:
AMERICAN THORACIC SOCIETYDOCUMENTS
Official American Thoracic Society Clinical Practice Guidelines:Diagnostic Evaluation of Infants with Recurrent orPersistent WheezingClement L. Ren*, Charles R. Esther, Jr.*, Jason S. Debley, Marianna Sockrider, Ozge Yilmaz, Nikhil Amin,Alia Bazzy-Asaad, Stephanie D. Davis, Manuel Durand, Jeffrey M. Ewig, Hasan Yuksel, Enrico Lombardi, Terry L. Noah,Peggy Radford, Sarath Ranganathan, Alejandro Teper, Miles Weinberger, Jan Brozek, and Kevin C. Wilson; on behalfof the ATS Ad Hoc Committee on Infants with Recurrent or Persistent Wheezing
THIS OFFICIAL CLINICAL PRACTICE GUIDELINE OF THE AMERICAN THORACIC SOCIETY (ATS) WAS APPROVED BY THE ATS BOARD OF DIRECTORS, MARCH 2016
Background: Infantile wheezing is a common problem, but there areno guidelines for the evaluation of infants with recurrent or persistentwheezing that is not relieved or prevented by standard therapies.
Methods: An American Thoracic Society–sanctioned guidelinedevelopment committee selected clinical questions related touncertainties or controversies in the diagnostic evaluation ofwheezing infants. Members of the committee conductedpragmatic evidence syntheses, which followed the Grading ofRecommendations, Assessment, Development, and Evaluation(GRADE) approach. The evidence syntheses were used to informthe formulation and grading of recommendations.
Results: The pragmatic evidence syntheses identified few studiesthat addressed the clinical questions. The studies that were identifiedconstituted very low-quality evidence, consisting almost exclusivelyof case series with risk of selection bias, indirect patient populations,
and imprecise estimates. The committee made conditionalrecommendations to perform bronchoscopic airway survey,bronchoalveolar lavage, esophageal pHmonitoring, and a swallowingstudy. It alsomade conditional recommendations against empiric foodavoidance, upper gastrointestinal radiography, and gastrointestinalscintigraphy. Finally, the committee recommended additionalresearch about the roles of infant pulmonary function testing and foodavoidance or dietary changes, based on allergy testing.
Conclusions: Although infantile wheezing is common, there is apaucity of evidence to guide clinicians in selecting diagnostic testsfor recurrent or persistent wheezing. Our committee made severalconditional recommendations to guide clinicians; however, additionalresearch that measures clinical outcomes is needed to improve ourconfidence in the effects of various diagnostic interventions and toallow advice to be provided with greater confidence.
Overview
Wheezing occurs commonly during infancy(1). In most cases, wheezing episodes aremild and easily treated (2). However,some infants will develop persistent orrecurrent wheezing, which is often severe(3). These infants are frequently referredto pediatric pulmonology specialists
for further evaluation and treatment.Guidelines for diagnostic testing exist forolder children with asthma (4), but suchguidelines are lacking for wheezing infants.In a 2009 survey of Assembly on Pediatricsmembers of the American Thoracic Society(ATS), infantile wheezing was one of thehighest ranked topics for which membersdesired a guideline. To address this
knowledge gap and interest, the ATSconvened a committee of pediatricpulmonologists with clinical and researchexperience in infantile wheezing to developevidence-based guidelines for the diagnosticevaluation of infantile wheezing.
For these guidelines, the committeedefined infantile wheezing as recurrent orpersistent episodes of wheezing in infants
*These authors contributed equally to this work.
ORCID ID: 0000-0003-4431-0644 (C.L.R.).
Correspondence and requests for reprints should be addressed to Clement L. Ren, M.D., Indiana University School of Medicine, Riley Hospital for Children,705 Riley Hospital Drive/ROC4270, Indianapolis, IN 46202. E-mail: [email protected]
This article has an online supplement, which is accessible from this issue’s table of contents at www.atsjournals.org
Am J Respir Crit Care Med Vol 194, Iss 3, pp 356–373, Aug 1, 2016
Copyright © 2016 by the American Thoracic Society
DOI: 10.1164/rccm.201604-0694ST
Internet address: www.atsjournals.org
356 American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016
ContentsOverviewIntroductionUse of These GuidelinesMethods
DefinitionProcess
ResultsQuestion 1: Should Infants withPersistent Wheezing despiteTreatment with Bronchodilators,Inhaled Corticosteroids, orSystemic Corticosteroids UndergoAirway Survey via FlexibleFiberoptic Bronchoscopy?
Question 2: Should Infants withPersistent Wheezing despiteTreatment with Bronchodilators,Inhaled Corticosteroids, orSystemic CorticosteroidsUndergo BronchoalveolarLavage?
Question 3: Should Infantswith Persistent Wheezingdespite Treatment with
Bronchodilators, InhaledCorticosteroids, or SystemicCorticosteroids Be Managedaccording to the Results ofInfant Pulmonary FunctionTesting Using the Raised-VolumeRapid ThoracoabdominalCompression Technique orClinical Assessment Alone?
Question 4: Should Infants withoutEczema Who Have PersistentWheezing despite Treatment withBronchodilators, InhaledCorticosteroids, or SystemicCorticosteroids Undergo EmpiricFood Avoidance?
Question 5: Should Infantswith Persistent Wheezing despiteTreatment with Bronchodilators,Inhaled Corticosteroids, orSystemic Corticosteroids Undergo24-Hour Esophageal pHMonitoring?
Question 6: Should Infantswith Persistent Wheezing despite
Treatment with Bronchodilators,Inhaled Corticosteroids, orSystemic Corticosteroids Undergoan Upper Gastrointestinal SeriesRather Than 24-Hour EsophagealpH Monitoring?
Question 7: Should Infantswith Persistent WheezingThat Is Not Relieved byBronchodilators, InhaledCorticosteroids, or SystemicCorticosteroids UndergoGastroesophagealScintigraphy Rather Than24-Hour EsophagealpH Monitoring?
Question 8: Should Infants withoutNeurologic Pathology withPersistent Wheezing That Is NotRelieved by Bronchodilators,Inhaled Corticosteroids, orSystemic CorticosteroidsUndergo a Swallowing FunctionStudy?
Limitations and Future Directions
less than 24 months old (herein referred toas “infants with persistent wheezing”). Theguidelines address diagnostic tests thatare frequently considered by pediatricpulmonologists and other clinicians whenevaluating infantile wheezing, but are eithercontroversial or a frequent source ofuncertainty. Diagnostic tests that aregenerally considered standard of care (e.g.,chest radiography) were not addressed.
The committee performed a pragmaticevidence synthesis and then used theGrading of Recommendations, Assessment,Development, and Evaluation (GRADE)approach (5) to formulate and grade thefollowing recommendations:
1. For infants with persistent wheezingdespite treatment with bronchodilators,inhaled corticosteroids, or systemiccorticosteroids, we suggest an airwaysurvey via flexible fiberoptic bronchoscopy(conditional recommendation, very lowquality of evidence).
2. For infants with persistent wheezing despitetreatment with bronchodilators, inhaledcorticosteroids, or systemic corticosteroids,we suggest bronchoalveolar lavage (BAL)(conditional recommendation, very lowquality of evidence).
3. We recommend research studies ininfants with persistent wheezing despitetreatment with bronchodilators, inhaled
corticosteroids, or systemic corticosteroids,which compare clinical outcomes amongthose who are managed according toresults of infant pulmonary functiontesting using the raised-volume rapidthoracoabdominal compression (RVRTC)method versus those who are managedaccording to clinical assessment alone.
4. A. For infants who do not have eczemabut have persistent wheezing despitetreatment with bronchodilators,inhaled corticosteroids, or systemiccorticosteroids, we suggest thatclinicians and caregivers not useempiric food avoidance or dietarychanges (conditional recommendation,very low quality of evidence).
B. We recommend research studies thatdetermine whether food avoidance ordietary changes guided by food allergytesting improves clinical outcomesin infants who do not have eczemabut have persistent wheezing despitetreatment with bronchodilators,inhaled corticosteroids, or systemiccorticosteroids.
5. For infants with persistent wheezingdespite treatment with bronchodilators,inhaled corticosteroids, or systemiccorticosteroids, we suggest 24-houresophageal pH monitoring (conditional
recommendation, very low quality ofevidence).
6. For infants with persistent wheezingdespite treatment with bronchodilators,inhaled corticosteroids, or systemiccorticosteroids, we suggest 24-houresophageal pH monitoring rather thanupper gastrointestinal radiography(conditional recommendation, very lowquality of evidence).
7. For infants with persistent wheezingdespite treatment with bronchodilators,inhaled corticosteroids, or systemiccorticosteroids, we suggest 24-houresophageal pH monitoring ratherthan gastrointestinal scintigraphy(conditional recommendation, very lowquality of evidence).
8. For infants with persistent wheezingdespite treatment with bronchodilators,inhaled corticosteroids, or systemiccorticosteroids, we suggest performingvideo-fluoroscopic swallowing studies(conditional recommendation, very lowquality of evidence).
Introduction
Wheezing during infancy is a commonclinical problem. In the Tucson Children’sRespiratory Study, a longitudinal birthcohort study of healthy full-term infants,
AMERICAN THORACIC SOCIETY DOCUMENTS
American Thoracic Society Documents 357
34% of children had at least one episode ofwheezing before age 3 years (1). In someinfants, this is a sign of early-onset asthma(6), whereas other infants may wheezebecause of diminished airway function orinnate immune responses (7, 8). For themajority of infants, these wheezing episodesare mild, episodic, and easily treated.However, some infants will develop severerecurrent or persistent wheezing. Guidelinesfor the evaluation and treatment of asthmain older children and the general approachto the evaluation of infantile wheezing havebeen published (4, 9), but no guidelines existfor the use of more specialized testing, suchas flexible fiberoptic bronchoscopy, in theevaluation of infants with persistentwheezing. Recognizing the need for clinicalguidance on the diagnostic evaluation ofwheezing infants, the ATS convened aguideline development committee ofpediatric specialists to conduct pragmaticevidence syntheses and then use the evidencesyntheses as the basis for recommendationsfor the evaluation of persistent wheezingin infancy.
Use of These Guidelines
These ATS guidelines are not meant toestablish a standard of care. Rather,they represent an effort to summarizeevidence and provide reasonable clinicalrecommendations based on that evidence.Clinicians, patients, third-party payers,other stakeholders, and the courts shouldnever view these recommendations asdictates. No guidelines or recommendationscan take into account all of the oftencompelling, unique individual clinicalcircumstances. Therefore, no one chargedwith evaluating clinicians’ actions shouldattempt to apply the recommendationsfrom these guidelines by rote or in a blanketfashion. These guidelines are not intendedto be a comprehensive review of theevaluation of infantile wheezing, butrather to provide evidence-basedrecommendations for a set of specializeddiagnostic tests frequently considered in theevaluation of this patient population.Clinicians will be able to use theserecommendations when consideringspecific diagnostic tests for the evaluation ofpersistent wheezing. Recommendationsfor order or selection of diagnostic testingare beyond the scope of this document, andsuch decisions will vary depending on
the specific clinical situation and parentpreferences.
Methods
DefinitionFor these guidelines, the committeedefined infantile wheezing as recurrent orpersistent episodes of wheezing in infantsless than 24 months old (herein, referred toas “infants with persistent wheezing”).This cutoff was chosen for two reasons:previous documents have addressedwheezing in preschool-aged children(3–5 yr old) (10), and wheezing on the basisof diminished airway function tends toimprove by age 3 years (1). The populationwas further limited to infants withpersistent wheezing despite treatment withrecommended first-line therapies ofbronchodilators, inhaled corticosteroids, orsystemic corticosteroids (11).
