ARTICLE

Diagnosis and Testing in Bronchiolitis

A Systematic Review
W. Clayton Bordley, MD, MPH; Meera Viswanathan, PhD; Valerie J. King, MD, MPH;
Sonya F. Sutton, BSPH; Anne M. Jackman, MSW; Laura Sterling, MD, MPH; Kathleen N. Lohr, PhD

Background: The diagnosis of bronchiolitis is based on are reports of treatment or prevention trials). Numer-
typical history and results of a physical examination. The ous studies demonstrate that rapid respiratory syncytial
indications for and utility of diagnostic and supportive virus tests have acceptable sensitivity and specificity, but
laboratory testing (eg, chest x-ray films, complete blood no data show that respiratory syncytial virus testing af-
cell counts, and respiratory syncytial virus testing) are fects clinical outcomes in typical cases of the disease. Sev-
unclear. enteen studies presented chest x-ray film data. Abnor-
malities on chest x-ray films ranged from 20% to 96%.
Objectives: To review systematically the data on diag- Insufficient data exist to show that chest x-ray films re-
nostic and supportive testing in the management of bron- liably distinguish between viral and bacterial disease or
chiolitis and to assess the utility of such testing. predict severity of disease. Ten studies included com-
plete blood cell counts, but most did not present spe-
Design: In conjunction with an expert panel, we gen- cific results. In one study, white blood cell counts cor-
erated admissibility criteria and derived relevant terms related with radiologically defined disease categories of
to search the literature published from 1980 to Novem- bronchiolitis.
ber 2002 in MEDLINE and the Cochrane Collaboration
Database of Controlled Clinical Trials. Trained abstrac- Conclusions: A large number of studies include diag-
tors completed detailed data collection forms for each ar- nostic and supportive testing data. However, these stud-
ticle. We summarized the data in tables after perform- ies do not define clear indications for such testing or the
ing data integrity checks. impact of testing on relevant patient outcomes. Given the
high prevalence of this disease, prospective studies of the
Results: Of the 797 abstracts identified, we present evi- utility of such testing are needed and feasible.
dence from 82 trials that met our inclusion criteria (17
are primary articles on diagnosis of bronchiolitis and 65 Arch Pediatr Adolesc Med. 2004;158:119-126

B
RONCHIOLITIS IS THE MOST bronchiolitis (eg, viral culture, immuno-
common lower respiratory fluorescence, and enzyme-linked immu-
tract infection in infants. Vir- nosorbent assays for RSV). The use of test-
tually all children have been ing is typically justified for 1 of the
exposed to respiratory syn- following reasons: ruling out other diag-
cytial virus (RSV), the cause of most bron- noses (eg, congestive heart failure or bac-
chiolitis cases, by their second birthday. terial pneumonia), first-time wheezing, co-
From the Departments of horting of hospitalized patients, deciding
Up to 3% of all children are hospitalized
Pediatrics and Surgery, Duke
with bronchiolitis in their first year of life.1 on treatment (eg, ribavirin), including pa-
University Medical Center,
Durham, NC (Dr Bordley); The diagnosis of bronchiolitis is based pri- tients in research protocols, or perform-
the Cecil G. Sheps Center for marily on typical history and results of a ing public health surveillance.
Health Services Research physical examination.2 Despite the high
(Drs Bordley, King, Sterling, prevalence of bronchiolitis, little consen- See also pages 111 and 127
and Lohr and Ms Jackman), the sus exists on the optimal management of
Department of Family Medicine the disease.3 There is significant varia- The clinical utility of specific etio-
(Drs King and Sterling), and tion in the use of supportive testing and logic testing in cases of bronchiolitis is un-
the School of Public Health
treatment of bronchiolitis.4,5 clear. Complete blood cell counts have
(Dr Lohr), University of North
Carolina at Chapel Hill; and
A variety of laboratory studies can poor test characteristics for determining
RTI International, Research provide supportive data for diagnosis. Ex- bacterial disease.6 Chest x-ray film find-
Triangle Park, NC amples include chest x-ray films, com- ings for bronchiolitis and pneumonia are
(Dr Viswanathan plete blood cell (CBC) counts, and spe- variable and nonspecific.7,8 Knowing that
and Ms Sutton). cific testing to determine the cause of RSV is the cause of bronchiolitis does little

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 158, FEB 2004 WWW.ARCHPEDIATRICS.COM
119

©2004 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Dahlgren Memorial Library-Georgetown University Medical Center User on 08/20/2015
 Table 1. Inclusion and Exclusion Criteria Table 2. Search Terms

Category Criteria Topic Search Terms

Study population Humans, infants, and children Exploded terms for Bronchiolitis, diagnosis, differential
Study settings and Inpatient, outpatient, home; all diagnosis diagnosis, thoracic radiography,
geography geographical locations subject to laboratory techniques, and procedures
publication language and study Exploded terms for Steroidal anti-inflammatory agents,
design criteria treatment steroids, bronchodilator agents,
Time period Systematic reviews, from 1966-2002; antiviral agents, antimicrobial cationic
individual studies, published from peptides, antibiotics, antimicrobials,
1980 through 2002* and anti-infective agents
Publication languages English only Exploded terms for Primary prevention, immunoglobulins,
Admissible evidence (study Original research studies that provide prophylaxis bronchiolitis (prevention and control),
design and other criteria) sufficient detail regarding methods isolation strategies, and patient
and results to enable use and isolation
abstraction of the data and results Study design for diagnosis Prospective studies, longitudinal studies,
For studies on RCTs with double-blinded, and cohort studies
diagnosis single-blinded, and crossover Study design for treatment Randomized controlled trial, single-blind
designs; non-RCTs with prospective and prophylaxis method, double-blind method, random
cohort designs allocation, and meta-analysis
For studies on RCTs, with double-blinded, Outcomes for diagnosis Fatal outcome, outcome and process
treatment and single-blinded, and crossover assessment (health care), outcome
prophylaxis designs; sample size appropriate for assessment (health care), and
the study question addressed (ie, treatment outcome
case reports or small case series; Outcomes for treatment Morbidity, mortality, and adverse effects
with ⬍10 subjects excluded) and prophylaxis or harms
Limiting terms for all Human, years 1980-2002, newborn
Abbreviation: RCT, randomized controlled trial. infant (age, birth to 1 mo), infant (age,
*1980 start point was based on consensus of the expert panel. 1-23 mo), or preschool child
(age, 2-5 y)

