Articles

Clinical features for diagnosis of pneumonia in children

younger than 5 years: a systematic review and meta-analysis
Clotilde Rambaud-Althaus, Fabrice Althaus, Blaise Genton, Valérie D’Acremont

Summary
Background Pneumonia is the biggest cause of deaths in young children in developing countries, but early diagnosis Lancet Infect Dis 2015
and intervention can effectively reduce mortality. We aimed to assess the diagnostic value of clinical signs and Published Online
symptoms to identify radiological pneumonia in children younger than 5 years and to review the accuracy of WHO March 11, 2015
http://dx.doi.org/10.1016/
criteria for diagnosis of clinical pneumonia. S1473-3099(15)70017-4
See Online/Comment
Methods We searched Medline (PubMed), Embase (Ovid), the Cochrane Database of Systematic Reviews, and http://dx.doi.org/10.1016/
reference lists of relevant studies, without date restrictions, to identify articles assessing clinical predictors of S1473-3099(15)70029-0
radiological pneumonia in children. Selection was based on: design (diagnostic accuracy studies), target disease Swiss Tropical and Public
(pneumonia), participants (children aged <5 years), setting (ambulatory or hospital care), index test (clinical features), Health Institute, University of
Basel, Basel, Switzerland
and reference standard (chest radiography). Quality assessment was based on the 2011 Quality Assessment of
(C Rambaud-Althaus MD,
Diagnostic Accuracy Studies (QUADAS-2) criteria. For each index test, we calculated sensitivity and specificity and, F Althaus MD, Prof B Genton MD,
when the tests were assessed in four or more studies, calculated pooled estimates with use of bivariate model and V D’Acremont MD); Department
hierarchical summary receiver operation characteristics plots for meta-analysis. of Ambulatory Care and
Community Medicine,
University of Lausanne,
Findings We included 18 articles in our analysis. WHO-approved signs age-related fast breathing (six studies; pooled Lausanne, Switzerland
sensitivity 0·62, 95% CI 0·26–0·89; specificity 0·59, 0·29–0·84) and lower chest wall indrawing (four studies; 0·48, (C Rambaud-Althaus, F Althaus,
0·16–0·82; 0·72, 0·47–0·89) showed poor diagnostic performance in the meta-analysis. Features with the highest Prof B Genton, V D’Acremont);
and Infectious Disease Service,
pooled positive likelihood ratios were respiratory rate higher than 50 breaths per min (1·90, 1·45–2·48), grunting
Lausanne University Hospital,
(1·78, 1·10–2·88), chest indrawing (1·76, 0·86–3·58), and nasal flaring (1·75, 1·20–2·56). Features with the lowest Lausanne, Switzerland
pooled negative likelihood ratio were cough (0·30, 0·09–0·96), history of fever (0·53, 0·41–0·69), and respiratory rate (Prof B Genton)
higher than 40 breaths per min (0·43, 0·23–0·83). Correspondence to:
Dr C Rambaud-Althaus, Health
Intervention Unit, EPH, Swiss
Interpretation Not one clinical feature was sufficient to diagnose pneumonia definitively. Combination of clinical
Tropical and Public Health
features in a decision tree might improve diagnostic performance, but the addition of new point-of-care tests for Institute, Basel 4051,
diagnosis of bacterial pneumonia would help to attain an acceptable level of accuracy. Switzerland
clotilde.rambaud@unibas.ch
Funding Swiss National Science Foundation.

Introduction published in the late 1980s and validated by other studies in

In developing countries, pneumonia is the largest cause of the 1990s. Since then, no major innovation has been made
deaths in children younger than 5 years.1 Early in pneumonia diagnosis and no accurate point-of-care test
identification and treatment of patients with pneumonia is available to identify children who would benefit from
cases is fundamental to reduce mortality. Identification of antibiotics. With the rapid spread of antibiotic resistance
which pneumonia cases need antibiotic treatment among worldwide, there is rising concern about overprescription
the large number of children presenting with respiratory of antibiotics resulting from insufficient specificity of the
symptoms is a challenge because cough is reported in two WHO criteria used to classify acute respiratory infections.6,7
thirds of children attending outpatient facilities in low- Here, we assess the diagnostic value of clinical signs
income countries.2 Chest radiograph, the current gold and symptoms in identification of children younger than
standard for pneumonia diagnosis,3 is not available in 5 years (excluding infants <2 months) with radiological
resource-poor settings where the burden of disease is the pneumonia. This evaluation might help to generate more
highest. Even when available, chest radiograph cannot be accurate clinical scores from which to make decisions
done for all coughing children because of the very high about the necessity of further investigation by chest
frequency of this complaint and the potential long-term radiograph or antibiotic treatment for children presenting
effects of exposure to x-rays. Therefore, clinical predictors with respiratory symptoms in low-resource ambulatory
are used to identify children who should receive an care facilities.
antibiotic drug or undergo assessment by chest radiograph.
Since the late 1980s, pneumonia diagnosis in developing Methods
countries has relied on the presence of cough, fast Search strategy
breathing, and chest indrawing, as recommended by We did a systematic literature search in Medline
WHO.4,5 This recommendation was based on studies (PubMed), Embase (Ovid), and the Cochrane Database

