays and the need for extensive rehabilitation.2 VAP is the most common noso-
comial infection in mechanically ventilated patients,1 occurring in up to 32% of pedi-
atric ICU (PICU) patients who require mechanical ventilation (MV) for more than

Disclosure: The authors have no financial relationships or conflicts of interest to report.

a
Section of Critical Care, Department of Pediatrics, University of Colorado Denver, School of
Medicine, Children’s Hospital Colorado, 13121 East 17th Avenue, MS8414, Aurora, CO 80045,
USA; b Department of Epidemiology, Colorado School of Public Health, University of Colorado
Denver Anschutz Medical Campus, 13001 East 17th, B119, Aurora, CO 80045, USA
* Corresponding author.
E-mail address: Peter.Mourani@childrenscolorado.org

Pediatr Clin N Am - (2017) -–-

http://dx.doi.org/10.1016/j.pcl.2017.06.005 pediatric.theclinics.com
0031-3955/17/ª 2017 Elsevier Inc. All rights reserved.
2 Mourani & Sontag

24 hours.3,4 VAP is associated with a 2-fold to 3-fold increase in mortality in ventilated

children,5,6 and increases total hospitalization costs and resource utilization,
increasing duration of MV by 5 to 11 days and PICU length of stay by 11 to
34 days.3,4,6,7 The suspicion and/or the diagnosis of VAP remain the primary reason
for antibiotic administration in the PICU.8 Thus, VAP remains a significant obstacle
to the management of pediatric critical illnesses and injuries.
The traditional view of the pathogenesis of pneumonia was based on the premise
that the lung was sterile, and that infection occurred by the introduction of pathogens
either by direct inoculation from the environment or through the blood stream. How-
ever, recent evidence has revealed that the lungs contain diverse microbial popula-
tions.9,10 The endogenous microbiota are likely critical regulators of both pathogen
behavior and host responses in the airways. Infection, therefore, occurs as a result
of the combination of dysbiosis and failure of the host immune response. Because
the lungs are inhabited by microbial populations (akin to the intestinal tract albeit in
far lower density,9–12), the simple paradigm of a single pathogen gaining access to
a sterile lower respiratory tract to cause infection has evolved. Pathogens are intro-
duced to a pre-existing and often complex microbial community9,11 that either allows,
facilitates, or hinders the potential pathogen, and consequently determines whether
progression to VAP occurs. Limited understanding of the microbial and host factors
associated with VAP pathogenesis has precluded development of truly effective pre-
vention and treatment strategies.13,14
This article briefly discusses the changing criteria by which VAP is diagnosed and
the limitations of these definitions for both surveillance and clinical research attempt-
ing to identify risk factors for and the pathogenesis of VAP. It also discusses the pros-
pects for a deeper understanding of the pathogenesis of VAP in the new era of
metagenomics and proteomics, together with the recent evidence about the role of
microbiome in health and disease.

SURVEILLANCE DEFINITION

In the absence of a readily available microbiologic gold standard, standardized clinical

criteria for VAP were first developed in 2002 by the Centers for Disease Control (CDC)
and National Nosocomial Infections Surveillance (NNIS) to allow for consistent diag-
nosis and reporting.15 The CDC/NNIS definitions, which underwent some minor mod-
ifications during the subsequent decade, relied on combinations of radiographic,
clinical, and laboratory evidence in the consideration of a VAP diagnosis. Because
traditional culture from respiratory samples depends on the chosen medium, positive
cultures from these methods were optional parts of the pediatric criteria.16–19 When
rigorously applied, CDC/NNIS VAP criteria have been shown to be associated with
predictably poor outcomes.20–22 However, application of these criteria is time-
consuming and inconsistent because definitions of components of the criteria are sub-
jective and imprecise.23,24 Several studies demonstrate this variability. Among 64
PICUs reporting data to CDC’s National Healthcare Safety Network (NHSN) between
2007 and 2012, the incidence of VAP decreased from 1.9 to 0.7 per 1000 ventilator-
days.25 However, other published prospective pediatric studies report rates as high
as 7.1 per 1000 ventilator-days in the United States.6 That different studies report
markedly different VAP rates suggest that diverse populations of children have unique
risk for VAP or that interpretation of the criteria by investigators is subjective. Similarly,
from 2006 to 2012, adult medical ICUs reporting data to NHSN demonstrated a
decline in VAP rates from 3.1 to 0.9 per 1000 ventilator-days, whereas other evalua-
tions have concluded that VAP rates have not declined during this period with a
 New VAP Paradigm 3

consistent rate of approximately 10% of ventilated patients.26 The data suggested