to change the clinical course, the management, or the

prognosis. Supportive testing, nevertheless, is common
and is associated with significant cost in the care of in- RESULTS
fants with this disease.9 Some institutions have devel-
oped evidence-based guidelines specifically to decrease We reviewed 797 abstracts identified using the search strat-
the use of RSV enzyme-linked immunosorbent assays and egy. Of these, 17 are primary articles on diagnosis of bron-
supportive diagnostic testing.10 chiolitis. None of these studies was designed specifically
The present study was part of a larger systematic to measure the utility of diagnostic or supportive testing.
review of the literature on the diagnosis and treatment However, considerable data on diagnosis and testing were
of bronchiolitis. We herein attempt to determine the found in the 65 treatment and prevention trials identi-
effectiveness of diagnostic tools and supportive testing fied, so these are also included in our results.
for diagnosing bronchiolitis in infants. A companion The studies dealing with diagnosis fell into the fol-
report presents the data on the treatment and preven- lowing 5 categories: (1) case definitions and inclusion
tion of bronchiolitis.11 criteria used in the clinical trials; (2) viral causes of bron-
chiolitis when all subjects underwent testing; (3) com-
METHODS parison of various virus isolation techniques; (4) predic-
tors of disease severity, complications, or both; and (5)
In conjunction with an expert panel, we generated inclusion studies in which standardized tests were performed on
and exclusion criteria (Table 1) and derived relevant terms all patients as part of their evaluation (eg, chest x-rays
(Table 2) to search the literature in MEDLINE and the Coch-
and CBC counts).
rane Collaboration Database of Controlled Clinical Trials. For
all studies, key inclusion criteria consisted of outcomes that were
clinically relevant and could be abstracted. Meta-analyses were CASE DEFINITION AND INCLUSION CRITERIA
included in the search to examine their lists of included and
excluded studies. We conducted hand searches of the refer- The challenge of this literature is the fact that bronchi-
ence lists of relevant included articles to ensure that we did not olitis is a clinical diagnosis based on typical history and
exclude important work. In addition, we consulted with the findings on physical examination. There is no specific
technical expert advisory group about any studies that were un- diagnostic test or gold standard that confirms the diag-
der way but not yet published. Our search was last updated in nosis or excludes other diseases that may be clinically
November 2002. Two additional studies published during this similar (eg, bacterial pneumonia). We reviewed the case
this article’s review process were included.12,13
definitions and inclusion criteria used in the clinical trials
Trained abstractors completed detailed data collection forms
for each included study. We summarized the information in and found that most definitions were quite similar. Forty-
tables after reviewing the data collection forms against the ar- three trials used tachypnea in the case definition or in-
ticles. Senior study personnel (W.C.B. and V.J.K.) performed clusion criteria; 42 used wheezing; 37 used oxygen satu-
data integrity checks by reviewing the articles a second time ration; and 32 used retractions. However, many studies
against the evidence tables. simply stated that infants with signs and symptoms con-

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 158, FEB 2004 WWW.ARCHPEDIATRICS.COM
120

©2004 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Dahlgren Memorial Library-Georgetown University Medical Center User on 08/20/2015
 Table 3. Studies Examining the Accuracy of Virological Tests

Source Gold Standard Tests Compared Results

21
Ahluwalia et al, 1987 Viral culture of NPA and EIA, IFA on NPA and NPS EIA-NPA: Sn = 69%, Sp = 100%; EIA-NPS:
NPS specimens Sn = 61%, Sp = 100%; IFA-NPA: Sn = 61%,
Sp = 89%; and IFA-NPS: Sn = 52%, Sp = 78%
Chattopadhya et al,22 1992 Viral culture IFA, EIA, EIA by blocking test IFA: Sn = 89%, Sp = 92%, EIA: Sn = 94%, Sp = 74%;
and EIA by blocking test: Sn = 94%, Sp = 77%
Eugene-Ruellan et al,23 Viral culture and/or IFA PCR 97% Concordance
1998
Ong et al,22 2001 IFA PCR IFA detected 27 cases; PCR detected 28 cases
Warner et al,25 1990 Viral culture and/or IFA EIA Sn = 86%, Sp = 91%

Abbreviations: EIA, enzyme immunoassays; IFA, direct immunofluorescence assay; NPA, nasopharyngeal aspirate; NPS, nasopharyngeal suction;
PCR, polymerase chain reaction; Sn, sensitivity; Sp, specificity.