ProcessThe co-chairs (C.L.R. and C.R.E.) wereconfirmed by the ATS Assembly onPediatrics, Program Review Subcommittee,and Board of Directors. A guidelinedevelopment committee was then assembled,which consisted of pediatric clinicians andresearchers with expertise in the evaluation ofwheezing during infancy. All members of thecommittee disclosed and were vetted forpotential conflicts of interest according to therules and procedures of the ATS. Thecommittee then developed clinical questions,using the PICO (Patient, Intervention,Comparator, and Outcomes) framework.Each question was the basis of a pragmaticevidence synthesis, which consisted ofsearching the Medline and CINAHL(Cumulative Index to Nursing and AlliedHealth Literature) databases on the basis ofprespecified search criteria, selecting studiesbased on prespecified selection criteria, andappraising and summarizing the evidenceaccording to the GRADE approach. Theevidence syntheses were used as the basis forthe formulation of recommendations, whichwas based on consideration of the balance ofbenefits versus harms and burdens, quality ofevidence, patient preferences, and cost andresource use. The recommendations weregraded according to the GRADE approach.The specifics of the PICO framework,outcomes, and other methods aredescribed in greater detail in the onlinesupplement.
Results
Question 1: Should Infants withPersistent Wheezing despiteTreatment with Bronchodilators,Inhaled Corticosteroids, or SystemicCorticosteroids Undergo AirwaySurvey via Flexible FiberopticBronchoscopy?
Summary of evidence. Our literature searchdid not identify any studies that comparedwheezing infants undergoing airway survey viabronchoscopy with wheezing infants who didnot undergo airway survey. Therefore, ourrecommendation is based on 10 case seriesthat collectively included 1,364 patients andreported that 452 of the 1,364 patients (33%)who underwent airway survey for respiratorysymptoms were found to have an anatomicabnormality known to cause wheezing(Table 1) (12–21). Lesions includedtracheomalacia, bronchomalacia,tracheobronchomalacia, vascular rings,vascular slings, and airway compression by avascular structure. No major complicationswere reported in any of the case series, withminor complications such as transienthypoxemia described in 5–10% of subjects.
Infants with wheezing due totracheomalacia, bronchomalacia, ortracheobronchomalacia are typicallymanaged by observation alone if wheezing isthe only abnormality or the associatedsymptoms are mild, because the vastmajority of infants improve over timewith conservative therapy (22). Infantswith wheezing due to tracheomalacia,bronchomalacia, or tracheobronchomalaciaoccasionally require an intervention(e.g., positive airway pressure, surgery, orstenting) because of accompanying life-threatening airway obstruction, respiratoryfailure, recurrent pneumonias, or failureto thrive. Positive airway pressureimmediately decreases respiratory distress,restores airway patency, and improvespulmonary function according to multiplesmall case series and case reports (23–30).Surgery (most commonly, aortopexy)relieves obstruction in virtually all patientswith tracheomalacia, but is less effective inpatients with tracheobronchomalacia orbronchomalacia according to small caseseries (31–37). This was illustrated by a caseseries in which 21 of 21 patients (100%)had tracheomalacia corrected by aortopexy,but only 1 of 4 patients (25%) with
AMERICAN THORACIC SOCIETY DOCUMENTS
358 American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016
Tab
le1.
Airw
aySurve
y:Qua
lityAss
essm
entan
dSum
maryof
Find
ings
Qua
lityAss
essm
ent
Sum
maryofFind
ings
No.
Study
Des
ign
Limitations
Indirec
tnes
sInco
nsistenc
yIm
precision
Pub
lication
Bias
Importan
ceof
Outco
me
Qua
lity
of
Eviden
ce
Freq
uenc
ywithwhich
bronc
hosc
opyiden
tifies
anan
atomical
lesionkn
ownto
caus
ewhe
ezing
10*
Cas
ese
ries†
Serious
‡Serious
xNon
eSerious
jjUnd
etec
ted
Not
apresp
ecified
outcom
eVerylow
The10
case
serie
sco
llectively
includ
ed1,36
4patients.
452of
1,36
4patients(33%
)were
foun
dto
have
anatom
ical
abno
rmalities
know
nto
caus
ewhe
ezing.
Thelarges
tse
riesinclud
ed88
5patients.
Whe
nthis
stud
ywas
remov
ed,thees
timatewas
unch
ange
d.
Com
plications
wererarely
reported.
Freq
uenc
yofwhe
ezingaftertrea
tmen
t27
¶Cas
ese
ries†
and
case
reports
Serious
**Serious
††
Non
eSerious
‡‡
Und
etec
ted
Critical
Verylow
90%
ofpatientswith
trac
heom
alac
ia,bron
chom
alac
ia,
ortrac
heob
ronc
homalac
iaim
prove
dwith
timealon
e.88
–10
0%of
patientswith
vascular
rings
,vascular
slings
,or
vascular
compress
ionof
theairw
ays
improve
dwith
surgery.
Com
plications
occu
rred
in10
%an
dmortalityin
,5%
.
*References12–2
1.
†Nostudiesincludedacontrolgroup.Therefore,theyare
allcase
serie
s.‡Lim
itatio
nswere
serio
usbecause
ofprobable
selectio
nbiasrelatedto
whounderw
entbronchoscopyandwhodid
not.Most
studiesdid
notprovidedetails
ofhow
theydecidedwhento
perform
bronchoscopywith
airw
aysu
rveyandwhennotto
perform
bronchoscopywith
airw
aysu
rvey.
x Indire
ctness
wasserio
usbecause
fewstudieswere
limitedso
lelyto
wheezinginfants.Most
includedinfants
with
resp
iratory
symptomsorsigns,su
chasstrid
or,wheezing,cough,resp
iratory
distress
orfailure,orrecurrentinfectio
ns.
jj Imprecisionwasserio
usbecause
sample
size
wassm
all(,
100)in
allbutoneofthestudiesthatwere
reviewed.
¶References23–4
2and44–5
0.
**Lim
itatio
nswere
serio
usbecause
ofprobable
selectio
nbiasrelatedto
whowaseventually
treatedwith
positiveairw
aypressure,airw
aystentin
g,andsu
rgery,aswellaswhichtypeof
surgery
wasused.
††Indire
ctness
wasserio
usbecause
few
studieswere
limitedso
lelyto
wheezinginfants.Most
includedinfants
beingtreatedforavarie
tyofresp
iratory
symptomsorsigns(e.g.,strid
or,
wheezing,cough,resp
iratory
distress
orfailure,orrecurrentinfectio
ns)
causedbyanatomicalabnorm
alities.
‡‡Im
precisionwasserio
usbecause
thesample
size
wassm
all(,
100)in
allthestudiesthatwere
reviewed.
AMERICAN THORACIC SOCIETY DOCUMENTS
American Thoracic Society Documents 359
tracheobronchomalacia was corrected byaortopexy (31). Airway stenting has beenused to improve airway obstruction ininfants (38–42), but complicationsincluding formation of granulation tissue,migration, or erosion occurred in 50% ofcases and were potentially associatedwith death in 2 of 22 infants (39, 40).Newer approaches include directtracheobronchopexy (43).
In contrast, wheezing due to vascularrings, vascular slings, and airwaycompression by a vascular structure isunlikely to self-resolve, and surgicalcorrection is performed for symptomaticpatients. According to seven case series,improvement in respiratory symptomswas seen in 88–100% of patients, andcomplete resolution was seen in more than50% of patients. Recurrent laryngeal nerveinjury was the most common surgicalcomplication and occurred in less than 10%of patients. More serious complications suchas aortoesophageal fistula, heart failure, orwound infection associated with mortalityoccurred in less than 5% of patients (44–50).
Taken together, the evidence indicatesthat an anatomic abnormality known tocause wheezing can be identified by airwaysurvey in approximately 33% of patientswith respiratory symptoms, and in thecommittee’s clinical experience more than90% of such patients will improve becauseeither their condition is self-limited orsurgery can correct the abnormality.Thus, about 30% of patients are likely tobenefit from an airway survey, eitherthrough direct intervention (surgery) or byavoiding unnecessary tests and treatmentsfor a benign, self-limited condition.Identification of airway malacia may alsohelp in management of infants believedto have concomitant asthma, becauseb-agonists may adversely affect airwaydynamics in these children (51). Thecommittee has very low confidence(i.e., quality of evidence) in the accuracy ofthese estimated effects, because the caseseries had probable selection bias and mostseries looked at infants who underwentbronchoscopy for respiratory symptoms,not specifically wheezing.
Rationale. Bronchoscopy with airwaysurvey that identifies an anatomical cause ofwheezing confers several potential benefits.Finding tracheomalacia, bronchomalacia, ortracheobronchomalacia usually leads toconservative management, which has a highsuccess rate and other benefits including
relief from the burden, cost, and potentialharms of further diagnostic testing; probablereductions in the use of ineffectivemedications (bronchodilators or systemiccorticosteroids) and the frequency ofphysician visits; and parental reassurance,given the high likelihood that the conditionwill spontaneously resolve. Findingvascular rings, vascular slings, and airwaycompression by a vascular structure leads tosurgical therapy with an 88–100% successrate. In the judgment of the committee,the possibility that approximately 30% ofinfants who undergo airway survey willbenefit far exceeds the burdens and costof bronchoscopy, as well as the potentialharms (i.e., complications due tobronchoscopy are rare and complications dueto subsequent therapy range from zero forconservative management to approximately10% for surgery). The recommendation forairway survey is conditional because the lowquality of evidence provides little certaintythat the benefits of airway survey exceed theburdens, costs, and harms. There are alsoemerging data on neurodevelopmental risksof anesthesia that need to be considered (52).In addition, parental preferences regardinginvasive procedures tend to be highlyindividualized.
Recommendation 1. For infants withpersistent wheezing despite treatment withbronchodilators, inhaled corticosteroids, orsystemic corticosteroids, we suggest airwaysurvey via flexible fiberoptic bronchoscopy(conditional recommendation, very lowquality of evidence).
Question 2: Should Infants withPersistent Wheezing despiteTreatment with Bronchodilators,Inhaled Corticosteroids, or SystemicCorticosteroids UndergoBronchoalveolar Lavage?
Summary of evidence. Our literature searchdid not identify any studies that comparedwheezing infants undergoing BAL withwheezing infants who did not undergo BAL.Therefore, our recommendation is based ondata from 20 case series, identified in ourliterature search, showing that 14–80% ofinfants (40–60% in most studies) withrecurrent or persistent wheezing produce apositive BAL culture (Table 2) (12, 16, 18,53–70). No complications were reported inany of the case series.
Patients with a positive BAL culturetypically receive a prolonged course of
antibiotic therapy, and indirect evidencefrom a randomized trial of 50 children withproductive cough presumed to be caused bybacterial bronchitis found that the coughresolved in 48% of children who receivedantibiotic therapy, compared with only 16%of those who did not receive antibiotics (71).The trial likely underestimated the effectsof antibiotics in patients with bacterialbronchitis because children did not need tohave a confirmed bacterial infection to beenrolled in the trial; patients withoutbacterial bronchitis are unlikely to haveresponded to antibiotic therapy and,therefore, their inclusion would have madeantibiotic therapy appear less effective.