www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4 1

 Articles

of Systematic Reviews (CDSR), without date or independently did the search using a two-step process:
language restrictions. We did our first search on first, assessing the title and abstract, and second,
Sept 30, 2013, with an update on Nov 6, 2014. In assessing the full text, using the seven selection criteria
Medline, we used the following search terms: listed in the panel. Any disagreement was resolved
“pneumonia”[MeSH terms] in combination with: through discussion and consensus.
“predictive value of tests”[MeSH terms] OR “sensitivity
and specificity”[MeSH terms] OR “reproducibility of Quality assessment
results”[MeSH terms] OR “diagnostic test” OR We assessed the quality of selected studies and potential
“diagnostic tests” OR “physical examination”[MeSH risk of bias with the 2011 revised version of the Quality
terms] OR “medical history taking”[MeSH terms], and Assessment of Diagnostic Accuracy Studies (QUADAS-2),8
See Online for appendix the following age filters: “infant 1–23 months” and adapted to the review question (appendix), as recom-
“preschool child 2–5 years”. In Embase, we used mended by the Cochrane Collaboration. For this Article,
“diagnostic accuracy”/exp OR “predictor variable”/exp all clinical index tests were considered to be appropriate,
OR “breathing rate”/exp in combination with irrespective of the way they had been assessed by the
“pneumonia” OR “lower respiratory tract infection” clinician, except for respiratory rates, which we required
OR “respiratory tract infection”, and with “child”. We to be measured by observation in 1 min in a calm child, as
did an additional manual search of the reference lists recommended by WHO. Therefore, the index test domain
from eligible articles and identified reviews to complete of the QUADAS-2 was separated into two sections: one for
our search. Two reviewers (CR-A and VD’A) the respiratory rate (defining fast breathing) and one for
the other index tests. For the reference standard, the best
existing one for bacterial pneumonia is endpoint
Panel: Criteria for study selection pneumonia on chest radiograph (WHO criteria).3,9 For this
Design review, we selected all studies using chest radiograph as
We selected studies that assessed diagnostic accuracy, clinical predictors or derived reference standard, irrespective of criteria used for
prediction rules. Narrative reviews, letters, editorials, comments, and case series of fewer interpretation. When the interpretation criteria differed
than 20 patients were excluded. Systematic reviews and meta-analyses were considered for from WHO criteria, we reported in the methodological
their reference lists. quality assessment that there were great concerns about
the applicability of the results to the research question.
Target disease The quality assessment was done independently by two
Studies that assessed pneumonia. authors (CR-A and VD’A). Any disagreement was resolved
Participants through discussion and consensus.
Studies needed to include children aged between 2 months and 6 years (ie, although our
reference standard was younger than 5 years, we accepted some studies of age 5 years and Data extraction
younger). We excluded studies including only adults or only children younger than 6 months Data were extracted by one author (CR-A). A second
or older than 6 years. Studies were excluded of patients at higher risk of pneumonia because author (VD’A) cross-checked all extracted data compiled
of pre-existing immune suppression (due to HIV infection, neutropenia, and malnutrition), in a table (Microsoft Excel 2010) comparing them to the
comorbidities (cystic fibrosis, mechanical ventilation, and burn injuries), or because of original data available from the selected full texts (or in
restrictive inclusion criteria (including only patients with wheeze). the subset of data sent by authors when applicable) to
ensure that data were accurate. Identified errors were
Setting discussed and corrected. We recorded characteristics of
We selected studies including either ambulatory patients or patients admitted into the study (design, year of publication, study country, and
hospitals. Studies in intensive care units were excluded. Studies done in developed and health-care setting), study population (size, age range,
developing countries were both considered. inclusion and exclusion criteria, and proportion of
Index test patients with pneumonia), reference standard (chest
We selected studies that assessed clinical features (symptoms and physical signs). radiograph procedure, masking, number of readers, and
interpretation criteria), and index tests (definition,
Reference standard procedures, and link with inclusion criteria) on
We selected studies in which the reference standard for pneumonia diagnosis was based on a predefined forms. Index tests were categorised as related
chest radiograph, whatever the interpretation criteria. to demographic and environmental factors, symptoms,
Data reporting or signs. When it was not possible to establish whether
Studies were selected if reconstruction of the two-by-two table was possible. Studies that the index test was obtained through caretaker interview
included children older than 5 years of age were selected only if age stratified analyses were (symptom) or through clinical examination (sign), the
available (so that children older than 5 years could be excluded). When the data for this age index test was not included in the review and meta-
group were not available, authors were contacted and studies included when data provided. analysis. Combinations of signs were not considered in
When more than one article was published on the same patients’ population, we selected this review. Signs or symptoms assessed at different
only the most recent article with appropriate analyses to the review question. thresholds were considered as different index tests, with
a separate analysis for each threshold.