sistent with bronchiolitis were eligible for inclusion. Many COMPARISON OF VIROLOGICAL TESTS
authors referred to the classic historic definition of bron-
chiolitis by Court.14 Five studies examined the accuracy of various virologi-
Eligibility criteria in the clinical trials varied, espe- cal tests for RSV and other causative viruses.21-25 Table 3
cially with respect to criteria such as age, duration of symp- demonstrates that numerous tests for RSV exist and that
toms, comorbidities (eg, prematurity and chronic lung their test characteristics vary. The 2000 Red Book from
disease), history of previous wheezing, and severity of the American Academy of Pediatrics reports that the over-
disease. Specific study objectives determined most of these all sensitivity of the rapid antigen detection tests ranges
variations (eg, numerous studies included only infants from 80% to 90%.26 Data presented in Table 3 are con-
with bronchiolitis due to RSV). sistent with this estimate. Individual test manufacturers
Most trials measured disease severity as a baseline in- likely have additional, unpublished data on their own as-
dependent variable and as a dependent outcome (ie, change says, as they generally report test characteristics in the
in disease severity resulting from treatment). Disease se- package insert materials that accompany test kits. Our
verity was most commonly measured using clinical scales search strategy would not have identified this unpub-
(43 of the 65 clinical trials), but the variety of scales used lished data. In addition to looking at test agreement, Ahlu-
made comparing studies difficult. Some studies used clini- walia et al21 compared 2 methods of specimen collec-
cal scales validated in previous studies such as the Respi- tion and demonstrated that viral culture, enzyme
ratory Distress Assessment Instrument.15-18 Other re- immunoassays, and direct immunofluorescence assays
search teams created or modified scales for their particular all yielded positive results more often when performed
trial.19,20 Despite these differences, the clinical scales all in- on nasopharyngeal aspirates than when performed on na-
corporated measures of respiratory rate, respiratory ef- sopharyngeal swabs.
fort, severity of wheezing, and oxygenation. We identified no trials that addressed the question
of whether knowing RSV is the causative agent in bron-
IDENTIFICATION OF THE CAUSE chiolitis affects clinical outcomes.
OF BRONCHIOLITIS
PREDICTORS OF DISEASE SEVERITY
Many but not all of the included studies attempted to iden- OR COMPLICATIONS
tify the cause of enrolled cases of bronchiolitis. Twenty-
nine of the clinical trials enrolled only infants with posi- Four studies (Table 4) measured various predictors of
tive findings for RSV. Of the 56 treatment trials, 46 disease severity.7,27-29 Shaw et al28 examined historical el-
performed RSV testing on all subjects. In the 21 trials in ements, physical examination findings, and laboratory
which all patients underwent testing and were in- results and identified the following 5 clinically impor-
cluded, regardless of RSV status, the cases caused by RSV tant predictors of severe disease: ill or toxic appearance,
ranged from 26% to 95%. In 12 trials, patients under- oxygen saturation of less than 95%, gestational age of less
went testing for other viral causes (eg, parainfluenza vi- than 34 weeks, respiratory rate of greater than 70 breaths
ruses) in addition to RSV, but most reported results as per minute, and age younger than 3 months. Mulhol-
the percentage with positive findings for RSV vs other land et al27 correlated clinical findings with disease se-
viruses. verity defined by pulse oximetry findings and arterial
The techniques for identifying RSV as the causative blood gas measurements. Young age, cyanosis, crackles,
agent of bronchiolitis included viral cultures, rapid anti- and oxygen saturation of less than 90% all predicted more
gen detection tests (eg, direct immunofluorescence assay severe disease. Dawson et al7 studied the relationship be-
and enzyme immunoassays), polymerase chain reaction, tween clinical severity based on clinical scales with the
and measurements of acute and convalescent antibody ti- degree of radiological changes on chest x-rays. The au-
ters. Rapid antigen detection tests for RSV were used most thors found no correlation. Wright et al29 examined the
frequently. In many studies, investigators performed vi- relationship between demographic characteristics, viral
ral cultures on cases with negative findings for RSV. shedding and antibody responses, and disease severity

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 158, FEB 2004 WWW.ARCHPEDIATRICS.COM
121

©2004 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Dahlgren Memorial Library-Georgetown University Medical Center User on 08/20/2015
 Table 4. Studies Measuring Predictors of Disease Severity

Source Outcome Predicted Indicators Examined Predictors

7
Dawson et al, 1990 Clinical score (mild, moderate, CXR findings (ie, hyperinflation, atelectasis, and There was no correlation between CXR
severe, or very severe) infiltrates) findings and disease severity
Mulholland et al,27 Severity at the time of admission as Demographics, cyanosis, crackles, chest wall Indicators of severity at time of
1990 assessed by means of oximetry in drawing, RR ⬎50 breaths/min, Liver admission: young age, cyanosis, and
and arterial blood gas results; O2 ⬎2 cm below costal margin, SaO2 ⬍90%, crackles; predictors of oxygen
requirements during admission PaO2 ⬍60 mm Hg, PaCO2 ⬎45 mm Hg, and requirement during admission: young
RSV status age, cyanosis, crackles, high RR, chest
wall indrawing, SaO2 ⬍90%, PaCO2
⬎45 mm Hg, PaCO2 ⬎45 mm Hg, and
PaO2 ⬍60 mm Hg
Shaw et al,28 1991 Mild disease (defined as alert, Historical information: cyanosis or apnea, The following 6 independent clinical and
active, and able to take fluids gestational age, age ⬍3 mo, decreased oral laboratory findings were strongly
throughout their disease, no O2 intake, perinatal complications, and URI associated with more severe disease
therapy, etc) vs severe disease symptoms ⬍3 d; physical examination and using multiple-factor analysis: ill or
(defined as all others without observations: ill or toxic appearance, Yale toxic appearance, oxygen saturation
mild disease) Observation Scale score ⱖ10, accessory ⬍95%, gestational age ⬍34 wk, RR
muscle use, clinical asthma score ⱖ2, RR, ⱖ70 breaths/min, and age ⬍3 mo
and rales; and laboratory: pulse oximetry
while quiet, pulse oximetry while sucking,
CXR findings of atelectasis or hyperaeration,
and isolation of RSV
Wright et al29 (2002) Illness severity in hospitalized Historical information: age; laboratory: serum History of BPD or congenital heart
infants with RSV bronchiolitis neutralizing antibody titer; RSV shedding disease and younger age
measured by (1) sum of
respiratory illness scores,
(2) duration of O2 therapy
(3) length of ventilatory support

Abbreviations: BPD, bronchopulmonary disease; CXR, chest x-ray film; RR, respiratory rate; RSV, respiratory syncytial virus; SaO2, arterial oxygen saturation;
URI, upper respiratory tract infection.