On the basis of the rates of BALinfection (40–60%) and symptomimprovement with antibiotic treatment(48%) described previously, we estimatethat 20–30% of children with persistentwheezing who undergo bronchoscopy withBAL will be found to have a lower airwaybacterial infection and that their symptomswill improve with antibiotic therapy. Thecommittee’s confidence in the estimatedeffects of BAL (i.e., the quality of evidence)is very low because it is based on prevalenceestimates derived from case series anda therapeutic effect estimated from arandomized trial, both of which had seriouslimitations. The case series were limitedby selection bias, indirectness of thepopulation (children with cough ratherthan infants with wheezing), and smallsample sizes with few events. Therandomized trial was similarly limited byindirectness of the population (childrenwith cough rather than infants withwheezing), indirectness of the outcome(cure of infection rather than improvementin wheezing), and imprecision (smallsample size with few events).
Rationale. To confirm or exclude lowerairway bacterial infection as the cause ofrecurrent or persistent wheezing, clinicianshave three options: (1) they can performBAL and then treat patients with confirmedbacterial infection with antibiotics;(2) they can empirically treat all patientswith empiric antibiotics; or (3) they can doneither. The committee judged the balanceof the benefits versus the burdens andrisks to be greater for the first option(i.e., 20–30% children improve aftertreatment of BAL-identified infection) thanfor either the second option (i.e., thesame infection cure rate, but 40–60% ofpatients receive unnecessary antibiotics
AMERICAN THORACIC SOCIETY DOCUMENTS
360 American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016
Tab
le2.
Bronc
hoalve
olar
Lava
ge:Qua
lityAss
essm
entan
dSum
maryof
Find
ings
Qua
lityAss
essm
ent
Sum
maryofFind
ings
No.
Study
Des
ign
Limitations
Indirec
tnes
sInco
nsistenc
yIm
precision
Pub
lication
Bias
Importan
ceof
Outco
me
Qua
lityof
Eviden
ce
Freq
uenc
ywithwhich
BALdetec
tsinfection(defi
nedas
apositive
culture)
20*
Cas
ese
ries†
Serious
‡Serious
xNon
ejj
Serious
¶Und
etec
ted
Not
apresp
ecified
outcom
eVerylow
Amon
gthe20
case
serie
siden
tified
,the
proportio
nof
BALsa
mplings
that
resu
ltedin
apos
itive
microbiologica
lcu
lture
rang
edfro
m14
to80
%;h
owever,
mos
tse
riesreportedthat
40–60
%of
BALsa
mplings
resu
ltedin
apos
itive
microbiologica
lculture.
Non
eof
these
riesreportedco
mplications
from
BAL.
Freq
uenc
yofwhe
ezingaftertrea
tmen
t1**
RCT
Serious
††
Serious
‡‡
Non
eSerious
¶Und
etec
ted
Critical
Verylow
50ch
ildrenwith
wet
coug
hdue
topresu
med
bac
teria
lbronc
hitis
were
trea
tedwith
either
antib
iotic
sor
noan
tibiotic
s.Amon
gthepatientswho
weretrea
ted,thecu
rerate
was
48%
.Amon
gthepatientswho
wereno
ttrea
ted,thecu
rerate
was
16%
.
Definitionofabbreviatio
ns:
BAL=bronchoalveolarlavage;RCT=randomizedcontrolledtrial.
*References6,10,12,and44–6
1.
†Nostudiesincludedacontrolgroup.Therefore,theyare
allcase
serie
s.‡Lim
itatio
nswere
serio
usbecause
ofprobable
selectio
nbiasrelatedto
whounderw
entBALandwhodid
not.Most
studiesdid
notprovidedetails
ofhow
theydecidedwhento
perform
bronchoscopywith
BALandwhennotto
perform
bronchoscopywith
BAL.
x Indire
ctness
wasserio
usbecause
few
studieswere
limitedso
lelyto
wheezinginfants.Most
includedolderchildrenandfrequently
coughoverla
ppedwith
wheezing.
jj Inconsistencywasnotaproblem.In
most
studies,
theprevalenceofpositiveBALcultu
reswas40–6
0%.
¶Im
precisionwasserio
usbecause
thesample
size
wassm
all(,
100)in
allthestudiesthatwere
reviewed.
**Reference71.
††Descrip
tionsoftheconcealm
entofallocatio
nandblindingoftheassessors
were
incomplete.
‡‡Indire
ctness
wasserio
usbecause
thepopulatio
nofinterest
iswheezinginfants,butthepopulatio
nstudiedwaschildrenwith
cough.In
addition,theoutcomeofinterest
waswheezing,but
theoutcomeofthestudieswascure
ofinfectio
n.
AMERICAN THORACIC SOCIETY DOCUMENTS
American Thoracic Society Documents 361
with their associated risks, such as fever,rash, anaphylaxis, acquisition of resistance,and change in gut microbiome [72]) or thirdoption (i.e., only 6.4–9.6% infection curerate). The committee recognized that theestimated cure rates for lower respiratoryinfection likely overestimate the cure rate forwheezing because some infants with lowerrespiratory tract bacterial infection haveadditional or alternative causes of wheezing;nonetheless, the committee still thoughtthat the risk of BAL is sufficiently smallthat the benefits probably outweigh theburdens and harms.
The strength of the recommendationfor BAL is conditional because thecommittee’s very low confidence in theestimated effects of BAL made it impossibleto be certain that the benefits of BALoutweigh the risks and burdens in themajority of patients. Moreover, BALrequires bronchoscopy, an invasiveprocedure requiring sedation, and it isuncertain that most families would wantbronchoscopy performed on their infant,despite persistent wheezing.
Recommendation 2. For infants withpersistent wheezing despite treatment withbronchodilators, inhaled corticosteroids, orsystemic corticosteroids, we suggest BAL(conditional recommendation, very lowquality of evidence).
Question 3: Should Infantswith Persistent Wheezingdespite Treatment with Bronchodilators,Inhaled Corticosteroids, or SystemicCorticosteroids Be Managedaccording to the Results of InfantPulmonary Function Testing Usingthe Raised-Volume RapidThoracoabdominal CompressionTechnique or Clinical AssessmentAlone?
Summary of evidence. Our literature searchrevealed 1,261 studies related to wheezingand pulmonary function tests (PFTs) inchildren. The overwhelming majority (1,226studies) were excluded because they enrolledchildren during later childhood, and theguideline development committee believedthat such evidence was too indirect to informjudgments for infants. Among the 35 studiesthat involved PFTs performed duringinfancy, only 2 (from the same cohort ofpatients at two different time points)described clinical outcomes afterthe assessment of bronchodilator
responsiveness (BDR) using the RVRTCtechnique (73, 74). Both studies reportedthat the presence of BDR identified by theRVRTC technique predicted future acuteexacerbations of wheezing requiringtreatment with systemic corticosteroids. Nostudies were identified that compared theeffects of management according to the BDRmeasured by the RVRTC technique versusmanagement based on clinical assessmentalone (i.e., no PFTs) on the clinical outcomesof interest (frequency of wheezing, frequencyof doctor visits, frequency of hospitalization,prescriptions for bronchodilators,prescriptions for inhaled or systemiccorticosteroids, parental stress, additionaldiagnostic testing, and inappropriatetherapy). Thus, there was no publishedevidence available to inform the guidelinedevelopment committee’s judgments.
Rationale. In the absence of publishedevidence, the guideline developmentcommittee turned to its collective clinicalexperience to try to answer the question.However, despite extensive discussion, theguideline development committee couldnot reach consensus on a clinicalrecommendation for or against infant PFTs,due to the paucity of evidence. Somemembers of the committee believed that theinformation derived from infant PFTs didnot justify the burdens and risks involved inperforming the test. Among the potentialbenefits of confirming or excluding BDR,the clinician may be directed away or towarddiagnostic testing that targets anatomicalcauses of wheezing, respectively. Among therisks and burdens of such testing are theneed for sedation; the risks associated withairway occlusion, gastric distention, andaerophagia; the additional personnel neededto monitor the infant during and after thetest; and the time and personnel needed toset up and conduct the test. Other membersof the guideline development committeebelieved that there are circumstances inwhich infant PFTs are clinically useful. Forexample, a restrictive pattern on the PFTmight lead clinicians to explore interstitiallung disease, andmarked gas trapping mightmotivate clinicians to evaluate the infantfurther for neuroendocrine hyperplasia ofinfancy, although wheezing is usually not acommon feature of this condition (75).
Recommendation 3. In infants withpersistent wheezing despite treatmentwith bronchodilators, inhaledcorticosteroids, or systemic corticosteroids,we recommend research studies that
compare clinical outcomes among infantswho are managed according to infant PFTperformed using the RVRTC techniqueversus those who are managed according toclinical assessment alone.
Question 4: Should Infants withoutEczema Who Have PersistentWheezing despite Treatment withBronchodilators, InhaledCorticosteroids, or SystemicCorticosteroids Undergo EmpiricFood Avoidance?
Summary of evidence. The NationalInstitute of Allergy and Infectious Diseaseshas published clinical guidelines on foodallergy in children with eczema, includingrecommendations for food allergy testingand avoidance in infants and children withthis condition (76). Therefore, we focusedour question on the role of food avoidancein infants without eczema. Our systematicreview identified four studies that assessedthe results of empiric food avoidance(Table 3). All of the studies measured ourprespecified outcome of frequency ofwheezing, but none measured any of ourother prespecified outcomes, includingfrequency of doctor visits, frequency ofhospitalization, prescriptions forbronchodilators, prescriptions for inhaledor systemic corticosteroids, parentalstress, additional diagnostic testing, andinappropriate therapy. A trial randomlyassigned 487 infants to receive either acow’s milk–free diet or a usual diet for atleast the initial 4 months of life and foundno difference in wheezing, eczema, ornasal discharge at 1 year (77). Fourhundred and forty-six of the infants werereassessed 6 years later. There were stillno differences in the incidence ofwheezing, asthma diagnoses, eczema, orallergic rhinitis (78). Another trialrandomly assigned 110 infants to receiveeither a partially hydrolyzed formula orstandard infant formula for the first4 months of life. There was no differencein the incidence of wheezing at 2 years,although eczema was more commonamong the infants who received astandard formula (79). Finally, aprospective cohort study monitored 6,905newborns through preschool age andfound no relationship between the earlyintroduction of potentially allergenicfoods (e.g., cow’s milk, egg, nuts, soy, orgluten) and either wheezing or eczema at
AMERICAN THORACIC SOCIETY DOCUMENTS
362 American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016
Tab
le3.
EmpiricFo
odAvo
idan
ce:Qua
lityAss
essm
entan
dSum
maryof
Find
ings
Qua
lityAss
essm
ent
Sum
maryofFind
ings
No.
StudyDes
ign
Limitations
Indirec
tnes
sInco
nsistenc
yIm
precision
Pub
lication
Bias
Importan
ceof
Outco
me
Qua
lityof
Eviden
ce
Inciden
ceofwhe
ezing(m
easu
redat
1yr)
1RCT*
Serious
†Serious
‡Non
eSerious
xNon
eCritical
Verylow
Thetrialo
f487
infantsfoun
dno
differen
cein
theinciden
ceof
whe
ezingduringthefirsty
ear
oflife:
84/232
(36%
)vs
.80
/242
(33%
)Inciden
ceofwhe
ezing(m
easu
redat
>1–4yr)
21RCTj
j 11ob
servationa
lstud
y¶Serious
**Serious
††
Non
eSerious
xNon
eCritical
Verylow
Therand
omized
trialo
f11
0infantsfoun
dno
differen
cein
theinciden
ceof
whe
ezing
at2yr
Theprosp
ectiv
eco
hortstud
yof
6,90
5ch
ildrenfoun
dno
differen
cein
theinciden
ceof
whe
ezingat
2,3,
or4yr
amon
gthos
ewho
were
introd
uced
toco
w’s
milk,
nuts,eg
g,so
y,or
gluten
earlier
orlaterthan
6mo
Inciden
ceofwhe
ezing(m
easu
redat
>4yr)
1RCT‡
‡Serious
xxSerious
‡Non
eSerious
xNon
eCritical
Verylow
Thetrialo
f446
infantsfoun
dno
differen
cein
theinciden
ceof
whe
ezingduringtheinitial
7yr
oflife:
148/21
5(69%
)vs.