2 www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4

 Articles

Statistical analysis Role of the funding source

For each individual index test identified in the chosen The funders of the study played no role in study design,
studies, we constructed a two-by-two table with use of the data collection, data analysis, data interpretation, report
data available in the publication or data provided by writing, or in the decision to submit the paper for
contacted authors. For each index test, we calculated publication. All authors had full access to all the data in
sensitivity and specificity with corresponding 95% CIs. the study. The corresponding author had final respon-
When an index test was assessed in four or more different sibility for the decision to submit for publication.
studies, we calculated pooled estimates of sensitivity and
specificity with a hierarchical random effects bivariate Results
logistic regression model (bivariate model). Because of Our search identified 1839 papers. Through the study
expected large differences between studies, heterogeneity selection process (figure 1), 18 articles12–19 were included
in test accuracy between studies was presumed; therefore in the review and underwent quality assessment using
we needed a random effects meta-analysis method that QUADAS-2 (detailed assessment of individual studies is
provided an estimate of the average accuracy of the test available in the appendix). One of the included articles
and described the variability in this effect, rather than a reported on two separate surveys done in different health
fixed effect approach that estimated an underlying facilities in the same country but using the same
common effect. The meta-analysis method also needed to procedures;15 data were extracted separately and counted
account for the correlation between sensitivity and as two distinct studies. Table 1 shows characteristics of
specificity because their relationship as the threshold
varies within and across studies. Therefore the bivariate
model was chosen and computed using the metandi 1790 articles identified in 147 articles identified in
initial search update in 2014
programme in Stata version 12,10 as recommended by the
Cochrane Collaboration for meta-analysis of diagnostic
accuracy studies.11 A minimum of four studies is required
for metandi analyses; we decided not to do meta-analyses 98 duplicates removed
for index tests assessed in fewer than four studies because
it would have little value. 1839 articles screened
We computed summary point estimates of sensitivity
and specificity, as well as the 95% confidence region
1794 excluded in initial screen
around the summary operating point and the 95% of title and abstract
prediction region. The prediction region shows the 1097 wrong design
extent of statistical heterogeneity between studies by 568 wrong target disease
38 wrong participants
depicting a region within which, assuming the model is 90 no clinical index test
correct, there is 95% confidence that the true sensitivity 1 wrong reference
standard
and specificity of a future study will lie.11 We pooled only
index tests with a common and clear definition and a
common threshold. Index tests that were necessary 45 articles identified for further assessment
inclusion criteria for the study (eg, cough in eight
studies) were not considered. For fast breathing that was
35 articles identified 18 excluded in abstract screen
assessed at different thresholds in the selected studies, a from reference lists 8 wrong design
hierarchical summary receiver operation characteristics of retrieved articles 6 wrong target disease
(HSROC) curve was computed using the Rutter and 4 wrong participants
Gatsonis HSROC model, as recommended by the
Cochrane Collaboration for analysis of index tests
62 full-text articles assessed for eligibility
assessed at different thresholds.11

Additional analyses on studies available in the 1990s 44 excluded

9 wrong design
We identified articles that provided the evidence for the 4 wrong participants
WHO definition of clinical pneumonia in the 1990s. 3 no clinical index test
Most of these articles could not be included in our review 11 wrong reference standard
14 insufficient data
because of inappropriate reference standards according 3 duplicate data
to our selection criteria. To better understand the
evidence that was available when the WHO definition
was established, we did a separate review and meta- 18 articles included in qualitative synthesis
analysis on the articles from 1990s, with pooled estimates
of sensitivity and specificity calculated with the bivariate Figure 1: Flow diagram of the study selection process
model for chest indrawing and age-related fast breathing. Only the first reason for exclusion (as ordered in panel 1) is reported.

www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4 3

 4
Setting Age range Patients Pneu- Inclusion criteria Exclusion criteria Reference standard Index tests
(N) monia
pre-
valence
Readers Masking Positivity criteria
Articles

Wafula et al Kenya; paediatric <5 years 377 67% Admission to the Patients with no chest 2 independent Not Lobar pneumonia or Symptoms: cough; signs: stridor, nasal flaring, chest
(1984)12 observation observation ward with radiograph traceable (attending reported bronchopneumonia indrawing, cyanosis, temperature >38°C, respiratory
ward features of ARI excluded from analyses radiologist and rate >50 breaths per min
a senior
radiologist)
Campbell The Gambia; 0–4 years 216 12% Cough, and one of: Not reported 1 paediatric Yes Lobar consolidation Symptoms: vomiting, refusing to feed, rapid
et al (1989)13 community respiratory rate >50 breaths radiologist breathing; signs: chest indrawing, nasal flaring,
per min, indrawing, respiratory rate >50 breaths per min, respiratory rate
wheeze, and stridor >60 breaths per min, heart rate >160 beats per min,
axillary temperature >37·5°C, temperature >38·5°C,
crepitation, bronchial breathing or reduced air entry,
rhonchi, grunting
Wafula et al Kenya; 5–59 150 60% History of cough <2 weeks Already on medication, 1 paediatric Not Lobar pneumonia or Symptoms: fever, rapid breathing, poor feeding; signs:
(1989)14 outpatient months heart failure, congenital radiologist reported bronchopneumonia respiratory rate >40 breaths per min, >60 breaths per
department heart disease, moderate to min, heart rate >140 beats per min, nasal flaring,
severe dehydration, cyanosis, chest indrawing, stridor, ronchi, crepitations,
metabolic disorders, and rectal temperature >37·5°C
chest deformities
Lucero et al Philippines; <5 years 199 69% Cough <3 weeks’ duration Not reported Not reported Not Not reported Signs: respiratory rate >40 breaths per min,
(1990) outpatient reported >50 breaths per min
research department
institute
Lucero et al Philippines; <5 years 199 29% Cough <1 week duration Not reported Not reported Not Not reported Signs: respiratory rate >40 breaths per min,
(1990) outpatient reported >50 breaths per min
health department
centre15
Harari et al Papua New 8 weeks to 185 30% Cough (n=95); cough and Wheeze, stridor, measles, 1 paediatric Yes Radiographic evidence Age <24 months; symptoms: fever, cough >2 days,
(1991)16 Guinea; 6 years respiratory rate ≥40 breaths and pertussis radiologist of pneumonia breath-less, poor feeding, poor sleeping; signs: axillary
outpatient per min (n=90) temperature >38°C, chest indrawing, nasal flaring,
department crepitations, cyanosis, respiratory rate >50/40 breaths
per min*, >50 breaths per min
Lozano et al Colombia; <3 years 200 65% Cough ≤7 days, living at Cardiovascular, 1 radiologist Yes Any kind of infiltrate Symptoms: fever, rapid breathing, difficult breathing,
(1994)17 emergency high altitude, and chest pulmonary, or (alveolar or interstitial) chest retractions, grunting, loss of appetite, food
department radiograph part of neurological congenital refusal (liquid, solid, breastfeeding), difficult to wake
emergency department defects; birth before term; up, abdominal distension, cold to the touch, seizures;
evaluation chronic diseases (asthma, signs: retractions, grunting, nasal flaring, respiratory
cancer, metabolic rate, abnormal respiratory sounds (wheezes,
disorders, crepitation, rhonchi, decreased breath sounds),
immunosuppression) abdominal distension, seizures
Dai et al China; 2–59 541 63% Cough Antibiotics received in 3 radiologists, Yes Criteria not specified; Signs: respiratory rate >50/40 breaths per min*, rales,
(1995)18 outpatient months past 4 weeks independently; 4 categories: nasal flaring, chest indrawing (lower chest wall),
department majority pneumonia, bronchitis, cyanosis of the tongue
opinion any abnormality, normal
prevailed
Palafox et al Mexico; clinical 3 days to 5 110 32% Cases: pneumonia clinical Symptoms >2 weeks, 1 radiologist Yes Presence of Signs: respiratory rate >60 breaths per min,
(2000)19 unit years diagnosis by a paediatrician; chronic diseases, genetic micronodular or >50 breaths per min, >40 breaths per min, chest
matched controls: next abnormalities, macronodular indrawing, alveolar rales
child with ARI (cough or neurological diseases, infiltrations or
rhinorrhoea, and infectious bronchial asthma, condensations in the
signs) without pneumonia septicaemia lung
(Table 1 continues on next page)