using data collected during the 2 RSV immunoglobulin infected children with or without bacteria in their secre-
clinical trials.20,29,30 Their report focused primarily on im- tions with data on the corresponding groups without virus
munologic responses to RSV, but demonstrated that infection. The x-ray film findings were normal signifi-
younger age, history of bronchopulmonary dysplasia, and cantly more often in the virus-positive–bacteria-negative
history of congenital heart disease were independently group than in the other groups. Alveolar pneumonia ap-
associated with more severe disease. pearing as lobar or segmental consolidations (“lobar” pneu-
Most textbooks cite young age, history of prematu- monia) was observed with equal frequency and without
rity or other comorbidities, toxic appearance at presenta- relation to bacterial findings in the virus-positive and virus-
tion, and rapid progression of symptoms as risk factors for negative groups. Roosevelt et al39 showed that the pres-
severe disease. The studies by Shaw et al,28 Mulholland et ence of chest x-ray film abnormalities was strongly cor-
al,27 and Wright et al29 support these assertions. related with the use of antibiotics, but did not examine the
effectiveness of antibiotic treatment in these patients. Swin-
UTILITY OF CHEST X-RAYS IN BRONCHIOLITIS gler et al8 examined the impact of chest x-ray films in acute
lower respiratory tract infections on clinical outcomes by
Seventeen studies obtained chest x-ray films on all pa- randomizing 522 infants aged 2 to 59 months to receive
tients (Table 5),7,12,19,20,28,30-41 but many clinical trials do or not to receive a chest radiograph. Children in the chest
not report chest x-ray film results. Two studies examined radiograph group were more likely to be diagnosed as hav-
the relationship between x-ray film abnormalities and dis- ing pneumonia or upper respiratory tract infections and
ease severity. In the trial by Shaw et al,28 the patients with were more likely to be treated with antibiotics; children
atelectasis were 2.7 times more likely (95% confidence in- who did not receive a chest radiograph were more likely
terval [CI], 1.97-3.70) to have severe disease than those to be diagnosed as having bronchiolitis. Despite these dif-
without this x-ray film finding. This association persisted ferences, the median time to recovery was 7 days in both
when it was included in a multivariable analysis. In con- groups.
trast, the data from Dawson et al7 demonstrated no corre-
lation between chest x-ray film findings and baseline dis- UTILITY OF CBC COUNTS IN BRONCHIOLITIS
ease severity as measured by a clinical severity scoring
system. Ten studies obtained CBC counts on all patients
Three studies compared chest x-ray films with cul- (Table 6).19,31,38,41-47 In most of these studies, however,
tures and management. In a prospective cohort of 128 in- the CBC results were not reported or used only to dem-
fants younger than 7 years with clinical lower respiratory onstrate that the treatment and control groups were simi-
tract infections, Friis et al35 obtained viral and bacterial stud- lar at baseline. Saijo et al31 correlated white blood cell
ies on nasopharyngeal secretions; they compared virus- counts in 120 RSV-positive infants with radiologically de-

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 158, FEB 2004 WWW.ARCHPEDIATRICS.COM
122

©2004 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Dahlgren Memorial Library-Georgetown University Medical Center User on 08/20/2015
 Table 5. Utility of Chest X-ray Films in Bronchiolitis

Source Purpose of Study Use of Chest X-ray Results

32
Bertrand et al, RCT of epinephrine vs salbutamol in Baseline assessment CXR results not reported
2001 hospitalized infants
Can et al,19 1998 RCT of salbutamol vs mist in the ED Baseline assessment CXR findings “consistent with bronchiolitis”
were present in 88%, 69%, and 73% of
the infants in the 3 study groups
Dawson et al,7 Cohort design specifically to look at the utility Baseline assessment No correlation between CXR findings and
1990 of routine CXRs in bronchiolitis by disease severity
examining the relationship between clinical
assessment (ie, mild, moderate, severe, or
very severe) and CXR findings (ie,
hyperinflation, atelectasis, and infiltrates)
Dobson et al,34 RCT of albuterol in hospitalized infants Baseline assessment CXR results not reported
1998
Friis et al,35 1990 Prospective cohort of children ⬍7 y designed Baseline assessment The results for children with bronchiolitis
to correlate CXR findings with viral and not reported separately, so no
bacterial studies conclusions can be drawn
Luchetti et al,36 RCT of porcine-derived surfactant in Baseline assessment and to CXR results not reported
1998 ventilated infants document clinical improvement
Meert et al,33 RCT of ribavirin in ventilated children with Baseline assessment CXR results not reported
1994 RSV
Nasr et al,37 RCT of recombinant human Baseline assessment and at study CXR improvement was a trial outcome (CXR
2001 deoxyribonuclease I (rhDNase) in end or time of hospital discharge scores improved in the treatment group
hospitalized infants but not in the control group); CXR
findings were not used to assess disease
severity or determine management
Patel et al,12 RCT of epinephrine vs albuterol vs saline Baseline assessment Pneumonia on CXR: 38% (epinephrine
2002 group), 42% (albuterol group), and 29%
(placebo group)
Rodriguez et RCT of RSVIg treatment of hospitalized young Baseline assessment repeated at CXR results not reported
al,20 1997 children at high risk for severe disease time of hospital discharge
Rodriguez et RCT of RSVIg treatment of previously healthy Baseline assessment repeated at CXR results not reported
al,30 1997 hospitalized children time of hospital discharge
Rodriguez et RCT of ribavirin in infants with RSV disease Baseline assessment CXR results for infants with bronchiolitis not
al,38 1987 (included patients with bronchiolitis, reported separately
pneumonia, and croup)
Roosevelt et al,39 RCT of dexamethasone in acute bronchiolitis Baseline assessment No data presented correlating CXR findings
1996 to disease severity; infiltrates seen in 32%
of treatment group and 20% of placebo
group, and 90% of infants with visible
infiltrates were treated with antibiotics vs
44% of those without these findings
Saijo et al,31 Compare laboratory findings (WBC, neutrophil Baseline assessment to define WBC, ESR, and CRP levels were all higher in
1996 count, ESR, and CRP) to radiographically disease categories (ie, lobar patients with RSV lobar pneumonia vs
defined categories of RSV lower respiratory pneumonia vs bronchopneumonia bronchiolitis or bronchopneumonia
tract infections vs bronchiolitis)
Schuh et al,40 RCT of albuterol in ED Baseline assessment CXR results not reported
1990
Shaw et al,28 Prospective cohort of 228 infants designed to Baseline assessment Overall, 58% had hyperaeration, and 9%
1991 predict mild disease (defined as alert, had atelectasis; findings in patients with
active, and able to take fluids throughout severe vs mild disease: atelectasis in
their disease, no O2 therapy, etc) vs severe 21%; vs 2% (RR, 2.7; 95% CI,
disease (defined as all others without mild 1.97-3.70), hyperaeration in 69% vs 52%
disease) (RR, 1.58; 95% CI, 1.03-2.42)
Taber et al,41 RCT of ribavirin in hospitalized infants Baseline assessment Hyperinflation in 24/26; peribronchial
1983 thickening in 25/26