157/23
1(68%
)Diagno
sisofas
thma
1RCT‡
‡Serious
xxSerious
‡Non
eSerious
xNon
eIm
portant
Verylow
Thetrialo
f446
infantsfoun
dno
differen
cein
theinciden
ceof
asthmadiagn
oses
duringthe
initial
7yr
oflife:
60/215
(28%
)vs
.81
/231
(35%
)
Definitionofabbreviatio
n:RCT=randomizedcontrolledtrial.
*Reference77.
†Concealm
entwasbyenvelopescontainingcolor-codedcardsandthecaregivers
were
notblinded.
‡Thequestionisaboutinfants
with
outeczemawhohave
refractory
wheezing;however,thetrialenrollednewborns.
Inaddition,thequestionasksaboutgeneralfoodavo
idance,butthetrial
employedonlycow’s
milk
avo
idance.
x Theopposite
endsoftheconfid
enceintervalwould
resu
ltin
differentclinicaldecisionsifreal.
jj Reference78.
¶Reference79.
**Therandomizedtrialdid
notreportconcealm
entandthecaregivers
were
notblinded;theobservatio
nalstudyreliedonquestionnaire
sthatretrosp
ectivelyassessedtheintroductio
nof
certain
foodsinto
thediet,creatin
gariskofrecallbias.
††Thequestionis
forinfants
with
outeczemawhohave
refractory
wheezing;however,therandomizedtrialenrollednewbornswith
afamily
history
ofatopyandtheobservatio
nalstudy
monito
redanynewborn.In
addition,thequestionasksaboutgeneralfoodavo
idance,butthetrialemployedonlycow’s
milk
avo
idance.
‡‡Reference80.
xxThiswasa7-yearfollow-upofMiske
llyandcolleagues(77).Thus,
ithadthesamelim
itatio
ns:
Concealm
entwasbyenvelopescontainingcolor-codedcardsandthecaregivers
were
not
blinded.In
addition,41ofthe487patients
droppedoutofthestudybetw
eenYears
1and7.
AMERICAN THORACIC SOCIETY DOCUMENTS
American Thoracic Society Documents 363
ages 2, 3, and 4 years. The study plans tomonitor the participants to adulthood (80).None of the studies evaluated the effects ofempiric food avoidance in a subgroup offood antigen IgE–positive infants.
Taken together, the evidence suggeststhat empiric food avoidance has no effecton the frequency of wheezing. However, itprovides very low confidence (i.e., quality ofevidence) in the estimated effects because therandomized trials were limited by risk of bias,indirectness of population and intervention,and imprecision, and the observational studywas limited by possible recall bias.
Rationale. The guideline developmentcommittee chose to include questionsregarding food avoidance and allergy testingbecause in the collective experience of thecommittee, parents of infants with persistentwheezing frequently raise this topic.Although there is evidence that respiratorysymptoms can be provoked by food antigensin infants with eczema (81), less is knownabout this relationship in infants withouteczema. The guideline developmentcommittee’s judgments were based on theimpact of empiric food avoidance onfrequency of wheezing, because our otherprespecified outcomes were not reported.The lack of beneficial effects due to empiricfood avoidance in any study, combinedwith the committee’s recognitionthat empiric food avoidance can beburdensome, led the committee to suggestthat empiric food avoidance not be used ininfants without eczema who have persistentwheezing despite standard therapy. Thestrength of the recommendation isconditional because the very low quality ofevidence prevented the committee frombeing certain about its judgments. In otherwords, although the committee believesthat there is no evidence that the desirableconsequences of empiric food avoidanceoutweigh the undesirable consequences inthe majority of patients, it recognizes thatthere may be clinical circumstances inwhich a trial of empiric food avoidance maybe reasonable for a minority of patientsfor whom the clinical history stronglycorrelates respiratory symptoms with foodexposure or in whom respiratory symptomsare elicited in a double-blind placebo-controlled food challenge.
Recommendation 4.
A. For infants without eczema who havepersistent wheezing despite treatment
with standard therapies, we suggestnot using empiric food avoidance(conditional recommendation, very lowquality of evidence).
B. We recommend research to determinewhether or not empiric food avoidance isbeneficial for the subgroup of infantswho are positive for IgE to food antigens.
Question 5: Should Infants withPersistent Wheezing despiteTreatment with Bronchodilators,Inhaled Corticosteroids, or SystemicCorticosteroids Undergo 24-HourEsophageal pH Monitoring?
Summary of evidence. Our systematic reviewdid not identify any randomized trialsor controlled observational studiesthat compared clinical outcomes amongthose who underwent 24-hour esophagealpH monitoring versus those who didnot. However, we did identify three caseseries that used 24-hour pH monitoringto determine the prevalence ofgastroesophageal reflux (GER) amongchildren with wheezing and also reported theclinical outcomes that followed treatmentof those with confirmed GER (Table 4)(82–84).
The most recent case series (83)enrolled 25 infants and children withasthma (88% had persistent wheezing)and performed 24-hour pH monitoringon all participants. GER was identified in19 of 25 (76%) infants and children.Participants with GER were treatedwith a proton pump inhibitor andreassessed at 3 months, at which timethere were statistically significantimprovements in symptoms (from 2.3to 0.4 symptoms per day), use ofbronchodilators (from 8.3 to 1.4 d perpatient), use of systemic steroids (from5.3 to 0.4 d per patient), frequencyof exacerbations (from 1.5 to 0.3exacerbations per patient), andhospitalizations (from 9.1 to 0.5 dper patient) compared with beforetreatment.
The case series confirmed three earlierseries. In the first (82), 36 infants andchildren with various respiratory disordersunderwent 24-hour pH monitoring. GERwas identified in 22 of 36 infants andchildren (61%) , including 4 of 6 infantsand children (67%) with wheezing. Amongthose 22 patients, 9 patients underwentfundoplication, after which symptoms
improved in 6 and resolved in 3. Theremaining 13 patients with GER weretreated with medical management; 9 hadsymptomatic improvement and 4 were lostto follow-up. In the second series (84), 12infants with persistent wheezing despitebronchodilator and antiinflammatorytherapy underwent 24-hour pHmonitoring, and all were confirmed tohave GER. They were subsequently treatedwith prokinetic agents and histaminereceptor blockers; six improved enough tono longer require antiasthma medications,two improved enough that they requiredonly intermittent antiasthma medications,and four failed to improve and underwentfundoplication. After fundoplication,three of the four patients no longerrequired antiasthma medications. In thethird series (85), 81 children withrecurrent pneumonias or chronic asthmaunderwent 24-hour pH monitoring, and38 (47%) were found to have GER. Fortypatients were treated for GER (2 on thebasis of alternative tests). Among the12 children who underwent medicalmanagement, 10 improved (83%). Amongthe 24 children who underwent surgicaltreatment, 22 improved (92%). Fourpatients were lost to follow-up. None ofthe case series reported any adverse effectsfrom the 24-hour pH monitoring orsubsequent therapy.
Taken together, the evidence indicatesthat GER exists in 47–100% of infants withpersistent wheezing and, if identified, morethan 83% (most estimates are in the90–100% range) will improve with medicalor surgical treatment. However, theevidence provides very low confidence inthe estimated effects. With respect toindirectness of the population, most studiesincluded older children with a mix ofrespiratory problems in addition towheezing (e.g., recurrent pneumonia,apnea, stridor, and cough) and did notevaluate the wheezing infant subgroup.With respect to indirectness of theintervention, there was variability in themethods used for 24-hour pH monitoring,including positioning of probes, patientpositioning, dietary restrictions, scoringcriteria, definitions of an abnormal study,and use of impedance data. pH probesdetect only acid reflux unless paired withimpedance; thus, not using impedancedata may underestimate episodes ofpostprandial reflux in infants with frequentfeeds and buffering of gastric contents (86).
AMERICAN THORACIC SOCIETY DOCUMENTS
364 American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016
Tab
le4.
Gas
troe
sopha
geal
Refl
ux:Qua
lityAss
essm
entan
dSum
maryof
Find
ings
Qua
lityAss
essm
ent
Sum
maryofFind
ings
No.
Study
Des
ign
Limitations
Indirec
tnes
sInco
nsistenc
yIm
precision
Pub
lication
Bias
Importan
ceof
Outco
me
Qua
lityof
Eviden
ce
Freq
uenc
ywithwhich
24-h
esopha
gea
lpH
monitoring
(gold
stan
dard)iden
tifies
GER
4Cas
ese
ries*
Serious
†Serious
‡Non
eSerious
xNon
eNot
apresp
ecified
outcom
eVerylow
24-h
esop
hage
alpH
mon
itorin
giden
tified
GER
in19
/25infantsan
dch
ildren(76%
),jj 2
2/36
infantsan
dch
ildren(61%
),¶12
/12infants
(100
%),**an
d38
/81ch
ildren
(47%
).††
Whe
ezing
4Cas
ese
ries*
Serious
†Serious
‡Non
eSerious
xNon
eCritical
Verylow
Aca
sese
riesof
25infantsan
dch
ildren
with
presu
med
asthmafoun
dGERin
19/25(76%
).After
trea
tmen
tof
the
GER,s
ymptomsdec
reas
edfrom
2.3
to0.4sy
mptomsper
day
and
exac
erbations
dec
reas
edfrom
1.5to
0.3ex
acerbations
per
patient.jj
Aca
sese
riesof
36infantsan
dch
ildren
with
recu
rren
tresp
iratory
symptoms
foun
dGER
in22
/36(61%
).Amon
gthos
ewith
GER,9/22
(41%
)un
derwen
tfund
oplicationwith
subse
que
ntim
prove
men
tor
reso
lutio
nin
all,an
d13
/22(59%
)weremed
ically
man
aged
with
improve
men
tin
the9who
wereno
tlost
tofollo
w-up.¶
Aca
sese
riesof
12infantswith
persisten
twhe
ezingfoun
dGER
in12
/12(100
%);aftertrea
tmen
tof
the
GER,6/12
(50%
)no
long
erne
eded
med
ications
forwhe
ezing,
2/12
(17%
)ne
eded
med
ications
only
interm
itten
tly,an
d4/12
(33%
)ne
eded
fund
oplication,
which
elim
inated
thene
edformed
ications
in3/4(75%
).**
Aca
sese
riesof
81ch
ildrenwith
recu
rren
tpne
umon
iasor
chronic
asthmafoun
dGER
in38
/81(47%
).Fo
rtypatientsweretrea
tedforGER
(2on
thebas
isof
alternativetests).
Amon
gthe12
child
renwho
underwen
tmed
ical
man
agem
ent,10
improve
d(83%
).Amon
gthe24
child
renwho
underwen
tsu
rgical
trea
tmen
t,22
improve
d(92%
).Fo
urpatientswerelost
tofollo
w-up.††
(Con
tinue
d)
AMERICAN THORACIC SOCIETY DOCUMENTS
American Thoracic Society Documents 365
Tab
le4.
(Con
tinue
d)
Qua
lityAss
essm
ent
Sum
maryofFind
ings
No.