that surveillance using previous definitions may be unreliable, and efforts to identify
more objective and consistently applied VAP criteria are justified.
A comprehensive revision of the VAP criteria was undertaken by the CDC/NNIS in
2013 to develop a more objective and easily applied definition. The revision created
a 3-tier definition algorithm for ventilator-associated events (VAEs). The first tier,
ventilator-associated condition (VAC), defines mechanically ventilated patients
following a sustained period of stability or improvement after intubation who then
experience a sustained respiratory deterioration (defined by elevations in positive-
end expiratory pressure [PEEP] or fractional inspired oxygen [FiO2]). The second tier,
infectious VAC (IVAC), describes patients with VAC and evidence of infection defined
by changes in temperature or white blood cell count (WBC) and the clinician’s decision
to initiate antibiotics. The third tier defines possible and probable VAP and requires
that patients with IVAC also have laboratory and/or microbiological evidence of respi-
ratory infection (Table 1).27
These definitions were only implemented for adult patients, whereas the diagnosis
of VAP in children continues to use the older criteria. Recent efforts by the CDC’s Pe-
diatric Ventilator-Associated Pneumonia Surveillance Definition Working Group,
based on work by the Pediatric Ventilator-Associated Conditions Study Team,28
have resulted in a new set of definitions patterned after the adult criteria. The main dif-
ference is that pediatric VAC determination is based on elevation of mean airway pres-
sure as opposed to PEEP, and the threshold increase in FiO2 is 0.25 versus 0.2 in
adults.29 The pediatric definition eliminated temperature and WBC because most pa-
tients with VAC met the derangement thresholds in these categories. Thus, instead of
a category of IVAC, the pediatric criteria suggest categories of (1) pediatric VAC with
antimicrobial use (pediatric AVAC), and (2) antimicrobial VAC with a positive respira-
tory diagnostic test (pediatric PVAP). This proposal is being pilot tested in select cen-
ters in 2017 with a goal of full implementation in 2018.
There have been few studies comparing the application of these definitions. In a
single-center retrospective study of 58 children ventilated longer than 48 hours, 70
evaluations for different definitions of ventilator-associated infection were applied.
Six subjects met the 2008 CDC/NHSN pediatric VAP definition, and 5 met the
2013 adult IVAC definition. Only 1 subject satisfied both definitions.30 In another
single-center prospective study conducted over a 6-month period that included
325 invasively ventilated children, application of the new 2013 adult VAE criteria
resulted in 7 subjects classified with VAC, 6 with IVAC, and 3 with possible VAP.
By comparison, 4 had VAP based on 2008 pediatric CDC definitions.31 The investi-
gators did not comment on the overlap between definitions. In an unpublished review
of a cohort of 133 critically ill children ventilated for longer than 72 hours at our center
(Mourani, PM and Sontag, MK), we found 12 (9%) met 2013 adult VAC criteria, of
whom 5 met IVAC and 1 met possible VAP criteria. In comparison, only 5 met the
newly proposed pediatric VAC criteria (no subjects satisfied the pediatric possible
or probable VAP definition), whereas 27 subjects (20%) met the 2008 pediatric
CDC VAP criteria. Only 5 of the 12 adult VAC criteria subjects and 3 of the 5 pro-
posed pediatric VAC subjects met the 2008 pediatric CDC VAP criteria. In contrast,
22 of 27 subjects diagnosed by 2008 pediatric CDC VAP criteria met neither the 2013
adult nor the newly proposed CDC pediatric VAC criteria. Thus, these definitions
seem to identify different cohorts of children with seemingly little overlap. All seem
to be associated with worse outcomes, including increased duration of MV and
length of stay. The question remains which criteria identifies the subjects who are
most amenable to targeted prevention strategies.
 4
Mourani & Sontag
Table 1
Comparison of recent surveillance definitions for ventilator-associated pneumonia

Respiratory Microbiological
Criteria Chest Radiograph Changes Deterioration Temperature White Cell Count Antibiotic Therapy Evidence
2008 Pediatric New or increasing Combination of >38.5 C or 4000 or NA Can be used but not
Criteria infiltrates. consolidation, nonspecific <36.5 C 15,000 cells/mL required
or cavitation (need to be worsening of gas
present 2 radiographs exchange and
for patients with pre- clinical evidence of
existing lung disease), increased
consolidation secretions, cough,
and rales
2013 Adult Criteriaa NA — — — — —
VAC — Increase in FiO2 0.2 or NA NA — —
PEEP 3 cm H2O
IVAC — Increase in FiO2 0.2 or >38 or <36 C 4000 or New agent started —
PEEP 3 cm H2O 12,000 cells/mL and continued
4 d
Possible VAP — Increase in FiO2 0.2 or >38 C or <36 C 4000 or New agent started Purulent respiratory
PEEP 3 cm H2O 12,000 cells/mL and continued secretions or positive
4 d respiratory specimen
culture
 Probable VAP — Increase in FiO2 0.2 or >38 C or <36 C 4000 or New agent started Purulent respiratory
PEEP 3 cm H2O 12,000 cells/mL and continued secretions and
4 d positive quantitative
respiratory specimen
culture or positive
pleural culture, lung
histopathology,
legionella test, or
viral test
2017 Pediatric NA — NA NA — —
VAP Criteriaa
VAC — Increase in FiO2 0.25 — — — —
or MAP 4 cm H2O
AVAC — Increase in FiO2 0.25 — — New agent started —
or MAP 4 cm H2O and continued
4 d
VAP — Increase in FiO2 0.25 — — New agent started Same criteria as adult
or MAP 4 cm H2O and continued possible or probable
4 d VAP

Abbreviations: AVAC, antimicrobial VAC; MAP, mean airway pressure; NA, not applicable.
a
Patient has a baseline period of stability or improvement on the ventilator, defined by 2 or more days of stable or decreasing daily minimum FiO2, PEEP (for
adult), or MAP (for pediatric) values. The baseline period is defined as the 2 calendar days immediately preceding the first day of increased daily minimum FiO2,
PEEP (adult), or MAP (pediatric).

New VAP Paradigm

5
6 Mourani & Sontag

RISK FACTORS FOR VENTILATOR-ASSOCIATED PNEUMONIA

The airway microbiome may play a major role in regulating both bacterial pathogen
outgrowth and host immune responses.32 As such, risk factors for developing VAP pri-
marily affect airway colonization or the host response, including sedation or neuro-
muscular blockers that suppress the cough reflex, duration of MV, continuous
enteral nutrition, antibiotic exposures, immunosuppression, and foreign bodies in
the airway (bronchoscopy and replacement of the endotracheal tube).3,4,6,7,33–35
Although enteral nutrition was identified as a risk factor in some of these studies, a pro-
spective cohort study including 1245 subjects from an international group of PICUs
did not find an association between the route of delivery or duration of enteral nutrition
and VAP. However, they did find that the use of acid-suppression medications was
associated with increase in the odds of developing VAP.36
The relationship between acid-suppression and VAP has been examined in multiple
studies in adults. These studies have yielded inconsistent findings. In a case-control
study evaluating potential risk factors for VAP in adults, stress ulcer prophylaxis was
not found to be associated with risk for VAP.37 A recent randomized controlled study
of pantoprazole or placebo in ventilated adults who were suitable for enteral nutrition
found no difference in the rate of VAP.38 In contrast, a meta-analysis conducted in
2011 found that use of either proton pump inhibitor or histamine-2 receptor antagonists
increases the risk of hospital-acquired pneumonias.39 Few studies have directly exam-
ined the relationship between acid suppression and changes in the gastric microbial
content, which many have hypothesized as a possible mechanism by which acid sup-
pression alters risk for VAP. In studies of general proton pump inhibitor use, increases in
gastric pH have been associated with gastric bacterial overgrowth.40 One meta-
analysis comparing histamine-2 receptor antagonists to sucralfate for ulcer prophylaxis
in mechanically ventilated adults found that VAP was decreased among the subjects
treated with sucralfate.41 They also found that sucralfate use had lower rates of gastric
colonization (quantitative culture of at least 1 gastric specimen had more than 100 col-
ony-forming units/mL) compared with histamine-2 receptor antagonist use. In a study
examining the gastric and oropharyngeal microbiota via traditional culture techniques
in elderly subjects fed via nasogastric tubes, the investigators found remarkable simi-
larity in the bacterial communities isolated from both sites.42 Isolation of pathogenic
bacteria was correlated with higher pH of gastric specimens, suggesting that acid sup-
pression may lead to higher risk of pneumonia due to aspiration. Recent applications of
molecular microbial identification will likely advance the pace of such investigations.