Abbreviations: CI, confidence interval; CRP, C-reactive protein; CRX, chest x-ray film; ED, emergency department; ESR, erythrocyte sedimentation rate; O2,
oxygen; RCT, randomized controlled trial; RR, relative risk; RSV, respiratory syncytial virus; RSVIg, RSV immunoglobulin; WBC, white blood cell count.

fined categories of lung disease (ie, lobar pneumonia vs studies reporting CBC data demonstrated their utility in
bronchopneumonia vs bronchiolitis). They found that a diagnosing bronchiolitis or guiding therapy.
white blood cell count of greater than 15 000/µL and a
neutrophil count of greater than 10 000/µL were more COMMENT
likely in children with lobar pneumonia or broncho-
pneumonia than in children with bronchiolitis. The 3 dis- Evaluating diagnostic tests for bronchiolitis is problem-
ease categories were defined radiologically. None of the atic because it is a disease that is diagnosed clinically. Thus,

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 158, FEB 2004 WWW.ARCHPEDIATRICS.COM
123

©2004 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Dahlgren Memorial Library-Georgetown University Medical Center User on 08/20/2015
 Table 6. Utility of CBCs in Bronchiolitis

Source Purpose of Study Use of CBC in Study Results

42
Barry et al, 1986 RCT of ribavirin in acute bronchiolitis Baseline assessment, completion of CBC results not reported
study
Can et al,19 1998 RCT of salbutamol sulfate vs mist Baseline assessment Mean WBC, neutrophil, eosinophil, Hb,
and Hct levels similar in 3 study groups
Chipps et al,43 1993 RCT of interferon alfa-2a in hospitalized Baseline assessment, day 5 of study CBC results not reported
infants
De Boeck et al,44 RCT of dexamethasone hospitalized Baseline assessment No difference in leukocyte count and
1997 infants eosinophilia between treatment groups
Friis et al,45 1984 RCT of antibiotics in treatment of Baseline assessment CBC results for bronchiolitis vs
pneumonia and bronchiolitis pneumonia not compared
Kjolhede et al,46 RCT of vitamin A in ALRI Baseline assessment CBC results not reported
1995
Kong et al,47 1993 RCT of Chinese herbs in hospitalized Baseline assessment CBC results not reported
infants
Rodriguez et al,38 RCT of ribavirin in infants with RSV Baseline assessment No differences between treatment groups
1987 disease (included patients with
bronchiolitis, pneumonia, and croup)
Saijo et al,31 1996 Finding of lobar pneumonia vs WBC ⬎15 ⫻ 103/µL; neutrophil The percentages of all 4 indicators were
bronchopneumonia vs bronchiolitis in count ⬎ 10 ⫻ 103/µL; ESR ⬎30 higher in patients with RSV lobar
hospitalized infants with RSV ALRI mm/h; and CRP level ⬎3.0 mg/dL pneumonia vs bronchiolitis or
bronchopneumonia
Taber et al,41 1983 RCT of ribavirin in hospitalized infants Baseline assessment, time of No differences between treatment groups;
discharge, and follow-up no differences from admission to
discharge to follow-up

Abbreviations: ALRI, acute lower respiratory tract infection; CBC, complete blood cell; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate;
Hb, hemoglobin; Hct, hematocrit; RCT, randomized controlled trial; RSV, respiratory syncytial virus; WBC, white blood cell.

there is no gold standard against which to compare test- viral infections, it could be argued that chest x-ray films
ing strategies. Numerous tests for RSV, the leading cause are more likely to lead to inappropriate antibiotic use than
of bronchiolitis, exist, but the clinical utility of RSV test- to improved clinical outcomes.
ing has not been demonstrated for any category of pa- Very few data exist on the utility of CBC counts. Al-
tients. The question of whether RSV testing is necessary though many of the treatment trials collected data on CBC
in patients with bronchiolitis is of interest from the clini- counts, their results were either used to demonstrate that
cal and utilization points of view. Although such testing treatment and control groups were similar or not re-
is commonly used to document the cause of bronchio- ported at all. Complete blood cell counts are commonly
litis, knowing the cause rarely changes clinical manage- used to assist in determining whether a patient has bac-
ment or outcomes. terial disease. The large body of literature on febrile in-
Many institutions require RSV testing of all infants fants demonstrates that the test characteristics for CBC
admitted to the hospital; the rationale is to allow cohort- counts vary greatly on the the basis of the cutoff used and
ing of patients to decrease nosocomial infections. How- that elevated white blood cell counts alone have low speci-
ever, no direct evidence from randomized, controlled trials ficity and positive predictive value. Three studies have
shows that this strategy prevents nosocomial transmis- looked at bacteremia in febrile infants with bronchio-
sion of RSV in children.48 A more logical strategy, fol- litis. Greenes and Harper50 found that the rate of bacter-
lowed by many infection control policies, is to isolate all emia was 1 (0.2%) in 411 for subjects with bronchio-
infants with acute lower respiratory tract infection, re- litis. Purcell and Fergie51 reviewed the medical records
gardless of the cause. of 2396 infants admitted to a single hospital and found
Data from clinical trials demonstrate that large num- that 1.6% had positive findings in cultures of blood, urine,
bers of infants with bronchiolitis have abnormalities on or spinal fluid. Both of these studies were retrospective,
chest x-ray films. However, chest x-ray films do not dis- and CBC count results were not presented. Kupperman
criminate well between bronchiolitis and other forms of et al52 prospectively studied 163 infants with bronchiol-
lower respiratory tract infection. Although textbooks sug- itis and fever and found a 0% rate of bacteremia (95%
gest that chest x-ray films be considered in the manage- CI, 0%-1.9%) and a 1.9% rate of urinary tract infections.
ment of bronchiolitis, specific indications are lacking.49 Our review has several limitations. First, all system-
The data in this review suggest that, in mild disease, chest atic reviews are at risk for publication bias.53 We searched
x-ray films offer no information that is likely to affect treat- the largest and most relevant databases for published stud-
ment and should not be routinely performed. Data from ies but did not seek unpublished data or data main-
the studies by Roosevelt et al39 and Swingler et al8 dem- tained by pharmaceutical companies. Second, no search
onstrate that chest x-ray films may lead to the use of an- strategy is guaranteed to return all relevant studies. Ad-
tibiotics, although this was not the focus of either of these ditional studies may be indexed under terms not used
studies. Given that most children with bronchiolitis or in our search. To decrease the likelihood of missing im-
other forms of acute lower respiratory tract infection have portant studies, we asked a technical advisory group of