www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4

 Setting Age range Patients Pneu- Inclusion criteria Exclusion criteria Reference standard Index tests
(N) monia
pre-
valence
Readers Masking Positivity criteria
(Continued from previous page)
Rothrock USA; emergency <5 years 329 20% Chest radiograph ordered as Urgent portable chest 1 of the Unclear “Pneumonia” or Chief complaint: cough, fever difficulty breathing,
et al (2001)20 department part as emergency radiograph; trauma, attending “infiltrate” on altered mental status; symptoms: respiratory distress;
department evaluation foreign body ingestion, or senior board- radiologist’s report; signs: respiratory distress, rales, diminished breath
submersion injury certified excluding “isolated sounds, respiratory rate >50/40 breaths per min*
radiologists at atelectasis”, “pleural
the time of effusion”, or “elevated
emergency hemidiaphragm”
department
evaluation
Shamo’on Jordan; inpatient <6 years 147 61% Admitted with clinical Immune deficiency, 1 radiologist Yes Lobar pneumonia or Symptoms: cough, poor feeding, signs: chest
et al (2004)21 department pneumonia (cough with known asthma, foreign bronchopneumonia indrawing, grunting, diminished air entry, crepitation,
tachypnea [respiratory rate body aspiration, chemical wheezes, respiratory rate >50/40 breaths per min*;
>50/40 breaths per min*], pneumonitis, failure to not defined: fever
indrawing, or wheezing) thrive or malnutrition and
severe URTI
Mahabee- USA; emergency 2–59 510 8·6% Cough and at least one of: Currently taking 2 paediatric Not No predefined criteria; Age >12 months; breastfed, daycare or preschool,
Gittens et al department months laboured, rapid, or noisy antibiotics, smoke radiologists reported suggestive of sibling, smokers at home; symptoms: illness duration
(2005)22 breathing; chest or inhalation, foreign body independently pneumonia: confluent >48 h; signs: respiratory rate, grunting, nasal flaring,
abdominal pain; or fever aspiration, chest trauma, opacification without retractions, decreased breath sounds, crackles,
asthma, bronchiolitis, volume loss, peripheral wheezing, oximetry
cystic fibrosis, sickle cell rather than central
disease, and chronic opacification, pleural
cardiopulmonary disease effusion
Hazir et al Pakistan; 2–59 1782 14% WHO clinical non-severe Underlying chronic illness, 2 paediatric Yes Radiological evidence of Age >12 months; symptoms: fever, cough, difficult
(2006)23 outpatient months pneumonia (cough and/or history of 3 or more radiologists pneumonia by at least breathing, poor feeding, vomiting, diarrhoea, past
department difficult breathing, and fast episodes of wheeze or independently; 2 of 3 radiologists using history of wheeze, illness duration >3 days; signs:
breathing [50/40 breaths acute bronchial asthma, in case of predefined WHO criteria. wheeze
per min*]), without lower antibiotic use during disagreement
chest wall indrawing, and previous 48 h chest

www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4

without any danger signs radiograph
with readable chest read by a third
radiograph available radiologist
Enwere et al The Gambia; 40 days to 3941 17% History of cough or Serious chronic illness, 1 paediatrician Yes WHO endpoint Symptoms: fever, cough, chest pain, difficult
(2007)24 outpatient 30 breathing difficulty and previous DPT vaccination, and pneumonia, other breathing, fast breathing, poor feeding, vomiting,
department months either suspicion of severe received pneumococcal 1 paediatric infiltrates and diarrhoea, convulsion; signs: appears sick, chest
pneumonia (study’s first vaccination (intervention radiologist abnormalities. indrawing, crepitation, rhonchi, bronchial breathing,
18 months), or raised group), absence of a independently hospital admission
respiratory rate or readable chest radiograph
indrawing (study’s last
27 months) or both
Puumalainen Philippines; 6 weeks to 1195 15% WHO clinical pneumonia Not reported 2 radiologists Yes WHO criteria for primary WHO clinical pneumonia definitions only
et al (2008)25 inpatient 23 (non-severe, severe, and independently endpoint consolidation
department months very severe)
Muangchana Thailand; <5 years 1396 7% Admitted to hospital with Absence of parents or 1 radiologist Yes WHO criteria: presence Age ≤12 months, age ≤3 years; symptoms: illness
et al (2009)26 inpatient suspected pneumonia guardian consent of either primary duration >2 days, illness duration >4 days; signs:
department diagnosed by a physician endpoint pneumonia, or temperature >38°C, temperature >39°C
pleural infiltration
(Table 1 continues on next page)
Articles