THE IMPACT OF OMICS TECHNOLOGY ON INFECTION-RELATED RESEARCH

The historical approach to diagnosis and treatment of infections relies on cultivation of

the infecting organism. The 1 organism–1 disease paradigm for microbial involvement
in disease has been successful for diagnosis and treatment of many acute infectious
diseases. However, this paradigm may not adequately apply to all acute pneumonias,
especially in the context of VAP, which may occur as a superinfection to a community-
acquired bacterial or viral lower respiratory tract infection (LRTI). For critically ill chil-
dren on mechanical ventilators, targeting a single microbe isolated from a tracheal
aspirate or other lung specimen can be misleading because it overlooks the complex
process by which that organism arrived in the lower airways and its role in disease.
Study of the microbiome, the genomic content of the microbiota, is a novel
approach to expand understanding of the complexities of human environments nor-
mally colonized with microbial communities, such as the gut and the lung. Although
genomics is the study of the DNA content of an organism, metagenomics applies
 New VAP Paradigm 7

genomic technologies and bioinformatics tools to study all the genomes recovered
from a specific environmental sample. In reference to human health, this emerging
powerful set of tools analyzes nucleic acids isolated from specimens collected from
organ systems to provide simultaneous and sensitive quantification of the bacterial,
fungal, and viral constituents of the microbiome.43–48 One approach to identification
of the microbial constituents of a sample is called shotgun sequencing. With this tech-
nique, DNA extracted from a sample is randomly sheared, and the resulting short se-
quences (>700 base pairs) are deciphered and then reconstructed into a consensus
sequence that is aligned to known genomic databases to identify the organism from
which it was extracted. Next-generation sequencing (NGS), using platforms such as
the Illumina MiSeq or HiSeq (Illumina Inc, San Diego, CA), generate much shorter
DNA fragments (<700 base pairs), but the limitation is offset by the significantly larger
number of sequence reads generated. NGS is now the most often used technology
because it has become the most economical choice. Many studies of human health
have focused only on the bacterial constituents of the microbiota, acknowledging
the limited contribution of viruses and fungi in many of the human compartments.
For these studies, an even cheaper and more rapid sequencing alternative to NGS en-
tails sequencing of the 16S rRNA gene of bacteria, which is highly conserved between
different species but also contains hypervariable regions that presents species-
specific signature sequences for identification of bacteria.
There are several bioinformatics processes that are required to decipher the large
data generated by these technologies. For example, a human gut microbiome project
identified 3.3 million genes assembled from more than 565 gigabases.49 The initial
sequence reads contain many redundant, low-quality, and human sequences that
need to be removed in a filtering step before being inserted in to assembly programs
that constructs the reads into specific genomes. Once assembled into genomes, the
data are moved through informatics pipelines to determine which sequences belong
to which species from the sample. Identification is performed through 2 main tech-
niques. The first matches sequences to those that are already publicly available in
sequence databases. For microbes that lack matching homologous sequences in
these databases, the second method uses intrinsic characteristics of the sequences
to predict the coding regions of bacterial species.
Many diseases have now been associated with changes in composition of bacterial
communities in the gut.50–52 Specifically, microbiota of limited diversity and relative
absence of commensal organisms are associated with increased inflammation, barrier
permeability, and disease states.53,54 Metagenomic techniques applied to investiga-
tions of the lung have demonstrated that the lungs of healthy children and adults
are rich with bacteria, contradicting the traditional theory that human lower airways
are sterile.9,10,12 These studies also demonstrate that microbial communities play an
important role in human physiology.55–58 The gut microbiota in children has been
shown to have a strong influence on the development and function of the immune sys-
tem and has been postulated to modulate the risk of subsequent asthma risk in infants
infected with respiratory syncytial virus.59
In addition to molecular detection of microbial communities, metagenomics can
facilitate investigations of the host response to the microbiome by providing simulta-
neous evaluations of both the microbial and human messenger RNA (mRNA), also
known as the transcriptome. Further, comprehensive proteomic and metabolomic
evaluations have the potential to add considerable depth to these investigations by
not only detecting the presence of organisms but also by describing what they are do-
ing and how they are interacting with each other and the host.60 Proteomics refers to
the study of the structure and function of proteins within cells, tissues, and organisms.
8 Mourani & Sontag

Analogous to the microbiome, the proteome refers to all expressed proteins in a spe-
cific environment, typically a cell, tissue, or organism. All organisms produce small
molecules (<1200 Da), known as metabolites, that are by-products of life-sustaining
biochemical process. Metabolomics is the study of these small molecules in relation
to the physiology or pathophysiology of organisms. Mass spectroscopy is the usual
method to identify proteins and metabolites. In this method, ions are generated
from proteins or metabolites, which can be used to separate them according to their
mass-to-charge ratio to create a mass spectrum. Several analytical techniques, each
with their respective advantages and limitations may be used to identify specific pro-
teins or metabolites. Application of these strategies to investigations of the intestinal
tract generated the concept that dysbiosis leads to generation of disease-
promoting microbiome or pathobiome,61 increasing risk for infection, as well as pro-
moting perturbations of the immune system and persistent organ dysfunction.60
Combining these approaches simultaneously has been used in a study of subjects
with human immunodeficiency virus and pneumonia, in which the investigators iden-
tified that the composition of the lower airway microbiome is associated with discrete
local host immune responses and systemic metabolic signatures.32 Further, these
complex phenotypes were directly associated with mortality.
Translating these approaches to interrogate the relationships between an intubated
individual’s respiratory microbiota, the host’s immune response, and the underlying
disease process will likely provide important insights into the pathogenesis of VAP,
help identify at-risk children early in the course of illness, and suggest new and
more effective prevention and therapeutic strategies to improve outcomes and reduce
resource utilization and costs.