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 158, FEB 2004 WWW.ARCHPEDIATRICS.COM
124

©2004 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Dahlgren Memorial Library-Georgetown University Medical Center User on 08/20/2015
 What This Study Adds Quality, Rockville, Md (contract 290-97-0011). The full re-
port is available at: http://www.ahrq.gov/clinic/evrptfiles
.htm#bronch. The authors of this article are responsible for
Despite the large number of clinical trials and prospec- its content, including any clinical or treatment recommen-
tive cohort studies of bronchiolitis, evidence-based in- dations.
dications for RSV and other supportive testing do not ex-
We thank Marian James, PhD, the Agency for Health-
ist. The data available suggest that such testing does not
alter clinical outcome. This systematic review justifies a care Research and Quality task order officer, for her assis-
prospective clinical trial to address these questions that tance with this project. In addition, we thank the following
uses clinically relevant outcomes such as hospitaliza- members of our Technical Expert Advisory Group, who pro-
tion rates, length of hospital stay, time to complete re- vided input and advice for the full evidence report on which
covery, costs of care, and consequences of false-positive this article is based: Henry L. Dorkin, MD, Cystic Fibrosis
and false-negative test findings. Center, Newton, Mass; Bernard Ewigman, MD, MSPH,
School of Medicine, University of Missouri–Columbia; Glenn
Flores, MD, Boston University School of Medicine, Boston,
Mass; Anne Haddix, PhD, Rollins School of Public Health,
content experts to review our final list to identify miss- Emory University, Atlanta, Ga; Allan S. Lieberthal, MD,
ing studies. Third, our exclusion of non-English studies Southern California–Permanente Medical Group, Pan-
may have introduced bias as well, although most sys- orama City, Calif; Jonathan L. Temte, MD, PhD, Depart-
tematic reviews published in the United States use similar ment of Family Medicine, University of Wisconsin, Madi-
exclusions. Finally, we were not able to analyze quanti- son; and Steve Wegner, MD, NC Access, Inc, Morrisville,
tatively the data in this review because of the heteroge- NC. We are indebted to our colleagues at RTI International
neity of the studies. and the University of North Carolina at Chapel Hill for their
Despite the high prevalence of bronchiolitis, little support in the development of this article. At RTI Interna-
consensus exists on optimal management.3,4,54 Diagnos- tional, we thank Amanda Honeycutt, PhD, and John Wit-
tic and supportive testing is common, but data demon- tenborn for their work on the cost-effectiveness analysis of
strating appropriate indications and efficacy of such test- the full report; Loraine Monroe for superior secretarial as-
ing do not exist. Wilson et al 9 demonstrated wide sistance; Nash Herndon, MA, for editing expertise; and Linda
institutional variations in the care of hospitalized in- Lux, MPA, and Philip Salib for technical assistance on the
fants with bronchiolitis that were not explained by dis- project. At the University of North Carolina, we acknowl-
ease severity. These variations correlated significantly with edge Sonya Harris-Hayward, MD, and Mary Maniscalo, MD,
hospital costs and length of stay. for data abstraction; Cheryl Coon, PhD, for methods ab-
Perlstein et al,10 Adcock et al,55 and Kotagal et al56 have straction; and Joy Harris and Donna Curasi for superior re-
all demonstrated that evidence-based guidelines can be used search assistance.
to decrease the frequency of RSV testing, chest x-rays, and Corresponding author: W. Clayton Bordley, MD, MPH,
bronchodilator use in infants hospitalized with bronchi- Division of Emergency Medicine, Department of Surgery,
olitis. These studies demonstrated significant decreases in Duke University Medical Center, DUMC Box 3096, Durham,
length of stay and no changes in readmission rates. These NC 27710 (e-mail: [email protected]).
studies did not purport to test the utility of RSV testing,
chest x-ray films, or CBC counts, but their findings sug- REFERENCES
gest that the routine use of such testing is unnecessary.
Additional prospective trials in emergency departments and
1. Glezen WP, Taber LH, Frank AL, Kasel JA. Risk of primary infection and reinfec-
other outpatient settings will help to validate these find- tion with respiratory syncytial virus. AJDC. 1986;140:543-546.
ings. 2. Orenstein DM. Bronchiolitis. In: Behrman RE, Kliegman R, Jenson HB, eds. Nel-
We recognize that, in some clinical situations, the son Textbook of Pediatrics. 16th ed. Philadelphia, Pa: WB Saunders Co; 2000:
cause of an infant’s illness can significantly affect the need 1285.
3. Meissner HC. Uncertainty in the management of viral lower respiratory tract dis-
for additional workup; examples include infants younger ease. Pediatrics. 2001;108:1000-1003.
than 2 months with fever and signs of lower respiratory 4. Wang EE, Law BJ, Boucher FD, et al. Pediatric Investigators Collaborative Net-
tract disease. Complete blood cell counts and chest x-rays work on Infections in Canada (PICNIC) study of admission and management varia-
can be useful in patients with unusual clinical courses tion in patients hospitalized with respiratory syncytial viral lower respiratory tract
infection. J Pediatr. 1996;129:390-395.
or severe disease. However, in most infants with bron-
5. Mallory MD, Shay DK, Garrett J, Bordley WC. Bronchiolitis management pref-
chiolitis, the limited evidence available does not sup- erences and the influence of pulse oximetry and respiratory rate on the decision
port routine use of RSV testing, chest x-ray films, or CBC to admit. Pediatrics [serial online]. 2003;111:e45-e51. Accessed June 1, 2003.
counts. Given the high prevalence of bronchiolitis, pro- 6. Avner JR, Baker MD. Management of fever in infants and children. Emerg Med
spective trials of diagnostic and supportive testing are Clin North Am. 2002;20:49-67.
7. Dawson KP, Long A, Kennedy J, Mogridge N. The chest radiograph in acute bron-
feasible and needed. Clinicians are understandably re- chiolitis. J Paediatr Child Health. 1990;26:209-211.
luctant to change management practices without high- 8. Swingler GH, Hussey GD, Zwarenstein M. Randomized controlled trial of clinical
quality evidence to guide them. outcome after chest radiograph in ambulatory acute lower-respiratory infection
in children. Lancet. 1998;351:404-408.
Accepted for publication October 2, 2003. 9. Willson DF, Horn SD, Hendley JO, Smout R, Gassaway J. Effect of practice varia-
tion on resource utilization in infants hospitalized for viral lower respiratory ill-
This study was conducted by the RT International/ ness. Pediatrics. 2001;108:851-855.
University of North Carolina Evidence-based Practice Cen- 10. Perlstein PH, Kotagal UR, Bolling C, et al. Evaluation of an evidence-based guide-
ter under contract to the Agency for Healthcare Research and line for bronchiolitis. Pediatrics. 1999;104:1334-1341.