5
 6
Articles

Setting Age range Patients Pneu- Inclusion criteria Exclusion criteria Reference standard Index tests
(N) monia
pre-
valence
Readers Masking Positivity criteria
(Continued from previous page)
Sigaúque Mozambique; 0–23 634 43% Admitted to hospital with Evidence of asthma, 2 primary Not Evidence of Age ≤12 months; symptoms: duration of fever,
et al (2009)27 inpatient months cough or difficult breathing congenital heart disease, readers and 1 reported consolidation or pleural duration of cough; signs: nasal flaring, crepitations,
department and fast breathing neonatal asphyxia, and external WHO effusion: confirmed wheezing or rhonchi, vomiting, prostration,
(respiratory rate >50/40 chronic respiratory radiologist pneumonia; non- hypoxaemia, temperature ≥37·5°C, temperature
breaths per min*), and disorders endpoint pneumonia: >39°C
chest indrawing (WHO interstitial or normal
clinical definition for severe chest radiograph
pneumonia)
Bilkis et al Uruguay and 1–4 years† 178† 69% Fever or history of fever Chronic respiratory 2 paediatric Yes Pulmonary Symptoms: fatigue, loss of appetite, loss of sleep,
(2010)28 Argentina; during the past 48 h, and disease, congenital radiologists, consolidation or cough, chest pain, abdominal pain, vomiting; signs:
emergency clinically suspected cardiopathy, together asymmetric infiltrate temperature >39°C, grunting, intercostal retraction,
department pneumonia oesophagogastric reflux, (diagnoses nasal flaring, wheezing, rales, decreased breath sounds
tumoural disease, cerebral correlated with
palsy, immunodeficiency, diagnosis of
asthmatic crisis requiring the evaluating
treatment, pneumonia in paediatricians)
the last 2 months, use of
antibiotics in the last
15 days; chest radiograph
already taken and
interpreted
Wingerter USA; emergency ≤5 years 2008 16% Chest radiograph done for Pre-existing medical 1 of the Not Definite pneumonia if Symptoms: history of fever; signs: temperature ≥38°C,
et al (2012)29 department clinical suspicion of disorders with increased attending reported “consolidation”, wheezing, WHO classification for pneumonia
pneumonia risk for pneumonia: sickle- radiologists at “infiltrate”, or
cell disease, complex the time of “pneumonia” on chest
congenital heart disease, emergency radiograph report;
immunodeficiency, department conservative definition
chronic lung disease other evaluation of pneumonia also
than asthma (ie, cystic included “atelectasis
fibrosis or broncho- versus infiltrate”,
pulmonary dysplasia) or a “atelectasis versus
severe neurological pneumonia”, or “likely
disorder atelectasis but cannot
exclude pneumonia”

All studies were cross-sectional, except for Palafox and colleagues,19 which was a case-control study. ARI=acute respiratory infection. LRTI=lower respiratory tract infection. URTI=upper respiratory tract infection. DPT=diphtheria, pertussis, and tetanus.
*Respiratory rate >50 breaths per min in children aged 2–11 months and >40 breaths per min in children aged 12–59 months. †Subgroup of participants aged below 5 years, provided by contacted author.

Table 1: Characteristics of included studies

www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4

 Articles

the 19 studies. 16 studies were done in low-income and Of the 78 clinical index tests, 18 were assessed in four
middle-income countries—ie, Asia (seven), Africa (five), different studies or more. The most frequently assessed
Latin America (three), and the Middle East (one), and index tests were: fast breathing (12 studies, with four
three took place in the USA. Seven studies were done in different definitions), chest indrawing (10 studies, with
outpatient departments, seven in emergency departments seven different definitions), nasal flaring (eight studies),
or related structures, four in inpatient departments, and elevated temperature (seven studies, with seven
one study recruited patients at community level. Except different definitions), crepitations (seven studies),
for one case-control study,19 all studies included history of fever (six studies), and wheezing (six studies).
consecutive patients with acute respiratory infection. In For each index test considered in four studies or more,
these studies, the inclusion criterion was cough alone the HSROC plot of point estimates of sensitivity and
(five), clinical suspicion of pneumonia as per specified specificity with 95% confidence and 95% prediction
criteria (seven), and suspicion of pneumonia without regions, computed using the bivariate method, are
specified criteria (six; table 1). The proportion of shown in figure 2A (symptoms) and figure 2B (signs).
radiological pneumonia in the studied populations varied Table 3 shows pooled estimates of each diagnostic
across studies from 7% to 69% (median 30%; IQR 15–63). performance measure (sensitivity, specificity, and
Table 2 shows quality assessment of included studies positive and negative likelihood ratio) for these index
(potential bias and applicability concerns). tests. Although most of the index tests showed poor
From the 19 studies, we found 57 different clinical diagnostic performance with a high degree of
features assessed for their accuracy in diagnosis of heterogeneity, some respiratory danger signs had
radiological pneumonia: five related to demographic and specificities higher than 0·80, such as cyanosis (two
environmental factors, 32 to symptoms, and 20 to signs. studies, 0·98 [95% CI 0·93–1·00] and 0·94 [0·89–0·96]),
Age, duration of illness, duration of cough, heart rate, stridor (two studies, 0·92 [0·86–0·96] and 1·00
respiratory rate, and temperature were assessed using [0·94–1·00]), and grunting (five studies, 0·87
different thresholds. Seven different definitions of chest [0·65–0·96]). Classic auscultation signs, such as
indrawing were used. 78 index tests were assessed in the crepitations, showed poor accuracy (table 3). Among
19 studies. The number of index tests assessed per study other auscultation signs, bronchial breathing had high
ranged from one to 25 (median eight, IQR four to 12). The specificity in two studies 0·97 [0·93–0·99] and 0·97
appendix contains coupled forest plots for each study and [0·95–0·98]). Two symptoms had high sensitivity with
the estimated sensitivity and specificity with 95% CI for little heterogeneity: history of fever (pooled sensitivity of
each of the 78 index tests. six studies, 0·94, 0·88–0·97) and cough (five studies,