Study
Des
ign
Limitations
Indirec
tnes
sInco
nsistenc
yIm
precision
Pub
lication
Bias
Importan
ceof
Outco
me
Qua
lityof
Eviden
ce
Use
ofbronc
hodila
tors
1Cas
ese
riesjj
Serious
†Serious
‡‡
Non
eSerious
xxNon
eIm
portant
Verylow
Aca
sese
riesof
25infantsan
dch
ildren
with
presu
med
asthmafoun
dGERin
19/25(76%
).After
trea
tmen
tof
the
GER
,use
ofbron
chod
ilatorsde
creased
from
8.3to
1.4dpe
rpa
tient.jj
Use
ofsy
stem
icsteroids
1Cas
ese
riesjj
Serious
†Serious
‡‡
Non
eSerious
xxNon
eIm
portant
Verylow
Aca
sese
riesof
25infantsan
dch
ildren
with
presu
med
asthmafoun
dGERin
19/25(76%
).After
trea
tmen
tof
the
GER
,use
ofbron
chod
ilatorsde
creased
from
5.3to
0.3dpe
rpa
tient.jj
Hosp
italizations
1Cas
ese
riesjj
Serious
†Serious
‡‡
Non
eSerious
xxNon
eIm
portant
Verylow
Aca
sese
riesof
25infantsan
dch
ildren
with
presu
med
asthmafoun
dGERin
19/25(76%
).After
trea
tmen
tof
the
GER
,use
ofbron
chod
ilatorsde
creased
from
9.1to
0.5dpe
rpa
tient.jj
Definitionofabbreviatio
n:GER=gastroeso
phagealreflux.
*References82–8
5.
†There
wasnoprocess
toensu
rethatpatients
were
consecutivelyorrandomlyincluded;thus,
selectio
nbiasin
favo
rofinfants
clinically
susp
ectedofhavingGERislikely.
‡Thequestionis
specifically
aboutinfants
with
wheezing;however,threeofthecase
serie
sincludedolderchildrenandsymptomsotherthanwheezing.
x There
were
only154patients
inthefourcase
serie
scombined.
jj Reference83.
¶Reference84.
**Reference82.
††Reference85.
‡‡Thequestionis
specifically
aboutinfants
with
wheezing;however,thecase
serie
sincludedolderchildrenandasthmasymptomsotherthanwheezing.
xxThecase
serie
sincludedonly25infants
andchildren,ofwhom
only19were
treatedforGER.
AMERICAN THORACIC SOCIETY DOCUMENTS
366 American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016
Rationale. The guideline developmentcommittee believed that the balance ofbenefits versus risks, burdens, and costfavors 24-hour pH monitoring inmost infants who have persistentwheezing despite bronchodilator andantiinflammatory therapy. Specifically,among such infants who undergo 24-hourpH monitoring, 67–100% will be found tohave GER and nearly all will improvesubstantially with treatment, withoutrequiring further diagnostic testing. Theprocedure is well tolerated by the vastmajority of patients and, although itssemiinvasive nature and potentialneed for inpatient admission may beconcerning to some parents, thecommittee believed that most familieswould be willing to have the test done.Although combined pH and impedanceprobe monitoring has become the standardat most centers, the available evidencelargely predates widespread use ofimpedance probes. Therefore, the committeewas unable to comment specifically on thevalue of impedance monitoring.
An alternative to 24-hour pHmonitoring is an empiric trial of antiacidtherapy. However, in up to one-thirdpatients receiving empiric therapy, theantiacid therapy is inappropriate and incursunnecessary cost, burden, and risk. Inaddition, the rate of treatment success islikely to be lower among empirically treatedpatients because those with GER thatrequires fundoplication may be incorrectlyconsidered nonresponders. In that case, itmay be presumed that GER is not acontributor and the parents may never beoffered potentially curative surgical therapy.Furthermore, studies in older patientssuggest that proton pump inhibitor therapyis linked to increased risk of pneumonia(87). Although a similar risk has notbeen reported in infants, a normal pH-monitoring study could potentially reduceany risks associated with proton pumpinhibitor therapy.
The strength of our recommendation isconditional because the very low quality ofevidence provided little confidence in theestimated benefits and harms reported bythe case series. As a result, the committeecould not be certain about its judgmentsregarding the balance of benefits versusharms, burdens, and cost.
Recommendation 5. For infants withpersistent wheezing that is not relieved bybronchodilators, inhaled corticosteroids, or
systemic corticosteroids, we suggest 24-houresophageal pH monitoring (conditionalrecommendation, very low quality ofevidence).
Question 6: Should Infantswith Persistent Wheezing despiteTreatment with Bronchodilators,Inhaled Corticosteroids, or SystemicCorticosteroids Undergo an UpperGastrointestinal Series Rather Than24-Hour Esophageal pH Monitoring?
Summary of evidence. The guidelinedevelopment committee next asked whetheran upper gastrointestinal (UGI) series is anacceptable alternative to 24-hour esophagealpH monitoring, which we considered thereference standard. Our systematic reviewdid not identify any randomized trials orcontrolled observational studies thatcompared clinical outcomes among thosewho underwent a UGI series with those whounderwent 24-hour esophageal pHmonitoring. However, it did identify threestudies that evaluated the accuracy of a UGIseries in detecting GER in infants and childrenwith wheezing (Table 4) (82, 85).
In the only study that used 24-hour pHmonitoring as the reference standard, 79children (age, 2–17 yr) who had difficult-to-control asthma underwent 24-houresophageal pH monitoring. GER wasidentified in 58 of 79 children (73%). Abarium swallow study was then performed,which identified GER with a sensitivityand specificity of 46 and 82%, respectively(88).
The other two studies used variousreference standards, but reported enoughdata to enable us to estimate the sensitivityof a UGI series in the detection of GERrelative to 24-hour esophageal monitoring.In a study of infants and children (age,2 mo–10.5 yr) with recurrent respiratorydisorders (82), 22 of 36 (61%) were foundto have GER by 24-hour pH monitoring,and 15 of 35 (42%) were found to haveGER by UGI series. Assuming that patientsin whom GER was detected by UGI seriesalso had GER detected by 24-hour pHmonitoring, the sensitivity of a UGI serieswould be 68%. In a study of 82 infants andchildren (5 mo–16 yr) with recurrentpneumonia or chronic asthma (85), 40 werefound to have GER on the basis of studycriteria; of these, 30 of 40 had positive UGIseries results and 38 of the 39 infants whohad 24-hour pH monitoring showed
positive results (one patient did notundergo pH monitoring). On the basis ofthese numbers, the sensitivity of a UGIseries would be 75%, compared with 97%for pH monitoring. The sensitivity of a UGIseries appears to be similarly poor amongchildren without respiratory symptoms(89).
These accuracy tests constitute verylow-quality evidence, meaning that theyprovide very low confidence in their results.The poor quality of evidence reflects thefact that the studies did not enrollconsecutive patients, and it was not reportedwhether there was legitimate uncertaintyabout the presence or absence of GER.
Rationale. The primary advantages ofperforming a UGI series rather than 24-houresophageal pH monitoring are that a UGIseries can be performed less invasively and inless time. A less frequent advantage is thatUGI series occasionally demonstratepertinent anatomical abnormalities, such ashiatal hernias or esophageal indentationsuggestive of a vascular ring. Thedisadvantages of a UGI series includeradiation exposure, the need for patientcooperation, and its semiinvasive nature.
The guideline development committeemade the a priori decision that the benefitsof a UGI series would outweigh both thedisadvantages of a UGI series and theconsequences of incorrect results ifthe false-negative rate was less than 10%(i.e., sensitivity greater than 90%) and thefalse-positive rate was less than 10%(i.e., specificity greater than 90%). In otherwords, assuming a prevalence of GER ofroughly 60%, the committee would accept40 false-positive results and 60 false-negative results for every 1,000 patientstested. The acceptable false-negative andfalse-positive rates are both relatively smallbecause 24-hour pH monitoring is notoverly risky or burdensome.
The evidence indicates that thesensitivity (68–79%) and specificity (82%)of UGI series are insufficient to warrant theuse of UGI series as an alternative to24-hour esophageal pH monitoring. Therecommendation against UGI series isconditional because the very low quality ofevidence does not provide sufficientconfidence in the estimated sensitivity andspecificity to be certain that a UGI seriesis not a worthwhile alternative. Themeaning of a conditional recommendationis that it is right for most patients, butmay not be right for a sizable minority
AMERICAN THORACIC SOCIETY DOCUMENTS
American Thoracic Society Documents 367
in certain situations. As an example, aUGI series can be a valuable tool foridentifying vascular rings or slings and maybe considered if such malformationsare suspected. A UGI series can also beconsidered in circumstances in which24-hour pH monitoring is not a practicaloption.
Recommendation 6. For infants withpersistent wheezing that is not relieved bybronchodilators, inhaled corticosteroids,or systemic corticosteroids, we suggest24-hour esophageal pH monitoringrather than a UGI series (conditionalrecommendation, very lowquality of evidence).
Question 7: Should Infants withPersistent Wheezing That Is NotRelieved by Bronchodilators, InhaledCorticosteroids, or SystemicCorticosteroids UndergoGastroesophageal ScintigraphyRather Than 24-Hour Esophageal pHMonitoring?
Summary of evidence. The guidelinedevelopment committee next asked whethergastroesophageal scintigraphy is anacceptable alternative to 24-hour pHmonitoring, which we considered thereference standard. Our systematic reviewdid not identify any randomized trials orcontrolled observational studies thatcompared clinical outcomes among thosewho underwent scintigraphy with those whounderwent 24-hour esophageal pHmonitoring. However, it did identify fourstudies that evaluated gastroesophagealscintigraphic detection of GER in infantsand children with wheezing.
In the only study that used 24-hourpH monitoring as the reference standard,79 children (age, 2–17 yr) who haddifficult-to-control asthma underwent24-hour esophageal pH monitoring.Gastroesophageal scintigraphy identifiedGER with a sensitivity and specificity of15 and 73%, respectively (88). Anotherstudy of infants with wheezing used clinicalhistory and a response to anti-GER therapyas the reference standard instead of24-hour pH monitoring. It found thatgastroesophageal scintigraphy detectedGER with a sensitivity and specificity of58 and 85%, respectively, when a historycompatible with GER was used as thereference standard, and with a sensitivityand specificity of 79 and 50%, respectively,when a response to anti-GER therapy
was used as the reference standard (90).Finally, two studies did not comparegastroesophageal scintigraphy with areference standard, but rather, reportedthat the technique identified GER in 22%of infants and children (age, 3 mo–4 yr)who presented with recurrent wheezingor vomiting (91) and in 26% of infantsand children (age, 6 mo–6 yr) whopresented with difficult-to-treat asthma(92); these yields were lower than the67–100% described previously for 24-houresophageal pH monitoring. The sensitivityof gastroesophageal scintigraphy appears tobe similarly poor among infants andchildren without respiratory symptoms(89).
These accuracy studies constitute verylow quality of evidence, meaning that theyprovide very low confidence in theirestimated effects. The poor quality ofevidence reflects the fact that the studies didnot enroll consecutive patients, and it wasnot reported whether there was legitimateuncertainty about the presence or absence ofGER. Moreover, there was indirectness ofthe population because our focus was onwheezing infants, but many of the studiesenrolled older children.
Rationale. The primary advantages ofgastroesophageal scintigraphy ratherthan 24-hour esophageal pH monitoringare that scintigraphy can be performedless invasively and in less time. Thedisadvantages of scintigraphy areprimarily radiation exposure (albeit lessthan that required for a UGI series) andhigh cost.