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 158, FEB 2004 WWW.ARCHPEDIATRICS.COM
125

©2004 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Dahlgren Memorial Library-Georgetown University Medical Center User on 08/20/2015
 11. King VJ, Viswanathan M, Bordley WC, et al. Pharmacologic treatment of bron- versus salbutamol in hospitalized infants with bronchiolitis. Pediatr Pulmonol.
chiolitis in infants and children: a systematic review. Arch Pediatr Adolesc Med. 2001;31:284-288.
2004;158:127-137. 33. Meert KL, Sarnaik AP, Gelmini MJ, Lieh-Lai MW. Aerosolized ribavirin in me-
12. Patel H, Platt RW, Pekeles GS, Ducharme FM. A randomized, controlled trial of chanically ventilated children with respiratory syncytial virus lower respiratory
the effectiveness of nebulized therapy with epinephrine compared with albuterol tract disease: a prospective, double-blind, randomized trial. Crit Care Med. 1994;
and saline in infants hospitalized for acute viral bronchiolitis. J Pediatr. 2002; 22:566-572.
141:818-824. 34. Dobson JV, Stephens-Groff SM, McMahon SR, Stemmler MM, Brallier SL, Bay
13. Wainwright C, Altamirano L, Cheney M, et al. A multicenter, randomized, double- C. The use of albuterol in hospitalized infants with bronchiolitis. Pediatrics. 1998;
blind, controlled trial of nebulized epinephrine in infants with acute bronchiol- 101:361-368.
itis. N Engl J Med. 2003;349:27-35. 35. Friis B, Eiken M, Hornsleth A, Jensen A. Chest X-ray appearances in pneumonia
14. Court SD. The definition of acute respiratory illnesses in children. Postgrad Med and bronchiolitis: correlation to virological diagnosis and secretory bacterial find-
J. 1973;49:771-776. ings. Acta Paediatr Scand. 1990;79:219-225.
15. Randolph AG, Wang EE. Ribavirin for respiratory syncytial virus lower respira- 36. Luchetti M, Casiraghi G, Valsecchi R, Galassini E, Marraro G. Porcine-derived
tory tract infection: a systematic overview. Arch Pediatr Adolesc Med. 1996;150: surfactant treatment of severe bronchiolitis. Acta Anaesthesiol Scand. 1998;42:
942-947. 805-810.
16. Klassen TP, Rowe PC, Sutcliffe T, Ropp LJ, McDowell IW, Li MM. Randomized 37. Nasr SZ, Strouse PJ, Soskolne E, et al. Efficacy of recombinant human deoxy-
trial of salbutamol in acute bronchiolitis. J Pediatr. 1991;118:807-811. ribonuclease I in the hospital management of respiratory syncytial virus bron-
17. Menon K, Sutcliffe T, Klassen TP. A randomized trial comparing the efficacy of chiolitis. Chest. 2001;120:203-208.
epinephrine with salbutamol in the treatment of acute bronchiolitis. J Pediatr. 38. Rodriguez WJ, Kim HW, Brandt CD, et al. Aerosolized ribavirin in the treatment
1995;126:1004-1007. of patients with respiratory syncytial virus disease. Pediatr Infect Dis J. 1987;6:
18. Schuh S, Coates AL, Binnie R, et al. Efficacy of oral dexamethasone in outpa- 159-163.
tients with acute bronchiolitis. J Pediatr. 2002;140:27-32. 39. Roosevelt G, Sheehan K, Grupp-Phelan J, Tanz RR, Listernick R. Dexametha-
19. Can D, Inan G, Yendur G, Oral R, Gunay I. Salbutamol or mist in acute bronchi- sone in bronchiolitis: a randomized controlled trial. Lancet. 1996;348:292-295.
olitis. Acta Paediatr Jpn. 1998;40:252-255. 40. Schuh S, Canny G, Reisman JJ, et al. Nebulized albuterol in acute bronchiolitis.
20. Rodriguez WJ, Gruber WC, Groothuis JR, et al. Respiratory syncytial virus im- J Pediatr. 1990;117:633-637.
mune globulin treatment of RSV lower respiratory tract infection in previously 41. Taber LH, Knight V, Gilbert BE, et al. Ribavirin aerosol treatment of bronchiolitis
healthy children. Pediatrics. 1997;100:937-942. associated with respiratory syncytial virus infection in infants. Pediatrics. 1983;
21. Ahluwalia G, Embree J, McNicol P, Law B, Hammond GW. Comparison of na- 72:613-618.
sopharyngeal aspirate and nasopharyngeal swab specimens for respiratory syn- 42. Barry W, Cockburn F, Cornall R, Price JF, Sutherland G, Vardag A. Ribavirin aero-
cytial virus diagnosis by cell culture, indirect immunofluorescence assay, and sol for acute bronchiolitis. Arch Dis Child. 1986;61:593-597.
enzyme-linked immunosorbent assay. J Clin Microbiol. 1987;25:763-767. 43. Chipps BE, Sullivan WF, Portnoy JM. Alpha-2A-interferon for treatment of bron-
22. Chattopadhya D, Chatterjee R, Anand VK, Kumari S, Patwari AK. Lower respira- chiolitis caused by respiratory syncytial virus. Pediatr Infect Dis J. 1993;12:653-
tory tract infection in hospitalized children due to respiratory syncytial (RS) vi- 658.
rus during a suspected epidemic period of RS virus in Delhi. J Trop Pediatr. 1992; 44. De Boeck K, Van der Aa N, Van Lierde S, Corbeel L, Eeckels R. Respiratory syn-