MICROBIAL INTERACTIONS AFFECTING RISK FOR VENTILATOR-ASSOCIATED

PNEUMONIA

Emerging evidence about how organisms interact with each other (including viral-
bacterial interactions), environmental elements, and the host have shed light on the
enormous complexity of infection pathogenesis. Microorganisms have evolved the
ability to detect local environmental signals such pH, metabolites, microbial popula-
tion density (quorum sensing), and host immune cells.62–64 These sensing mecha-
nisms allow pathogens to regulate expression of their virulence factors, masking
infection until the advantage to them is greatest. Bacteria can also express structural
appendages and secrete products that bind pathogen-recognition receptors on
epithelial cells, which then transduce specific downstream pathways that impair local
and systemic immune function.65 Critical illness alone has been demonstrated to alter
the intestinal microbiome by diminishing populations and activity of commensal bac-
teria.66 Pathogen populations then proliferate as a result of unimpeded strategies to
either evade detection of the host or render immune cells ineffective.67 The impact
of changes in the microbiome and bacterial virulence has not yet been routinely incor-
porated into the approach to infection.68
Numerous studies have demonstrated that viral LRTIs are frequently associated
with bacterial pneumonia, including VAP and invasive bacterial infections.69–76 The
highest rates of bacterial coinfections occur in patients admitted to the ICU.73 Pneu-
mococcal vaccine provides protection against influenza-related hospitalization,77
suggesting an important role for the bacterial infection in determining the severity of
illness in viral LRTI. Among children admitted to the PICU with influenza during the
2009 pandemic, those with Staphylococcus aureus coinfection had a higher mortality
risk.78 Rhinovirus infection in chronic obstructive pulmonary disease (COPD) patients
 New VAP Paradigm 9

is associated with an increase in the bacterial load of Haemophilus influenza, suggest-

ing that viral infection increases risk of secondary bacterial pneumonia.79
Historically, this complication of viral infection has been attributed to degradation of
epithelial barrier function by direct cytopathologic effects of the virus. Recent work,
however, has strongly implicated endogenous immune suppressive pathways that limit
inflammation-mediated damage and may permit bacterial outgrowth from the micro-
biota into clinical infection.80–84 Understanding the delicate balance between the im-
mune activation required to clear infections and the immune suppression needed to
reduce excessive inflammation-mediated damage is critically important for targeting
new strategies to reduce VAP and its associated morbidity and mortality (Fig. 1).

THE MICROBIOME AND HOST IMMUNE RESPONSE

The microbiome has a critical role in immune activation and host defense against in-
fections. The gut microbiome is crucial for priming of defenses against infection not
only in the gastrointestinal tract but also in distant organs, including the lung.83,85
Inflammasomes, receptors that regulate the activation of caspase-1 and propagate in-
flammatory responses to infectious organisms and molecules derived from the host,
are recognized to be vital elements of host defense and innate immunity. Recent ev-
idence suggests that systemic priming of inflammasome activation by endogenous
microbes changes the set point of innate immune responses to respiratory infection,
in effect priming host immune response.83,86 This process involves activation of
signaling via toll-like receptors, which then leads to production of type I interferon
(INF) and the precursor molecules pro-interleukin (IL)-1b and pro-IL-18. A second
insult, perhaps other products of damaged tissues, then activates caspase-1 and
caspase-mediated cleavage of the pro-forms into active IL-1b and IL-18, which are
released from the cell and upregulate several key cytokines and chemokines (a sub-
family of cytokines that serve as chemotactic factors for different cell types). Animal
models suggest that viral infections suppress immune responses and increase risk

Fig. 1. Endogenous airway bacteria achieve a balance between host immune activation and
suppression. Deleterious shifts in the airway ecology likely lead to VAP. Viral infection (and
many other triggering factors) may initiate these shifts by suppressing host immune re-
sponses that control the bacterial populations and prevent emergence of pathogens.
CXCL, of chemokine ligand; IL, interleukin; INF, interferon.
10 Mourani & Sontag

of bacterial infection. Influenza virus infections induce highly polarized type 1 T helper
cell responses, producing large amounts of INF-g and IL-10.84 In mouse models,
INF-g contributes to postviral immune suppression by impeding alveolar macrophage
function, leading to reduced bacterial clearance.87 Influenza-mediated type I INF pro-
duction also attenuates production of chemokine ligand (CXCL)-1 and CXCL2, impair-
ing neutrophil responses and predisposing to secondary bacterial infection.88 CD200
ligand is a membrane glycoprotein expressed on epithelial cells or other apoptotic
cells that activates CD200 receptor (R) on airway macrophages, suppressing their
activity. Expression of CD200 and CD200R are increased in experimental animals
with viral respiratory infection, and CD200R knockout mice are less susceptible to
postviral bacterial infection, suggesting that the CD200 system is a critical mediator
of postviral immune suppression.89 Commensal microbes, in particular Staphylo-
coccal species, attenuate the immune-mediated lung injury and cytokine storm trig-
gered by influenza infection by promoting the M2 immune-suppressive macrophage
phenotype in airway and alveolar macrophages, leading to increased production of
IL-10 and transforming growth factor (TGF)-b, and improved survival.90 IL-10 and
TGF-b have also been implicated in bacterial superinfection in other studies.91,92 In
addition to priming the immune system to prevent infection, the airway microbiota
may also modulate the response to established infection. Microbial interactions with
the host immune system are quite complex and may also depend on timing of expo-
sure. For example, as these data suggest, Staphylococcus may prime the immune
system to protect the host from influenza infection, but other studies (as previously
noted) suggest that influenza and Staphylococcal coinfections portend a higher risk
of mortality. These relationships deserve further study.

APPLICATION OF THE NEW INFECTION PARADIGM TO VENTILATOR-ASSOCIATED

PNEUMONIA

Ventilator-associated tracheobronchitis can be an intermediate step between airway

dysbiosis and the development of VAP,93–95 suggesting that investigations of the
changing ecology of the large airways in relation to infection are likely to augment un-
derstanding of the pathogenesis of VAP. The gold standard VAP diagnostic method of
biopsy and direct culture of lung tissue is not possible for most ventilated children due
to its invasive nature and risks. Yet, traditional endotracheal aspirate culture tech-
niques are imprecise and have led to wide-spread antibiotic use, a practice which
could select for drug-resistant pathogens in the airway and in other body compart-
ments.96–98 VAP is the most common indication for antibiotic use in the PICU, ac-
counting for almost half of all antibiotic days.8 Yet, rather than mitigating the
adverse impact of VAP, present approaches to clinical cultures and antibiotic use
may actually increase risk for VAP on an individual and unit level by depleting protec-
tive commensal organisms and selecting for antibiotic resistant pathogens. The
increasing rates of antimicrobial resistance among VAP pathogens now lead many
clinicians to empirically treat suspected VAP patients with a combination of broad-
spectrum antibiotics, which likely perpetuates the cycle of increasing antibiotic resis-
tance and persistent dysbiosis of the host microbiome.
To break this adverse cycle, a new paradigm is desperately needed. Moving beyond
traditional culture methodologies to simultaneous employment of metagenomics, pro-
teomics, and metabolomics will arm clinicians with critical information that has been
previously unavailable. When the host organism is disturbed by disease, the profile
of proteome and metabolome often changes. These changes may be in response to
chronic disease that may render the host susceptible to infection or to a primary
 New VAP Paradigm 11