Risk of bias Applicability concerns

Patient All index tests Index test for Reference Flow and Patient All index tests Index test for Reference
selection but respiratory respiratory Standard timing selection but respiratory respiratory standard
rate rate rate rate
Wafula et al (1984)12 Unclear Unclear Unclear Low High Low Unclear Unclear High
Campbell et al (1989)13 Unclear Low Unclear Low High High Low Unclear Low
Wafula et al (1989)14 Unclear Low Low Unclear Unclear Low Low Low High
Lucero et al (1990)15 Unclear NA Low Unclear Unclear Low NA Unclear Unclear
Harari et al (1991)16 Unclear Low Low Unclear Unclear Low Low Unclear Unclear
Lozano et al (1994)17 Unclear Low High Low High Unclear Low Low High
Dai et al (1995)18 Low Low Low Low Unclear Low Low Low Unclear
Palafox et al (2000)19 High Low Low Low Unclear High Low Low High
Rothrock et al (2001)20 Unclear Low Unclear High High Unclear Low Unclear High
Shamo’on et al (2004)21 High Low NA Low Low High Low NA High
Mahabee-Gittens et al (2005)22 High Low High High High High Low High Unclear
Hazir et al (2006)23 High Low NA Low High High Low NA Low
Enwere et al (2007)24 High Low NA Low High High Low NA Low
Puumalainen et al (2008)25 High Low NA Low High High Low NA Low
Muangchana et al (2009)26 Unclear Low NA Low Low Unclear Low NA Low
Sigauque et al (2009)27 High Low NA Low High High Low NA Low
Bilkis et al (2010)28 Unclear Low High High High Unclear Low Unclear High
Wingerter et al (2012)29 Unclear Low Unclear High Low Unclear Low Unclear High

NA=not applicable.

Table 2: Quality assessment according to QUADAS-2: level of risk

www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4 7

 Articles

A
History of fever Cough Difficult breathing Rapid breathing Poor feeding Vomiting
1·0

0·8
Sensitivity

0·6

0·4

0·2

0
1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0

B
Nasal flaring Grunting Indrawing Respiratory rate Age-related fast breathing Respiratory rate
>40 breaths per min >50 breaths per min
1·0

0·8
Sensitivity

0·6

0·4

0·2

Decreased breath sounds Crepitations Wheezing Rales Rhonchi Temperature >38°C

1·0

0·8
Sensitivity

0·6

0·4

0·2

0
1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0 1·0 0·8 0·6 0·4 0·2 0
Specificity Specificity Specificity Specificity Specificity Specificity

Study estimate 95% confidence region

Summary point 95% prediction region

Figure 2: Hierarchical summary receiver operating characteristic (HSROC) plots for sensitivity versus specificity for six symptoms (A) and for 12 signs (B)
Each circle represents a study, with the size being proportional to the study size. The square represents the summary operating point of test performance and the zone outlines surrounding it represent
the 95% confidence and 95% prediction regions of this summary estimate respectively.

0·96, 0·91–0·98). Two features that are the backbone of geneous), respiratory rate higher than 40 breaths per
the WHO definition for clinical pneumonia4 (age-related min, and history of fever (table 3).
fast breathing and lower chest wall indrawing) had a Table 4 shows the performance of age-related fast
very high degree of heterogeneity in sensitivities and breathing and chest indrawing in the studies that
specificities (table 3). Fast breathing was assessed at generated the evidence leading to the adoption of WHO
different respiratory rate thresholds within and between clinical diagnosis of pneumonia in the 1990s.30 Sensitivity
studies. Figure 3 shows the HSROC curve of fast estimates for age-related fast breathing ranged
breathing at different thresholds. It suggests that none from 0·73 to 0·82 and specificities from 0·54 to 0·89 in
of the fixed thresholds is better than another and that five studies (table 4). For chest indrawing, sensitivity
having an age-related threshold does not improve the estimates ranged from 0·06 to 0·77 and specificity
accuracy of the diagnostic test. estimates from 0·39 to 1·00 (table 4).
In the meta-analysis, when looking at the likelihood
ratios (measures that are more meaningful for clinical Discussion
decisions), the symptoms and signs with the highest To our knowledge, our Article is the first systematic
pooled estimates of positive likelihood ratio were review with meta-analysis of clinical predictors of pneu-
respiratory rate higher than 50 breaths per min, grunting, monia in children. The comprehensive search, un-
lower chest indrawing, and nasal flaring (table 3). The impeded by date, country, or language restrictions,
features with the lowest pooled estimates of negative allowed the consideration of large amounts of data,
likelihood ratio were cough (although very hetero- compared with previous reviews.7,30,34,35 We considered