38:68-73. cytial virus bronchiolitis: a double-blind dexamethasone efficacy study. J Pedi-
23. Eugene-Ruellan G, Freymuth F, Bahloul C, Badrane H, Vabret A, Tordo N. Detection atr. 1997;131:919-921.
of respiratory syncytial virus A and B and parainfluenzavirus 3 sequences in respi- 45. Friis B, Andersen P, Brenoe E, et al. Antibiotic treatment of pneumonia and bron-
ratory tracts of infants by a single PCR with primers targeted to the L-polymerase chiolitis: a prospective randomized study. Arch Dis Child. 1984;59:1038-1045.
gene and differential hybridization. J Clin Microbiol. 1998;36:796-801. 46. Kjolhede CL, Chew FJ, Gadomski AM, Marroquin DP. Clinical trial of vitamin A
24. Ong GM, Wyatt DE, O’Neill HJ, McCaughey C, Coyle PV. A comparison of nested as adjuvant treatment for lower respiratory tract infections. J Pediatr. 1995;126:
polymerase chain reaction and immunofluorescence for the diagnosis of respi- 807-812.
ratory infections in children with bronchiolitis, and the implications for a cohort- 47. Kong XT, Fang HT, Jiang GQ, Zhai SZ, O’Connell DL, Brewster DR. Treatment of
ing strategy. J Hosp Infect. 2001;49:122-128. acute bronchiolitis with Chinese herbs. Arch Dis Child. 1993;68:468-471.
25. Warner JL, Whitehurst NJ, Todd SJ, Shalaby H, Wall LV. Comparison of direc- 48. Lozano JM, Wang E. Bronchiolitis. Clin Evid. June 2002;(7):272-282.
tigen RSV with viral isolation and direct immunofluorescence for the identifica- 49. Fleisher GR. Infectious disease emergencies. In: Fleisher GR, Ludwig S, eds. Text-
tion of respiratory syncytial virus. J Clin Microbiol. 1990;28:480-483. book of Pediatric Emergency Medicine. 4th ed. Philadelphia, Pa: Lippincott Wil-
26. American Academy of Pediatrics. Respiratory syncytial virus. In: Pickering LK, liams & Wilkins; 2000:725-794.
ed. 2000 Red Book: Report of the Committee on Infectious Diseases. 25th ed. 50. Greenes DS, Harper MB. Low risk of bacteremia in febrile children with recog-
Elk Grove Village, Ill: American Academy of Pediatrics; 2000:484. nizable viral syndromes. Pediatr Infect Dis J. 1999;18:258-261.
27. Mulholland EK, Olinsky A, Shann FA. Clinical findings and severity of acute bron- 51. Purcell K, Fergie J. Concurrent serious bacterial infections in 2396 infants and
chiolitis. Lancet. 1990;335:1259-1261. children hospitalized with respiratory syncytial virus lower respiratory tract in-
28. Shaw KN, Bell LM, Sherman NH. Outpatient assessment of infants with bron- fections. Arch Pediatr Adolesc Med. 2002;156:322-324.
chiolitis. AJDC. 1991;145:151-155. 52. Kuppermann N, Bank DE, Walton EA, Senac MO Jr, McCaslin I. Risks for bac-
29. Wright PF, Gruber WC, Peters M, et al. Illness severity, viral shedding, and an- teremia and urinary tract infections in young febrile children with bronchiolitis.
tibody responses in infants hospitalized with bronchiolitis caused by respiratory Arch Pediatr Adolesc Med. 1997;151:1207-1214.
syncytial virus. J Infect Dis. 2002;185:1011-1018. 53. Summarizing the evidence. In: Guyatt G, Rennie D. User’s Guides to the Medical
30. Rodriguez WJ, Gruber WC, Welliver RC, et al, Respiratory Syncytial Virus Im- Literature. Chicago, Ill: AMA Press; 2002:529-538.
mune Globulin Study Group. Respiratory syncytial virus (RSV) immune globulin 54. Mallory GBJ, Motoyama EK, Koumbourlis AC, Mutich RL, Nakayama DK. Bron-
intravenous therapy for RSV lower respiratory tract infection in infants and young chial reactivity in infants in acute respiratory failure with viral bronchiolitis. Pe-
children at high risk for severe RSV infections. Pediatrics. 1997;99:454-461. diatr Pulmonol. 1989;6:253-259.
31. Saijo M, Ishii T, Kokubo M, Murono K, Takimoto M, Fujita K. White blood cell 55. Adcock PM, Sanders CL, Marshall GS. Standardizing the care of bronchiolitis.
count, C-reactive protein and erythrocyte sedimentation rate in respiratory syn- Arch Pediatr Adolesc Med. 1998;152:739-744.
cytial virus infection of the lower respiratory tract. Acta Paediatr Jpn. 1996;38: 56. Kotagal UR, Robbins JM, Kini NM, Schoettker PJ, Atherton HD, Kirschbaum MS.
596-600. Impact of a bronchiolitis guideline: a multisite demonstration project. Chest. 2002;
32. Bertrand P, Aranibar H, Castro E, Sanchez I. Efficacy of nebulized epinephrine 121:1789-1797.

(REPRINTED) ARCH PEDIATR ADOLESC MED/ VOL 158, FEB 2004 WWW.ARCHPEDIATRICS.COM
126

©2004 American Medical Association. All rights reserved.

Downloaded From: http://archpedi.jamanetwork.com/ by a Dahlgren Memorial Library-Georgetown University Medical Center User on 08/20/2015