The guideline development committeemade an a priori decision that the advantagesof gastroesophageal scintigraphy wouldoutweigh the disadvantages associated withpotential incorrect results if the false-negativerate was less than 10% (i.e., sensitivity greaterthan 90%) and the false-positive rate was lessthan 10% (i.e., specificity greater than 90%).In other words, assuming a prevalence ofGER of roughly 60%, the committee wouldaccept 40 false-positive results and 60 false-negative results for every 1,000 patientstested. The acceptable false-negative andfalse-positive rates are both relatively smallbecause 24-hour pH monitoring is not overlyrisky or burdensome.
The evidence indicates that thesensitivity and specificity (15 and 73%,respectively) of gastroesophagealscintigraphy are insufficient to warrant theuse of scintigraphy as an alternative to
24-hour esophageal pH monitoring. Therecommendation against scintigraphy isconditional because the very low qualityof evidence does not provide sufficientconfidence in the estimated sensitivity andspecificity to be certain that scintigraphy isnot a worthwhile alternative.
Recommendation 7. For infants withpersistent wheezing that is not relieved bybronchodilators, inhaled corticosteroids, orsystemic corticosteroids, we suggest 24-houresophageal pH monitoring rather thangastrointestinal scintigraphy (conditionalrecommendation, very low quality ofevidence).
Question 8: Should Infants withoutNeurologic Pathology with PersistentWheezing That Is Not Relieved byBronchodilators, InhaledCorticosteroids, or SystemicCorticosteroids Undergo aSwallowing Function Study?
Summary of evidence. Our literature reviewdid not identify any randomized trials orcontrolled observational studies thatcompared clinical outcomes amongthose who underwent a swallowing functionstudy versus those who did not. However, itdid identify two case series that reported theprevalence of aspiration detected byvideo-fluoroscopic swallowing functionstudies in infants and children who didnot have chronic illnesses but did haverespiratory symptoms including wheezing(Table 5). Both series also reported theoutcomes of treatment (93, 94).
The first series enrolled 472 infants(age, ,1 yr) with either respiratorysymptoms or vomiting and performedfluoroscopic swallowing studies on each.Swallowing dysfunction was detected in63 of 472 infants (13%). Among theseinfants, 70% had tracheal aspirationand 30% had laryngeal penetration.Because the coordination of swallowingimproves with age among infants withoutchronic illnesses, the infants withswallowing dysfunction were managedby thickening the consistency of theirfood. Tracheal aspiration or laryngealpenetration was seen in 179 swallowingstudies with thin liquids, 61 studies withthickened liquids, and 14 studies withpureed food (93).
The second case series included 112infants (age, ,1 yr) with wheezing orintermittent stridor and performed video-
AMERICAN THORACIC SOCIETY DOCUMENTS
368 American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016
Tab
le5.
Asp
iratio
n:Qua
lityAss
essm
entan
dSum
maryof
Find
ings
Qua
lityAss
essm
ent
Sum
maryofFind
ings
No.
Study
Des
ign
Limitations
Indirec
t-ne
ssInco
nsistenc
yIm
precision
Pub
lication
Bias
Importan
ceof
Outco
me
Qua
lityof
Eviden
ce
Freq
uenc
ywithwhich
video
-fluo
rosc
opic
swallowingstud
ies(gold
stan
dard)iden
tify
aspirationdue
tosw
allowingdys
func
tion
2Cas
ese
ries*
Serious
†Serious
‡Non
eSerious
xNon
eNot
apresp
ecified
outcom
eVerylow
472infants(age
less
than
1yr)with
either
resp
iratory
symptomsor
vomiting
underwen
tfluo
rosc
opic
swallowingstud
ies;
swallowing
dys
func
tionwas
detec
tedin
63of
472infants(13%
)jj
122infants(age
less
than
1yr)with
either
whe
ezingor
interm
itten
tstrid
orun
derwen
tfluo
rosc
opic
swallowingstud
ies;
swallowing
dys
func
tionwas
detec
tedin
13of
112infants(12%
)¶
Whe
ezing(ass
esse
dbythesu
rrogateoutco
meofradiographictrac
heal
aspirationorlaryng
ealp
enetration)
2Cas
ese
ries*
Serious
†Serious
‡Non
eSerious
xNon
eCritical
Verylow
Inaca
sese
riesof
472infants(age
less
than
1yr)w
itheither
resp
iratory
symptomsor
vomiting
,trac
heal
aspira
tionor
laryng
ealp
enetratio
nwas
seen
in17
9sw
allowingstud
ies
with
thin
liquids,
61stud
ieswith
thicke
nedliq
uids,
and14
stud
ies
with
puree
dfood
,aris
kreduc
tionof
morethan
90%
jj
Inaca
sese
riesof
122infants(age
less
than
1yr)with
either
whe
ezing
orinterm
itten
tstrid
or,nine
infants
with
confi
rmed
swallowing
dys
func
tionweretrea
tedwith
athicke
neddiet,while
four
infants
hadtheiroral
feed
ings
stop
ped
and
rece
ived
naso
jejuna
lorga
strostom
yfeed
ings
temporarily.In
allo
fthe
infants,
thesw
allowingdys
func
tion
reso
lved
with
in3–
9mo¶
*References93and94.
†There
wasnoprocess
toensu
rethatpatients
were
consecutivelyorrandomlyincluded;thus,
selectio
nbiasin
favo
rofinfants
clinically
susp
ectedofhavingsw
allowingdysfunctio
nislikely.
‡Thequestionis
specifically
aboutinfants
with
wheezing;however,both
case
serie
sincludedinfants
with
symptomsotherthanwheezing.
x There
were
only584patients
inthetw
ocase
serie
scombined.
jj Reference93.
¶Reference94.
AMERICAN THORACIC SOCIETY DOCUMENTS
American Thoracic Society Documents 369
fluoroscopic swallowing function studies oneach. Swallowing dysfunction was detectedin 13 of 112 infants (12%). Nine infantswere treated with a thickened diet, and fourinfants had their oral feedings stopped andreceived nasojejunal or gastrostomyfeedings temporarily. In all of the infants,the swallowing dysfunction resolved within3–9 months (94).
Taken together, the evidence suggeststhat swallowing dysfunction, which isknown to cause wheezing, can be identifiedby video-fluoroscopic swallowing studiesin 10–15% of infants who do not havea chronic illness but have respiratorysymptoms. More than 90% of such patientswill improve with feeding interventionswhile waiting for the swallowingcoordination to improve with age. Thus,9–14% of patients who undergo video-fluoroscopic swallowing studies mayderive some benefit. The committee hasvery low confidence (i.e., quality ofevidence) in the accuracy of theseestimated effects because the studydesigns were case series (i.e., theywere uncontrolled); and therewas risk for indirectness (i.e., mostseries looked at infants who had avariety of respiratory symptoms, notspecifically wheezing).
Rationale. A video-fluoroscopicswallowing study confers several potentialbenefits. Finding swallowing dysfunctionusually leads to feeding modifications thatreduce aspiration by approximately 90%;a reduction in aspiration is a surrogateoutcome for persistent wheezing, stridor,cough, and pneumonia. Other benefitsinclude relief from the burden, cost, andpotential harms of further diagnostic testing;probable reductions in the use of ineffectivemedications (bronchodilators or inhaledcorticosteroids) and the frequency ofphysician visits; and parental reassurancegiven the high likelihood that the conditionwill spontaneously resolve. Limitationsinclude the need for infant/child
cooperation, cost, availability of speechpathologist, and the risk of aspirationduring the study. The committee judgedthat the desirable consequences outweighthe undesirable consequences and,therefore, suggests that infants withpersistent wheezing that has notresponded to conventional therapiesundergo a video-fluoroscopic swallowingstudy. The recommendation isconditional because the very low qualityof evidence provides little certainty thatthe benefits of a video-fluoroscopicswallowing study exceed the burdens,costs, and harms.
Recommendation 8. For infantswithout neurologic pathology withpersistent wheezing that is not relieved bybronchodilators, inhaled corticosteroids, orsystemic corticosteroids, we suggest aswallowing function study to evaluate foraspiration (weak recommendation, very lowquality of evidence).
Limitations and FutureDirections
A common theme throughout ourguideline development was the strikingpaucity of data regarding infantilewheezing. Despite how widespread andcommon this clinical problem is, we wereunable to find any large clinical studies thatused consistent case definitions andoutcomes. Most of the studies cited werecase series, providing the lowest quality ofevidence on the GRADE scale. Given thefrequency with which infantile wheezingoccurs, there is an urgent need for morerigorous research to be conducted in thisfield.
Although we used the GRADEmethodology, we rarely had patient-important outcomes that could bereliably linked to performance of thevarious diagnostic tests. As a result,we presumed that treatment strategies
based on a positive test would providetherapeutic benefit to the patient, butthis presumption and limited evidencereduced our ability to make strongrecommendations.
One clear need for future research isto determine whether implementation ofthese tests actually leads to treatmentthat improves patient-important outcomes.Outcome measures should include bothclinical responses and parental preferences,particularly regarding choices betweendiagnostic testing and empiric treatment.However, study design is complicated by thefact that a substantial fraction of infants withpersistent wheeze not responsive to standardtherapies have anatomic abnormalities thatmay not respond to any medical therapy.Routine incorporation of bronchoscopyinto clinical trials could address this issue, butlikely would be problematic given therelatively high costs and risks associated withthis procedure.
This issue highlights the fact thatmany current tests involve substantialcosts and/or risks that limit widespreaduse. Further research should addresswhether diagnosis could be achieved byless invasive tests, radiologic studies inlieu of bronchoscopy for anatomicabnormalities, or analysis of exhaled breathto detect markers of airway infection orreflux. Comparative effectiveness studiesand the development of clinicalpathways would also help cliniciansbetter evaluate infants with persistentwheezing.
In summary, this document providesguidelines that further two goals ofinterest to the ATS. First, they will aid thepediatric generalist or respiratoryspecialist in the management of the infantwith recurrent or persistent wheeze thatdoes not respond to conventionaltherapies. Second, they will serve toidentify the research needed to improvediagnosis and treatment of this vulnerablepopulation. n
This clinical practice guideline was prepared by an ad hoc subcommittee of the ATS Assembly on Pediatrics.
Members of the subcommittee are asfollows:
CLEMENT L. REN, M.D. (Co-Chair)CHARLES R. ESTHER, JR., M.D., PH.D. (Co-Chair)JASON S. DEBLEY, M.D., M.P.H.MARIANNA SOCKRIDER, M.D., M.P.H., D.P.H.OZGE YILMAZ, M.D.
NIKHIL AMIN, M.D.ALIA BAZZY-ASAAD, M.D.STEPHANIE D. DAVIS, M.D.MANUEL DURAND, M.D.JEFFREY M. EWIG, M.D.HASAN YUKSEL, M.D.ENRICO LOMBARDI, M.D.
TERRY L. NOAH, M.D.PEGGY RADFORD, M.D.SARATH RANGANATHAN, M.D.ALEJANDRO TEPER, M.D.MILES WEINBERGER, M.D.JAN BROZEK, M.D., PH.D.KEVIN C. WILSON, M.D.