8 www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4

 Articles

Number of Total Sensitivity (95% CI) Specificity (95% CI) Positive likelihood Negative likelihood 1/negative likelihood
studies population (n) ratio (95% CI) ratio (95% CI) ratio (95% CI)
Symptoms
History of fever 6 8260 0·94 (0·88–0·97) 0·12 (0·06–0·23) 1·06 (1·00–1·12) 0·53 (0·41–0·69) 1·89 (1·46–2·45)
Cough 5 6421 0·96 (0·91–0·98) 0·14 (0·03–0·46) 1·12 (0·90–1·39) 0·30 (0·09–0·96) 3·37 (1·04–10·89)
Difficult breathing 4 6070 0·60 (0·35–0·81) 0·52 (0·19–0·84) 1·26 (0·84–1·91) 0·76 (0·64–0·90) 1·32 (1·11–1·56)
Rapid breathing 4 4474 0·79 (0·75–0·82) 0·31 (0·17–0·49) 1·14 (0·88–1·46) 0·69 (0·39–1·25) 1·44 (0·80–2·60)
Poor feeding 7 4984 0·64 (0·39–0·83) 0·52 (0·3–0·73) 1·34 (1·17–1·54) 0·69 (0·55–0·86) 1·46 (1·16–1·83)
Vomiting 5 6723 0·36 (0·22–0·52) 0·7 (0·55–0·82) 1·17 (1·06–1·29) 0·93 (0·86–0·99) 1·08 (1·01–1·16)
Signs
Nasal flaring 8 2813 0·47 (0·28–0·66) 0·73 (0·52–0·87) 1·75 (1·20–2·56) 0·73 (0·59–0·89) 1·38 (1·12–1·69)
Grunting 5 1251 0·24 (0·10–0·47) 0·87 (0·65–0·96) 1·78 (1·10–2·88) 0·88 (0·78–0·99) 1·13 (1·01–1·28)
Temperature >38°C* 5 4631 0·56 (0·39–0·71) 0·55 (0·40–0·70) 1·25 (1·14–1·37) 0·80 (0·70–0·91) 1·26 (1·10–1·43)
Respiratory rate >40 breaths per min 4 1058 0·78 (0·54–0·91) 0·51 (0·38–0·63) 1·58 (1·37–1·84) 0·43 (0·23–0·83) 2·30 (1·20–4·41)
Respiratory rate >50 breaths per min 7 1834 0·53 (0·30–0·74) 0·72 (0·58–0·83) 1·90 (1·45–2·48) 0·65 (0·45–0·95) 1·53 (1·05–2·24)
Age-related fast breathing†‡ 6 3320 0·62 (0·26–0·89) 0·59 (0·29–0·84) 1·55 (0·44–5·42) 0·63 (0·16–2·55) 1·59 (0·39–6·42)
Crepitations 7 2510 0·53 (0·37–0·69) 0·58 (0·48–0·67) 1·26 (0·99–1·60) 0·81 (0·61–1·08) 1·23 (0·93–1·63)
Rales 4 1158 0·49 (0·32–0·67) 0·45 (0·22–0·70) 0·90 (0·42–1·90) 1·13 (0·48–2·62) 0·89 (0·38–2·06)
Rhonchi 4 1543 0·19 (0·04–0·57) 0·67 (0·24–0·93) 0·57 (0·36–0·91) 1·21 (0·88–1·67) 0·83 (0·60–1·14)
Decreased breath sounds 5 1364 0·22 (0·12–0·38) 0·76 (0·29–0·96) 0·93 (0·15–5·67) 1·02 (0·58–1·80) 0·98 (0·55–1·72)
Wheezing 6 4825 0·22 (0·18–0·25) 0·75 (0·66–0·82) 0·86 (0·63–1·17) 1·05 (0·95–1·16) 0·95 (0·86–1·06)
Lower chest indrawing‡ 4 1870 0·48 (0·16–0·82) 0·72 (0·47–0·89) 1·76 (0·86–3·58) 0·71 (0·38–1·35) 1·40 (0·74–2·65)

*Rectal temperature >38·0°C or axillary temperature >37·5°C. †Respiratory rate >60 breaths per min in children aged <2 months, >50 breaths per min in children 2–11 months, and >40 breaths per min in
children aged 12–59 months. ‡WHO criteria for pneumonia.

Table 3: Pooled estimates of diagnostic performance measures of each index test assessed in four studies or more

only data from children aged younger than 5 years, 1·0

allowing better targeting of the population of interest.
Methodological quality was assessed in duplicate and
based on a-priori defined rules using the latest version of
0·8
the QUADAS, which reduced subjectivity in the selection
of studies and index tests and allowed precise evaluation
of the risk of bias in several domains. Another strength
0·6
of our article is that the method used for analysis
Sensitivity

accounted for heterogeneity in results and for correlation

between sensitivity and specificity.
From the 18 selected articles, a large set of clinical indices 0·4