infection that may then promote organ dysfunction. Thus, simultaneous study of the
proteome and metabolome along with the microbiome, may enable the discovery of
biomarkers that further understanding of predisease and disease states. These data
may better inform when infection is present and by what pathogens, as well as identify
the risk before infection is present and point to strategies that can effectively prevent
infection by restoring commensal microbial communities, strengthening the host im-
mune response, and suppressing emergence of the detrimental pathobiome.
Early evaluations of the respiratory tract microbiota in pulmonary diseases, such as
cystic fibrosis, asthma, COPD, and lung transplant,9–11,48,99–105 reveal shifts in bacte-
rial ecology compared with healthy patients, suggesting that the composition of the
respiratory tract microbiota plays a direct role in pathophysiology of lung disease.
Based on these observations, it is not unreasonable to assume that similar conclu-
sions can be drawn for mechanically ventilated patients at risk for VAP. In fact, early
application of molecular detection methods to mechanically ventilated ICU patients
suggests this may be the case.106 The authors have been evaluating changes in the
airway microbiome in mechanically ventilated children. We have found that airway
samples collected on intubation exhibit a relatively diverse bacterial community; how-
ever, diversity decreases in samples collected in subsequent days and the airways
can rapidly become dominated by pathogenic bacteria before diagnosis of VAP
(Fig. 2). Identification of diminishing diversity and emergence of potential pathogens

Fig. 2. Bacterial composition of tracheal aspirate samples in mechanically ventilated chil-

dren as determined through 16S rRNA detection. Each bar represents a day of ventilation
with relative abundance of organisms at the genus level described by the different colors.
Tracheal aspirate samples are relatively diverse on intubation but may decrease over time
with pathogenic bacteria dominating the community before diagnosis of VAP, Staph,
Staphylococcus.
12 Mourani & Sontag

before infections occur may allow caregivers the opportunity to modulate the micro-
biota to increase diversity and/or limit specific pathogen outgrowth. Yet, the variability
in community composition, the virulence of the pathogens involved, and the host re-
sponses belies the concept that a single approach will universally apply to all patients.
Application of this new paradigm, incorporating simultaneous assessments of micro-
bial populations and their activity, as well as the host response, may enable a person-
alized strategy, uniquely suited to individual patients.

SUMMARY

Despite some evidence that VAP incidence is decreasing, other data suggest that it
remains a common and detrimental complication of MV support in children and is
the main indication for antibiotic use in the PICU. The limitations of traditional microbial
culture techniques and the lack of reliable and consistently applied diagnostic criteria
have hampered the progress of understanding the pathogenesis of VAP and, thus,
truly effective strategies to prevent this nosocomial infection. However, evolution of
metagenomic and proteomic technologies has opened the doors to new investiga-
tions that have changed the paradigm by which infections and their impact on human
health are understood. These studies have demonstrated that VAP likely results from a
complex interaction of microbes, the environment, and the host immune response.
Future studies evaluating the mechanisms of these interactions will improve under-
standing of the pathogenesis of VAP (and pneumonia, in general), enable identification
of at-risk patients early in the course of MV, and suggest specific preventive and ther-
apeutic interventions to improve outcomes in this high-risk population.

REFERENCES

1. Raymond J, Aujard Y. Nosocomial infections in pediatric patients: a European,

multicenter prospective study. European Study Group. Infect Control Hosp Epi-
demiol 2000;21(4):260–3.
2. Foglia EE, Fraser VJ, Elward AM. Effect of nosocomial infections due to
antibiotic-resistant organisms on length of stay and mortality in the pediatric
intensive care unit. Infect Control Hosp Epidemiol 2007;28(3):299–306.
3. Srinivasan R, Asselin J, Gildengorin G, et al. A prospective study of ventilator-
associated pneumonia in children. Pediatrics 2009;123(4):1108–15.
4. Elward AM, Warren DK, Fraser VJ. Ventilator-associated pneumonia in pediatric
intensive care unit patients: risk factors and outcomes. Pediatrics 2002;109(5):
758–64.
5. Bigham MT, Amato R, Bondurrant P, et al. Ventilator-associated pneumonia in
the pediatric intensive care unit: characterizing the problem and implementing
a sustainable solution. J Pediatr 2009;154(4):582–7.e2.
6. Gupta S, Boville BM, Blanton R, et al. A multicentered prospective analysis of
diagnosis, risk factors, and outcomes associated with pediatric ventilator-
associated pneumonia. Pediatr Crit Care Med 2015;16(3):e65–73.
7. Almuneef M, Memish ZA, Balkhy HH, et al. Ventilator-associated pneumonia in a
pediatric intensive care unit in Saudi Arabia: a 30-month prospective surveil-
lance. Infect Control Hosp Epidemiol 2004;25(9):753–8.
8. Fischer JE, Ramser M, Fanconi S. Use of antibiotics in pediatric intensive care
and potential savings. Intensive Care Med 2000;26(7):959–66.
9. Hilty M, Burke C, Pedro H, et al. Disordered microbial communities in asthmatic
airways. PLoS One 2010;5(1):e8578.
 New VAP Paradigm 13