AMERICAN THORACIC SOCIETY DOCUMENTS
370 American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016
Author Disclosures: O.Y. receivedconference travel support from AllergopharmaTurkey and GlaxoSmithKline. N.A. waspreviously employed by MannKind Corp. andhas been an employee of RegeneronPharmaceuticals since January 2016(subsequent to guideline completion). S.D.D.was on an advisory committee and aconsultant for Vertex Pharmaceuticals, wasa consultant for Eli Lilly and Co., and was aspeaker for ABCOMM in an activity supportedby Gilead. H.Y. was a speaker for
GlaxoSmithKline and Merck Sharp Dohme,and received conference travel support fromAllergopharma, GlaxoSmithKline, and Nutricia.E.L. was a speaker and on a data safetyand monitoring board for Chiesi, receivedconference travel support from Lusofarmaco,was on an advisory committee and a speakerfor Novartis, was a speaker for Sigma-Tau,and participated in meetings organizedby his institution that were sponsored byCareFusion, Cosmed, Italchimici, StewartItalia, and Valeas. C.L.R., C.R.E., J.S.D.,
M.S., A.B-A., M.D., J.M.E., T.L.N., P.R., S.R.,A.T., M.W., J.B., and K.C.W. reported norelationships with relevant commercialinterests.
Acknowledgment: The committee wouldlike to acknowledge the invaluableefforts of Angela Dixon from the Universityof Rochester (Rochester, NY) for herwork as the medical librarian on thisproject.
References
1. Martinez FD, Wright AL, Taussig LM, Holberg CJ, Halonen M, MorganWJ; Group Health Medical Associates. Asthma and wheezing in thefirst six years of life. N Engl J Med 1995;332:133–138.
2. Devulapalli CS, Carlsen KC, Haland G, Munthe-Kaas MC, Pettersen M,Mowinckel P, Carlsen KH. Severity of obstructive airways diseaseby age 2 years predicts asthma at 10 years of age. Thorax 2008;63:8–13.
3. Bacharier LB, Phillips BR, Bloomberg GR, Zeiger RS, Paul IM,Krawiec M, Guilbert T, Chinchilli VM, Strunk RC; ChildhoodAsthma Research and Education Network, National Heart, Lung,and Blood Institute. Severe intermittent wheezing in preschoolchildren: a distinct phenotype. J Allergy Clin Immunol 2007;119:604–610.
4. National Asthma Education and Prevention Program. Expert PanelReport 3 (EPR-3): guidelines for the diagnosis and management ofasthma—summary report 2007. J Allergy Clin Immunol 2007;120(5,Suppl):S94–S138. [Published erratum appears in J Allergy ClinImmunol 121:1330.]
5. Schunemann HJ, Jaeschke R, Cook DJ, Bria WF, El-Solh AA, Ernst A,Fahy BF, Gould MK, Horan KL, Krishnan JA, et al.; ATS DocumentsDevelopment and Implementation Committee. An official ATSstatement: grading the quality of evidence and strength ofrecommendations in ATS guidelines and recommendations. Am JRespir Crit Care Med 2006;174:605–614.
6. Skoner D, Caliguiri L. The wheezing infant. Pediatr Clin North Am 1988;35:1011–1030.
7. Martinez FD, Morgan WJ, Wright AL, Holberg CJ, Taussig LM. Diminishedlung function as a predisposing factor for wheezing respiratory illness ininfants. N Engl J Med 1988;319:1112–1117.
8. Guerra S, Lohman IC, Halonen M, Martinez FD, Wright AL. Reducedinterferon g production and soluble CD14 levels in early life predictrecurrent wheezing by 1 year of age. Am J Respir Crit Care Med 2004;169:70–76.
9. Kliegman RM, editor. Nelson textbook of pediatrics, 20th ed.Philadelphia, PA: Elsevier; 2016.
10. Brand PL, Baraldi E, Bisgaard H, Boner AL, Castro-Rodriguez JA,Custovic A, de Blic J, de Jongste JC, Eber E, Everard ML, et al.Definition, assessment and treatment of wheezing disorders inpreschool children: an evidence-based approach. Eur Respir J 2008;32:1096–1110.
11. Ducharme FM, Tse SM, Chauhan B. Diagnosis, management,and prognosis of preschool wheeze. Lancet 2014;383:1593–1604.
12. Aslan AT, Kiper N, Dogru D, Karagoz AH, Ozcelik U, Yalcin E.Diagnostic value of flexible bronchoscopy in children withpersistent and recurrent wheezing. Allergy Asthma Proc 2005;26:483–486.
13. Baraldi E, Donega S, Carraro S, Farina M, Barbato A, Cutrone C.Tracheobronchomalacia in wheezing young children poorlyresponsive to asthma therapy. Allergy 2010;65:1064–1065.
14. Cakir E, Ersu RH, Uyan ZS, Oktem S, Karadag B, Yapar O, Pamukcu O,Karakoc F, Dagli E. Flexible bronchoscopy as a valuable tool inthe evaluation of persistent wheezing in children. Int J PediatrOtorhinolaryngol 2009;73:1666–1668.
15. Hauk PJ, Krawiec M, Murphy J, Boguniewicz J, Schiltz A, Goleva E, LiuAH, Leung DY. Neutrophilic airway inflammation and associationwith bacterial lipopolysaccharide in children with asthma andwheezing. Pediatr Pulmonol 2008;43:916–923.
16. Le Bourgeois M, Goncalves M, Le Clainche L, Benoist MR, FournetJC, Scheinmann P, de Blic J. Bronchoalveolar cells in children , 3years old with severe recurrent wheezing. Chest 2002;122:791–797.
17. Masters IB, Chang AB, Patterson L, Wainwright C, Buntain H, DeanBW, Francis PW. Series of laryngomalacia, tracheomalacia,and bronchomalacia disorders and their associations withother conditions in children. Pediatr Pulmonol 2002;34:189–195.
18. Saglani S, Nicholson AG, Scallan M, Balfour-Lynn I, Rosenthal M,Payne DN, Bush A. Investigation of young children with severerecurrent wheeze: any clinical benefit? Eur Respir J 2006;27:29–35.
19. Saito J, Harris WT, Gelfond J, Noah TL, Leigh MW, Johnson R, DavisSD. Physiologic, bronchoscopic, and bronchoalveolar lavage fluidfindings in young children with recurrent wheeze and cough. PediatrPulmonol 2006;41:709–719.
20. Ullmann N, Sacco O, Gandullia P, Silvestri M, Pistorio A, Barabino A,Disma NM, Rossi GA. Usefulness and safety of double endoscopy inchildren with gastroesophageal reflux and respiratory symptoms.Respir Med 2010;104:593–599.
21. De Baets F, De Schutter I, Aarts C, Haerynck F, Van Daele S, DeWachter E, Malfroot A, Schelstraete P. Malacia, inflammation andbronchoalveolar lavage culture in children with persistent respiratorysymptoms. Eur Respir J 2012;39:392–395.
22. Laberge J-M, Puligandla P. Congenital malformations of the lungs andairways. In: Taussig LM, Landau LI, Le Souef PN, Martinez FD,Morgan WJ, Sly PD, eds. Pediatric respiratory medicine, 2nd ed.Philadelphia, PA: Mosby; 2008. pp. 907–942.
23. Davis S, Jones M, Kisling J, Angelicchio C, Tepper RS. Effect ofcontinuous positive airway pressure on forced expiratory flowsin infants with tracheomalacia. Am J Respir Crit Care Med 1998;158:148–152.
24. Wiseman NE, Duncan PG, Cameron CB. Management oftracheobronchomalacia with continuous positive airway pressure. JPediatr Surg 1985;20:489–493.
25. Reiterer F, Eber E, Zach MS, Muller W. Management of severecongenital tracheobronchomalacia by continuous positive airwaypressure and tidal breathing flow–volume loop analysis. PediatrPulmonol 1994;17:401–403.
26. Pizer BL, Freeland AP, Wilkinson AR. Prolonged positive airwaypressure for severe neonatal tracheobronchomalacia. Arch Dis Child1986;61:908–909.
27. Miller RW, Pollack MM, Murphy TM, Fink RJ. Effectiveness ofcontinuous positive airway pressure in the treatment ofbronchomalacia in infants: a bronchoscopic documentation. CritCare Med 1986;14:125–127.
28. Ferguson GT, Benoist J. Nasal continuous positive airway pressure inthe treatment of tracheobronchomalacia. Am Rev Respir Dis 1993;147:457–461.
29. Neijens HJ, Kerrebijn KF, Smalhout B. Successful treatment with CPAPof two infants with bronchomalacia. Acta Paediatr Scand 1978;67:293–296.
AMERICAN THORACIC SOCIETY DOCUMENTS
American Thoracic Society Documents 371
30. Panitch HB, Allen JL, Alpert BE, Schidlow DV. Effects of CPAP on lungmechanics in infants with acquired tracheobronchomalacia. Am JRespir Crit Care Med 1994;150:1341–1346.
31. Blair GK, Cohen R, Filler RM. Treatment of tracheomalacia: eight years’experience. J Pediatr Surg 1986;21:781–785.
32. Kamata S, Usui N, Sawai T, Nose K, Kitayama Y, Okuyama H, Okada A.Pexis of the great vessels for patients with tracheobronchomalacia ininfancy. J Pediatr Surg 2000;35:454–457.
33. Greenholz SK, Karrer FM, Lilly JR. Contemporary surgery oftracheomalacia. J Pediatr Surg 1986;21:511–514.
34. Morabito A, MacKinnon E, Alizai N, Asero L, Bianchi A. The anteriormediastinal approach for management of tracheomalacia. J PediatrSurg 2000;35:1456–1458.
35. Abdel-Rahman U, Ahrens P, Fieguth HG, Kitz R, Heller K, Moritz A.Surgical treatment of tracheomalacia by bronchoscopic monitoredaortopexy in infants and children. Ann Thorac Surg 2002;74:315–319.
36. Bullard KM, Scott Adzick N, Harrison MR. A mediastinal windowapproach to aortopexy. J Pediatr Surg 1997;32:680–681.
37. Delgado MD, Matute JA, Jimenez MA, Aguado P, Benavent MI,Filler RM, Berchi FJ. [The treatment of the tracheobronchomalaciain pediatric age] [article in Spanish]. Cir Pediatr 1997;10:65–69.
38. Tazuke Y, Kawahara H, Yagi M, Yoneda A, Soh H, Maeda K,Yamamoto T, Imura K. Use of a Palmaz stent for tracheomalacia:case report of an infant with esophageal atresia. J Pediatr Surg 1999;34:1291–1293.
39. Filler RM, Forte V, Chait P. Tracheobronchial stenting for the treatmentof airway obstruction. J Pediatr Surg 1998;33:304–311.
40. Furman RH, Backer CL, Dunham ME, Donaldson J, Mavroudis C,Holinger LD. The use of balloon-expandable metallic stents in thetreatment of pediatric tracheomalacia and bronchomalacia. ArchOtolaryngol Head Neck Surg 1999;125:203–207.
41. Casiano RR, Numa WA, Nurko YJ. Efficacy of transoral intraluminalWallstents for tracheal stenosis or tracheomalacia. Laryngoscope2000;110:1607–1612.
42. Tsugawa C, Nishijima E, Muraji T, Yoshimura M, Tsubota N, Asano H.A shape memory airway stent for tracheobronchomalacia inchildren: an experimental and clinical study. J Pediatr Surg 1997;32:50–53.
43. Bairdain S, Smithers CJ, Hamilton TE, Zurakowski D, Rhein L, FokerJE, Baird C, Jennings RW. Direct tracheobronchopexy to correctairway collapse due to severe tracheobronchomalacia: short-termoutcomes in a series of 20 patients. J Pediatr Surg 2015;50:972–977.
44. Shah RK, Mora BN, Bacha E, Sena LM, Buonomo C, Del Nido P,Rahbar R. The presentation and management of vascular rings: anotolaryngology perspective. Int J Pediatr Otorhinolaryngol 2007;71:57–62.