10. Sze MA, Dimitriu PA, Hayashi S, et al. The lung tissue microbiome in chronic
obstructive pulmonary disease. Am J Respir Crit Care Med 2012;185(10):
1073–80.
11. Erb-Downward JR, Thompson DL, Han MK, et al. Analysis of the lung micro-
biome in the “healthy” smoker and in COPD. PLoS One 2011;6(2):e16384.
12. Charlson ES, Bittinger K, Haas AR, et al. Topographical continuity of bacterial
populations in the healthy human respiratory tract. Am J Respir Crit Care Med
2011;184(8):957–63.
13. Foglia E, Meier MD, Elward A. Ventilator-associated pneumonia in neonatal and
pediatric intensive care unit patients. Clin Microbiol Rev 2007;20(3):409–25.
14. Venkatachalam V, Hendley JO, Willson DF. The diagnostic dilemma of ventilator-
associated pneumonia in critically ill children. Pediatr Crit Care Med 2011;12(3):
286–96.
15. Centers for Disease Control and Prevention guidelines: guidelines and proced-
ures for monitoring VAP. 2013. Available at: http://www.cdc.gov/nhsn/PDFs/
pscManual/6pscVAPcurrent.pdf. Accessed September 6, 2013.
16. Gauvin F, Dassa C, Chaı̈bou M, et al. Ventilator-associated pneumonia in intu-
bated children: comparison of different diagnostic methods. Pediatr Crit Care
Med 2003;4(4):437–43.
17. Labenne M, Poyart C, Rambaud C, et al. Blind protected specimen brush and
bronchoalveolar lavage in ventilated children. Crit Care Med 1999;27(11):
2537–43.
18. Gauvin F, Lacroix J, Guertin MC, et al. Reproducibility of blind protected bron-
choalveolar lavage in mechanically ventilated children. Am J Respir Crit Care
Med 2002;165(12):1618–23.
19. Zucker A, Pollack M, Katz R. Blind use of the double-lumen plugged catheter for
diagnosis of respiratory tract infections in critically ill children. Crit Care Med
1984;12(10):867–70.
20. Beyersmann J, Gastmeier P, Grundmann H, et al. Use of multistate models to
assess prolongation of intensive care unit stay due to nosocomial infection.
Infect Control Hosp Epidemiol 2006;27(5):493–9.
21. Safdar N, Dezfulian C, Collard HR, et al. Clinical and economic consequences
of ventilator-associated pneumonia: a systematic review. Crit Care Med 2005;
33(10):2184–93.
22. Suka M, Yoshida K, Uno H, et al. Incidence and outcomes of ventilator-
associated pneumonia in Japanese intensive care units: the Japanese nosoco-
mial infection surveillance system. Infect Control Hosp Epidemiol 2007;28(3):
307–13.
23. Cordero L, Ayers LW, Miller RR, et al. Surveillance of ventilator-associated pneu-
monia in very-low-birth-weight infants. Am J Infect Control 2002;30(1):32–9.
24. Emori TG, Edwards JR, Culver DH, et al. Accuracy of reporting nosocomial in-
fections in intensive-care-unit patients to the National Nosocomial Infections Sur-
veillance system: a pilot study. Infect Control Hosp Epidemiol 1998;19(5):
308–16.
25. Patrick SW, Kawai AT, Kleinman K, et al. Health care-associated infections
among critically ill children in the US, 2007-2012. Pediatrics 2014;134(4):
705–12.
26. Metersky ML, Wang Y, Klompas M, et al. Trend in ventilator-associated pneu-
monia rates between 2005 and 2013. JAMA 2016;316(22):2427–9.
27. Magill SS, Klompas M, Balk R, et al. Developing a new, national approach to sur-
veillance for ventilator-associated events*. Crit Care Med 2013;41(11):2467–75.
14 Mourani & Sontag

28. Cocoros NM, Kleinman K, Priebe GP, et al. Ventilator-associated events in neo-
nates and children–a new paradigm. Crit Care Med 2016;44(1):14–22.
29. Cocoros NM, Priebe GP, Logan LK, et al. A pediatric approach to ventilator-
associated events surveillance. Infect Control Hosp Epidemiol 2017;38(3):
327–33.
30. Beardsley AL, Nitu ME, Cox EG, et al. An evaluation of various ventilator-
associated infection criteria in a PICU. Pediatr Crit Care Med 2016;17(1):73–80.
31. Narayanan A, Dixon G, Chalkley S, et al. Ventilator-associated pneumonia in
children: comparing the plethora of surveillance definitions. J Hosp Infect
2016;94(2):163–4.
32. Shenoy MK, Iwai S, Lin DL, et al. Immune response and mortality risk relate to
distinct lung microbiomes in patients with HIV and pneumonia. Am J Respir
Crit Care Med 2017;195(1):104–14.
33. Fayon MJ, Tucci M, Lacroix J, et al. Nosocomial pneumonia and tracheitis in a
pediatric intensive care unit: a prospective study. Am J Respir Crit Care Med
1997;155(1):162–9.
34. Wright ML, Romano MJ. Ventilator-associated pneumonia in children. Semin Pe-
diatr Infect Dis 2006;17(2):58–64.
35. Bochicchio GV, Sung J, Joshi M, et al. Persistent hyperglycemia is predictive of
outcome in critically ill trauma patients. J Trauma 2005;58(5):921–4.
36. Albert BD, Zurakowski D, Bechard LJ, et al. Enteral nutrition and acid-
suppressive therapy in the PICU: impact on the risk of ventilator-associated
pneumonia. Pediatr Crit Care Med 2016;17(10):924–9.
37. Lewis SC, Li L, Murphy MV, et al. Risk factors for ventilator-associated events: a
case-control multivariable analysis. Crit Care Med 2014;42(8):1839–48.
38. Selvanderan SP, Summers MJ, Finnis ME, et al. Pantoprazole or placebo for
stress ulcer prophylaxis (POP-UP): randomized double-blind exploratory study.
Crit Care Med 2016;44(10):1842–50.
39. Eom CS, Jeon CY, Lim JW, et al. Use of acid-suppressive drugs and risk of
pneumonia: a systematic review and meta-analysis. CMAJ 2011;183(3):310–9.
40. Freedberg DE, Lebwohl B, Abrams JA. The impact of proton pump inhibitors on
the human gastrointestinal microbiome. Clin Lab Med 2014;34(4):771–85.
41. Huang J, Cao Y, Liao C, et al. Effect of histamine-2-receptor antagonists versus
sucralfate on stress ulcer prophylaxis in mechanically ventilated patients: a
meta-analysis of 10 randomized controlled trials. Crit Care 2010;14(5):R194.
42. Segal R, Pogoreliuk I, Dan M, et al. Gastric microbiota in elderly patients fed via
nasogastric tubes for prolonged periods. J Hosp Infect 2006;63(1):79–83.
43. Bexfield N, Kellam P. Metagenomics and the molecular identification of novel vi-
ruses. Vet J 2011;190(2):191–8.
44. Delwart EL. Viral metagenomics. Rev Med Virol 2007;17(2):115–31.
45. Haagmans BL, Andeweg AC, Osterhaus AD. The application of genomics to
emerging zoonotic viral diseases. PLoS Pathog 2009;5(10):e1000557.
46. Tang P, Chiu C. Metagenomics for the discovery of novel human viruses. Future
Microbiol 2010;5(2):177–89.
47. Mourani PM, Harris JK, Sontag MK, et al. Molecular identification of bacteria in
tracheal aspirate fluid from mechanically ventilated preterm infants. PLoS One
2011;6(10):e25959.
48. Wootton SC, Kim DS, Kondoh Y, et al. Viral infection in acute exacerbation of
idiopathic pulmonary fibrosis. Am J Respir Crit Care Med 2011;183(12):
1698–702.
 New VAP Paradigm 15

49. Qin J, Li R, Raes J, et al. A human gut microbial gene catalogue established by
metagenomic sequencing. Nature 2010;464(7285):59–65.
50. de La Serre CB, Ellis CL, Lee J, et al. Propensity to high-fat diet-induced obesity
in rats is associated with changes in the gut microbiota and gut inflammation.
Am J Physiol Gastrointest Liver Physiol 2010;299(2):G440–8.
51. van Vliet MJ, Harmsen HJ, de Bont ES, et al. The role of intestinal microbiota in
the development and severity of chemotherapy-induced mucositis. PLoS
Pathog 2010;6(5):e1000879.
52. Yang L, Lu X, Nossa CW, et al. Inflammation and intestinal metaplasia of the
distal esophagus are associated with alterations in the microbiome. Gastroen-
terology 2009;137(2):588–97.
53. Cani PD, Possemiers S, Van de Wiele T, et al. Changes in gut microbiota control
inflammation in obese mice through a mechanism involving GLP-2-driven
improvement of gut permeability. Gut 2009;58(8):1091–103.
54. Dimmitt RA, Staley EM, Chuang G, et al. Role of postnatal acquisition of the in-
testinal microbiome in the early development of immune function. J Pediatr Gas-
troenterol Nutr 2010;51(3):262–73.
55. Suerbaum S. Microbiome analysis in the esophagus. Gastroenterology 2009;
137(2):419–21.
56. Li P, Hotamisligil GS. Metabolism: host and microbes in a pickle. Nature 2010;
464(7293):1287–8.
57. Finkbeiner SR, Allred AF, Tarr PI, et al. Metagenomic analysis of human diarrhea:
viral detection and discovery. PLoS Pathog 2008;4(2):e1000011.
58. Kistler A, Avila PC, Rouskin S, et al. Pan-viral screening of respiratory tract infec-
tions in adults with and without asthma reveals unexpected human coronavirus
and human rhinovirus diversity. J Infect Dis 2007;196(6):817–25.
59. Lynch JP, Sikder MA, Curren BF, et al. The influence of the microbiome on early-
life severe viral lower respiratory infections and asthma-food for thought? Front
Immunol 2017;8:156.
60. Alverdy JC, Krezalek MA. Collapse of the microbiome, emergence of the path-
obiome, and the immunopathology of sepsis. Crit Care Med 2017;45(2):337–47.
61. Gilbert JA, Quinn RA, Debelius J, et al. Microbiome-wide association studies link
dynamic microbial consortia to disease. Nature 2016;535(7610):94–103.
62. Zaborina O, Lepine F, Xiao G, et al. Dynorphin activates quorum sensing quino-
lone signaling in Pseudomonas aeruginosa. PLoS Pathog 2007;3(3):e35.
63. Kohler JE, Zaborina O, Wu L, et al. Components of intestinal epithelial hypoxia
activate the virulence circuitry of pseudomonas. Am J Physiol Gastrointest Liver
Physiol 2005;288(5):G1048–54.
64. Patel NJ, Zaborina O, Wu L, et al. Recognition of intestinal epithelial HIF-1alpha
activation by pseudomonas aeruginosa. Am J Physiol Gastrointest Liver Physiol
2007;292(1):G134–42.
65. Schreiber F, Arasteh JM, Lawley TD. Pathogen resistance mediated by IL-22
signaling at the epithelial-microbiota interface. J Mol Biol 2015;427(23):3676–82.
66. Dickson RP. The microbiome and critical illness. Lancet Respir Med 2016;4(1):
59–72.
67. Koeppen K, Hampton TH, Jarek M, et al. A novel mechanism of host-pathogen
interaction through sRNA in bacterial outer membrane vesicles. PLoS Pathog
2016;12(6):e1005672.
68. Klingensmith NJ, Coopersmith CM. The gut as the motor of multiple organ
dysfunction in critical illness. Crit Care Clin 2016;32(2):203–12.
16 Mourani & Sontag