45. Greenhill ED, Skinner K. Impaired nursing students: an interventionprogram. J Nurs Educ 1991;30:379–381.
46. Woods RK, Sharp RJ, Holcomb GW III, Snyder CL, Lofland GK,Ashcraft KW, Holder TM. Vascular anomalies andtracheoesophageal compression: a single institution’s 25-yearexperience. Ann Thorac Surg 2001;72:434–438, discussion 438–439.
47. Koontz CS, Bhatia A, Forbess J, Wulkan ML. Video-assistedthoracoscopic division of vascular rings in pediatric patients. AmSurg 2005;71:289–291.
48. Al-Bassam A, Saquib Mallick M, Al-Qahtani A, Al-Tokhais T, Gado A,Al-Boukai A, Thalag A, Alsaadi M. Thoracoscopic division of vascularrings in infants and children. J Pediatr Surg 2007;42:1357–1361.
49. Suematsu Y, Mora BN, Mihaljevic T, del Nido PJ. Totally endoscopicrobotic-assisted repair of patent ductus arteriosus and vascular ringin children. Ann Thorac Surg 2005;80:2309–2313.
50. Kogon BE, Forbess JM, Wulkan ML, Kirshbom PM, Kanter KR. Video-assisted thoracoscopic surgery: is it a superior technique for thedivision of vascular rings in children? Congenit Heart Dis 2007;2:130–133.
51. Panitch HB, Keklikian EN, Motley RA, Wolfson MR, Schidlow DV. Effectof altering smooth muscle tone on maximal expiratory flows inpatients with tracheomalacia. Pediatr Pulmonol 1990;9:170–176.
52. Rappaport BA, Suresh S, Hertz S, Evers AS, Orser BA. Anestheticneurotoxicity—clinical implications of animal models. N Engl J Med2015;372:796–797.
53. Schellhase DE, Fawcett DD, Schutze GE, Lensing SY, Tryka AF.Clinical utility of flexible bronchoscopy and bronchoalveolar lavage inyoung children with recurrent wheezing. J Pediatr 1998;132:312–318.
54. Marguet C, Jouen-Boedes F, Dean TP, Warner JO. Bronchoalveolarcell profiles in children with asthma, infantile wheeze, chroniccough, or cystic fibrosis. Am J Respir Crit Care Med 1999;159:1533–1540.
55. Fayon M, Just J, Thien HV, Chiba T, Pascual L, Sandouk G, Grimfeld A.Bacterial flora of the lower respiratory tract in children with bronchialasthma. Acta Paediatr 1999;88:1216–1222.
56. Marguet C, Dean TP, Warner JO. Soluble intercellular adhesionmolecule-1 (sICAM-1) and interferon-g in bronchoalveolar lavagefluid from children with airway diseases. Am J Respir Crit Care Med2000;162:1016–1022.
57. Krawiec ME, Westcott JY, Chu HW, Balzar S, Trudeau JB, SchwartzLB, Wenzel SE. Persistent wheezing in very young children isassociated with lower respiratory inflammation. Am J Respir CritCare Med 2001;163:1338–1343.
58. Arnoux B, Bousquet J, Rongier M, Scheinmann P, de Blic J. Increasedbronchoalveolar lavage CD8 lymphocyte subset population inwheezy infants. Pediatr Allergy Immunol 2001;12:194–200.
59. Nagayama Y, Tsubaki T, Toba T, Nakayama S, Kiyofumi O. Analysis ofsputum taken from wheezy and asthmatic infants and children, withspecial reference to respiratory infections. Pediatr Allergy Immunol2001;12:318–326.
60. Chang AB, Cox NC, Faoagali J, Cleghorn GJ, Beem C, Ee LC,Withers GD, Patrick MK, Lewindon PJ. Cough and refluxesophagitis in children: their co-existence and airway cellularity.BMC Pediatr 2006;6:4.
61. Maclennan C, Hutchinson P, Holdsworth S, Bardin PG, Freezer NJ.Airway inflammation in asymptomatic children with episodic wheeze.Pediatr Pulmonol 2006;41:577–583.
62. Marchant JM, Masters IB, Taylor SM, Cox NC, Seymour GJ, Chang AB.Evaluation and outcome of young children with chronic cough. Chest2006;129:1132–1141.
63. Donnelly D, Critchlow A, Everard ML. Outcomes in children treated forpersistent bacterial bronchitis. Thorax 2007;62:80–84.
64. Byrnes C, Edwards E. Outcomes in children treated for persistentbacterial bronchitis. Thorax 2007;62:922–923, author reply 923.
65. Marchant JM, Gibson PG, Grissell TV, Timmins NL, Masters IB, ChangAB. Prospective assessment of protracted bacterial bronchitis:airway inflammation and innate immune activation. Pediatr Pulmonol2008;43:1092–1099.
66. Khoshoo V, Edell D, Mohnot S, Haydel R Jr, Saturno E, Kobernick A.Associated factors in children with chronic cough. Chest 2009;136:811–815.
67. Rosen R, Johnston N, Hart K, Khatwa U, Katz E, Nurko S. Higher rate ofbronchoalveolar lavage culture positivity in children with nonacidreflux and respiratory disorders. J Pediatr 2011;159:504–506.
68. Zgherea D, Pagala S, Mendiratta M, Marcus MG, Shelov SP,Kazachkov M. Bronchoscopic findings in children with chronic wetcough. Pediatrics 2012;129:e364–e369.
69. Kompare M, Weinberger M. Protracted bacterial bronchitis in youngchildren: association with airway malacia. J Pediatr 2012;160:88–92.
70. Chang AB, Redding GJ, Everard ML. Chronic wet cough: protractedbronchitis, chronic suppurative lung disease and bronchiectasis.Pediatr Pulmonol 2008;43:519–531.
71. Marchant J, Masters IB, Champion A, Petsky H, Chang AB.Randomised controlled trial of amoxycillin clavulanate in childrenwith chronic wet cough. Thorax 2012;67:689–693.
72. Gibson MK, Crofts TS, Dantas G. Antibiotics and the developinginfant gut microbiota and resistome. Curr Opin Microbiol 2015;27:51–56.
73. Debley JS, Stamey DC, Cochrane ES, Gama KL, Redding GJ. Exhalednitric oxide, lung function, and exacerbations in wheezy infants andtoddlers. J Allergy Clin Immunol 2010;125:1228–1234.e13.
AMERICAN THORACIC SOCIETY DOCUMENTS
372 American Journal of Respiratory and Critical Care Medicine Volume 194 Number 3 | August 1 2016
74. Elliott M, Heltshe SL, Stamey DC, Cochrane ES, Redding GJ, DebleyJS. Exhaled nitric oxide predicts persistence of wheezing,exacerbations, and decline in lung function in wheezy infants andtoddlers. Clin Exp Allergy 2013;43:1351–1361.
75. Kurland G, Deterding RR, Hagood JS, Young LR, Brody AS, Castile RG,Dell S, Fan LL, Hamvas A, Hilman BC, et al.; American ThoracicSociety Committee on Childhood Interstitial Lung Disease (chILD)and the chILD Research Network. An official American ThoracicSociety clinical practice guideline: classification, evaluation, andmanagement of childhood interstitial lung disease in infancy. Am JRespir Crit Care Med 2013;188:376–394.
76. Boyce JA, Assa’ad A, Burks AW, Jones SM, Sampson HA, Wood RA,Plaut M, Cooper SF, Fenton MJ, Arshad SH, et al. Guidelines for thediagnosis and management of food allergy in the United States:summary of the NIAID-sponsored expert panel report. Nutr Res2011;31:61–75.
77. Miskelly FG, Burr ML, Vaughan-Williams E, Fehily AM, Butland BK,Merrett TG. Infant feeding and allergy. Arch Dis Child 1988;63:388–393.
78. Chan YH, Shek LP, Aw M, Quak SH, Lee BW. Use of hypoallergenicformula in the prevention of atopic disease among Asian children.J Paediatr Child Health 2002;38:84–88.
79. Tromp II, Kiefte-de Jong JC, Lebon A, Renders CM, Jaddoe VW,Hofman A, de Jongste JC, Moll HA. The introduction of allergenicfoods and the development of reported wheezing and eczema inchildhood: the Generation R study. Arch Pediatr Adolesc Med 2011;165:933–938.
80. Burr ML, Limb ES, Maguire MJ, Amarah L, Eldridge BA, Layzell JC,Merrett TG. Infant feeding, wheezing, and allergy: a prospectivestudy. Arch Dis Child 1993;68:724–728.
81. James JM, Bernhisel-Broadbent J, Sampson HA. Respiratory reactionsprovoked by double-blind food challenges in children. Am J RespirCrit Care Med 1994;149:59–64.
82. Buts JP, Barudi C, Moulin D, Claus D, Cornu G, Otte JB. Prevalenceand treatment of silent gastro-oesophageal reflux in children withrecurrent respiratory disorders. Eur J Pediatr 1986;145:396–400.
83. Yuksel H, Yilmaz O, Kirmaz C, Aydoğdu S, Kasirga E. Frequency ofgastroesophageal reflux disease in nonatopic children with asthma-like airway disease. Respir Med 2006;100:393–398.
84. Eid NS, Shepherd RW, Thomson MA. Persistent wheezing andgastroesophageal reflux in infants. Pediatr Pulmonol 1994;18:39–44.
85. Berquist WE, Rachelefsky GS, Kadden M, Siegel SC, Katz RM,Fonkalsrud EW, Ament ME. Gastroesophageal reflux–associatedrecurrent pneumonia and chronic asthma in children. Pediatrics1981;68:29–35.
86. Ghezzi M, Silvestri M, Guida E, Pistorio A, Sacco O, Mattioli G, JasonniV, Rossi GA. Acid and weakly acid gastroesophageal refluxes andtype of respiratory symptoms in children. Respir Med 2011;105:972–978.
87. Fohl AL, Regal RE. Proton pump inhibitor–associated pneumonia: not abreath of fresh air after all? World J Gastrointest Pharmacol Ther2011;2:17–26.
88. Balson BM, Kravitz EK, McGeady SJ. Diagnosis and treatment ofgastroesophageal reflux in children and adolescents with severeasthma. Ann Allergy Asthma Immunol 1998;81:159–164.
89. Arasu TS, Wyllie R, Fitzgerald JF, Franken EA, Siddiqui AR, Lehman GA,Eigen H, Grosfeld JL. Gastroesophageal reflux in infants and childrencomparative accuracy of diagnostic methods. J Pediatr 1980;96:798–803.
90. Patra S, Singh V, Chandra J, Kumar P, Tripathi M. Diagnosticmodalities for gastro-esophageal reflux in infantile wheezers. J TropPediatr 2011;57:99–103.
91. Karaman O, Uzuner N, Değirmenci B, Uğuz A, Durak H. Results of thegastroesophageal reflux assessment in wheezy children. Indian JPediatr 1999;66:351–355.
92. Thomas EJ, Kumar R, Dasan JB, Kabra SK, Bal CS, Menon S,Malhothra A. Gastroesophageal reflux in asthmatic childrennot responding to asthma medication: a scintigraphic studyin 126 patients with correlation between scintigraphic andclinical findings of reflux. Clin Imaging 2003;27:333–336.
93. Mercado-Deane MG, Burton EM, Harlow SA, Glover AS,Deane DA, Guill MF, Hudson V. Swallowing dysfunction ininfants less than 1 year of age. Pediatr Radiol 2001;31:423–428.
94. Sheikh S, Allen E, Shell R, Hruschak J, Iram D, Castile R, McCoy K.Chronic aspiration without gastroesophageal reflux as a cause ofchronic respiratory symptoms in neurologically normal infants. Chest2001;120:1190–1195.
AMERICAN THORACIC SOCIETY DOCUMENTS
American Thoracic Society Documents 373