69. Cherian T, Simoes EA, Steinhoff MC, et al. Bronchiolitis in tropical south India.
Am J Dis Child 1990;144(9):1026–30.
70. Thorburn K, Harigopal S, Reddy V, et al. High incidence of pulmonary bacterial
co-infection in children with severe respiratory syncytial virus (RSV) bronchioli-
tis. Thorax 2006;61(7):611–5.
71. Stockman LJ, Reed C, Kallen AJ, et al. Respiratory syncytial virus and Staphy-
lococcus aureus coinfection in children hospitalized with pneumonia. Pediatr
Infect Dis J 2010;29(11):1048–50.
72. Resch B, Gusenleitner W, Mueller WD. Risk of concurrent bacterial infection in
preterm infants hospitalized due to respiratory syncytial virus infection. Acta
Paediatr 2007;96(4):495–8.
73. Hon KL, Leung E, Tang J, et al. Premorbid factors and outcome associated with
respiratory virus infections in a pediatric intensive care unit. Pediatr Pulmonol
2008;43(3):275–80.
74. Bezerra PG, Britto MC, Correia JB, et al. Viral and atypical bacterial detection in
acute respiratory infection in children under five years. PLoS One 2011;6(4):
e18928.
75. Madhi SA, Ludewick H, Kuwanda L, et al. Pneumococcal coinfection with hu-
man metapneumovirus. J Infect Dis 2006;193(9):1236–43.
76. Hodson A, Kasliwal M, Streetly M, et al. A parainfluenza-3 outbreak in a SCT unit:
sepsis with multi-organ failure and multiple co-pathogens are associated with
increased mortality. Bone Marrow Transplant 2011;46(12):1545–50.
77. Madhi SA, Klugman KP, Vaccine Trialist Group. A role for Streptococcus pneu-
moniae in virus-associated pneumonia. Nat Med 2004;10(8):811–3.
78. Randolph AG, Vaughn F, Sullivan R, et al. Critically ill children during the 2009-
2010 influenza pandemic in the United States. Pediatrics 2011;128(6):e1450–8.
79. Molyneaux PL, Mallia P, Cox MJ, et al. Outgrowth of the bacterial airway micro-
biome after rhinovirus exacerbation of chronic obstructive pulmonary disease.
Am J Respir Crit Care Med 2013;188(10):1224–31.
80. Avadhanula V, Rodriguez CA, Devincenzo JP, et al. Respiratory viruses augment
the adhesion of bacterial pathogens to respiratory epithelium in a viral species-
and cell type-dependent manner. J Virol 2006;80(4):1629–36.
81. McAuley JL, Hornung F, Boyd KL, et al. Expression of the 1918 influenza A virus
PB1-F2 enhances the pathogenesis of viral and secondary bacterial pneu-
monia. Cell Host Microbe 2007;2(4):240–9.
82. Hrincius ER, Hennecke AK, Gensler L, et al. A single point mutation (Y89F)
within the non-structural protein 1 of influenza A viruses limits epithelial cell
tropism and virulence in mice. Am J Pathol 2012;180(6):2361–74.
83. Ichinohe T, Pang IK, Kumamoto Y, et al. Microbiota regulates immune defense
against respiratory tract influenza A virus infection. Proc Natl Acad Sci U S A
2011;108(13):5354–9.
84. Braciale TJ, Sun J, Kim TS. Regulating the adaptive immune response to respi-
ratory virus infection. Nat Rev Immunol 2012;12(4):295–305.
85. Prakash A, Sundar SV, Zhu YG, et al. Lung ischemia-reperfusion is a sterile in-
flammatory process influenced by commensal microbiota in mice. Shock 2015;
44(3):272–9.
86. Ganal SC, Sanos SL, Kallfass C, et al. Priming of natural killer cells by nonmu-
cosal mononuclear phagocytes requires instructive signals from commensal mi-
crobiota. Immunity 2012;37(1):171–86.
87. Sun K, Metzger DW. Inhibition of pulmonary antibacterial defense by interferon-
gamma during recovery from influenza infection. Nat Med 2008;14(5):558–64.
 New VAP Paradigm 17

88. Shahangian A, Chow EK, Tian X, et al. Type I IFNs mediate development of
postinfluenza bacterial pneumonia in mice. J Clin Invest 2009;119(7):
1910–20.
89. Goulding J, Godlee A, Vekaria S, et al. Lowering the threshold of lung innate im-
mune cell activation alters susceptibility to secondary bacterial superinfection.
J Infect Dis 2011;204(7):1086–94.
90. Wang J, Li F, Sun R, et al. Bacterial colonization dampens influenza-mediated
acute lung injury via induction of M2 alveolar macrophages. Nat Commun
2013;4:2106.
91. van der Sluijs KF, van Elden LJ, Nijhuis M, et al. IL-10 is an important mediator of
the enhanced susceptibility to pneumococcal pneumonia after influenza infec-
tion. J Immunol 2004;172(12):7603–9.
92. Neill DR, Fernandes VE, Wisby L, et al. T regulatory cells control susceptibility to
invasive pneumococcal pneumonia in mice. PLoS Pathog 2012;8(4):e1002660.
93. Nseir S, Ader F, Marquette CH. Nosocomial tracheobronchitis. Curr Opin Infect
Dis 2009;22(2):148–53.
94. Craven DE, Chroneou A, Zias N, et al. Ventilator-associated tracheobronchitis:
the impact of targeted antibiotic therapy on patient outcomes. Chest 2009;
135(2):521–8.
95. Wheeler DS, Whitt JD, Lake M, et al. A case-control study on the impact of
ventilator-associated tracheobronchitis in the PICU. Pediatr Crit Care Med
2015;16(6):565–71.
96. Willson DF, Conaway M, Kelly R, et al. The lack of specificity of tracheal aspi-
rates in the diagnosis of pulmonary infection in intubated children. Pediatr Crit
Care Med 2014;15(4):299–305.
97. Willson DF, Kirby A, Kicker JS. Respiratory secretion analyses in the evaluation
of ventilator-associated pneumonia: a survey of current practice in pediatric crit-
ical care. Pediatr Crit Care Med 2014;15(8):715–9.
98. Singh N, Rogers P, Atwood CW, et al. Short-course empiric antibiotic therapy for
patients with pulmonary infiltrates in the intensive care unit. A proposed solution
for indiscriminate antibiotic prescription. Am J Respir Crit Care Med 2000;162(2
Pt 1):505–11.
99. Harris JK, De Groote MA, Sagel SD, et al. Molecular identification of bacteria in
bronchoalveolar lavage fluid from children with cystic fibrosis. Proc Natl Acad
Sci U S A 2007;104(51):20529–33.
100. Klepac-Ceraj V, Lemon KP, Martin TR, et al. Relationship between cystic fibrosis
respiratory tract bacterial communities and age, genotype, antibiotics and
Pseudomonas aeruginosa. Environ Microbiol 2010;12(5):1293–303.
101. Zemanick ET, Wagner BD, Sagel SD, et al. Reliability of quantitative real-time
PCR for bacterial detection in cystic fibrosis airway specimens. PLoS One
2010;5(11):e15101.
102. Bisgaard H, Hermansen MN, Buchvald F, et al. Childhood asthma after
bacterial colonization of the airway in neonates. N Engl J Med 2007;357(15):
1487–95.
103. Ege MJ, Mayer M, Normand AC, et al. Exposure to environmental microorgan-
isms and childhood asthma. N Engl J Med 2011;364(8):701–9.
104. Charlson ES, Diamond JM, Bittinger K, et al. Lung-enriched organisms and
aberrant bacterial and fungal respiratory microbiota after lung transplant. Am
J Respir Crit Care Med 2012;186(6):536–45.
18 Mourani & Sontag

105. Goleva E, Jackson LP, Harris JK, et al. The effects of airway microbiome on corti-
costeroid responsiveness in asthma. Am J Respir Crit Care Med 2013;188(10):
1193–201.
106. Kelly BJ, Imai I, Bittinger K, et al. Composition and dynamics of the respiratory
tract microbiome in intubated patients. Microbiome 2016;4:7.