were assessed, showing both the large panel of clinical

signs and symptoms that can be considered in children
and the poor consensus for the clinical sign with the 0·2
highest accuracy for diagnosis of pneumonia. Fast
breathing and chest indrawing were the most frequently
assessed clinical signs. These two signs are the cornerstone 0
1·0 0·8 0·6 0·4 0·2 0
of acute respiratory infection classification in the WHO
Specificity
pneumonia case management strategy, with antibiotic
Study estimate Age-related fast breathing
prescription being recommended for children with cough HSROC curve Respiratory rate >40 breaths per min
and fast breathing or lower chest wall indrawing (classified Respiratory rate >50 breaths per min
Respiratory rate >60 breaths per min
as non-severe pneumonia).5
Age-related fast breathing was adopted in the 1990s with Figure 3: Hierarchical summary receiver operating characteristics (HSROC)
sensitivity estimates ranging from 0·73 to 0·82, although curve of sensitivity versus specificity of fast breathing assessed at different
heterogeneous specificities were reported (from 0·54 to respiratory rate thresholds in 12 different studies
0·89; table 4). In our Article, age-related fast breathing
assessed in six different studies had poorer estimates of difference in the diagnostic accuracy of age-related fast
diagnostic performance than in previous studies, with breathing is possibly due to the difference in reference
highly heterogeneous sensitivities and specificities. This standard: in our Article, chest radiograph was used as a

www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4 9

 Articles

Age range Reference standard True False False True Sensitivity Specificity
positive negative positive negative (95% CI) (95% CI)
Age-related fast breathing*
Shann et al (1984)31 <5 years Crepitations 52 15 36 97 0·78 (0·66–0·87) 0·73 (0·65–0·80)
Cherian et al (1988)32 <5 years Crepitations or chest 204 46 47 385 0·82 (0·76–0·86) 0·89 (0·86–0·92)
radiograph
Harari et al (1991)16 8 weeks to 6 years Chest radiograph 41 15 47 82 0·73 (0·60–0·84) 0·64 (0·55–0·72)
Mulholland et al (1992) Philippines33 2–59 months Paediatrician 81 21 95 111 0·79 (0·70–0·87) 0·54 (0·47–0·61)
Mulholland et al (1992) Swaziland33 2–59 months Paediatrician 20 6 64 201 0·77 (0·56–0·91) 0·76 (0·70–0·81)
Lower chest wall indrawing
Shann et al (1984)31 <5 years Crepitations 4 63 0 133 0·06 (0·02–0·15) 1·00 (0·97–1·00)
Cherian et al (1988)32 <5 years Crepitations or chest 193 57 11 421 0·77 (0·71–0·82) 0·97 (0·95–0·99)
radiograph
Campbell et al (1989)13 0–4 years Chest radiograph 15 10 117 113 0·60 (0·39–0·79) 0·39 (0·32–0·46)
Harari et al (1991)16 8 weeks to 6 years Chest radiograph 18 38 16 113 0·32 (0·20–0·46) 0·88 (0·81–0·93)

*Respiratory rate >60 breaths per min in children aged <2 months, >50 breaths per min in children 2–11 months, and >40 breaths per min in children aged 12–59 months.

Table 4: Performance of age-related fast breathing and chest indrawing in studies that WHO used to decide on criteria for clinical pneumonia

reference standard, whereas in most of the early studies In our Article, chest indrawing, as in the studies done
(table 4), the reference standard was based on the subjective in the 1990s, also produced heterogeneous estimates of
assessment of a physician (this is also why these studies sensitivities, specificities, and likelihood ratios. Chest
were not selected for analysis here). Thus, these early indrawing is probably an early indicator of respiratory
studies probably included lower respiratory tract infections distress that can be due to different disorders, such as
other than pneumonia, such as bronchiolitis. pneumonia, but also bronchiolitis. Even if chest
At the time that these studies were done, giving indrawing is insufficient for diagnosis of radiological
antibiotics to all children with lower respiratory tract pneumonia, then it might still be useful to identify
infections and withholding them only for upper respiratory children that are at risk of hypoxaemia and would benefit
tract infections was perceived to be the best option. Since from oxygen therapy rather than provision of antibiotics.
pneumonia was a major cause of mortality in resource- Our Article has some limitations. First, to assess clinical
poor settings and no simple test was available for diagnosis predictors for the diagnosis of pneumonia in ambulatory
of bacterial pneumonia, WHO decided to use highly care, a study should ideally include all patients presenting
sensitive clinical criteria. The benefits of presumed lives to the health facility without pre-selection criteria. In all
saved through antibiotic treatment were estimated to our included studies only a subgroup of patients at higher
outweigh the risks of unnecessary treatment due to poor risk of pneumonia were chosen based on a constellation
specificity of the diagnostic criteria used. By contrast, in of symptoms and signs, and difficult-to-diagnose cases
our Article, we aimed to assess the clinical predictors for were potentially excluded. This inclusion might have
radiological pneumonia, considered an acceptable biased the diagnostic performance measures. Second, the
surrogate for bacterial infection, to identify children that interobserver agreement among clinicians on symptoms
really need antibiotic treatment. As a result, the pooled and signs, such as auscultation findings, can be very low.
estimates of likelihood ratios for age-related fast breathing This concern about reproducibility is common to all
are worse in our findings (positive likelihood ratio 1·55; diagnostic studies that assess clinical features. Finally,
negative likelihood ratio 0·63; table 3) than in the 1990s’ there was heterogeneity between our chosen studies in
surveys (pooled estimates from the data reported in table 4: terms of inclusion criteria, setting, and chest radiograph
positive likelihood ratio 2·92; negative likelihood ratio interpretation criteria. To do a meta-analysis in this
0·29), and therefore fast breathing might not be useful context, we used the bivariate and the Rutter and Gatsonis
clinically, at least on its own, to identify children in need of HSROC models, which account for the heterogeneity
antibiotics. This was also suggested in a recent study in inherent in diagnostic accuracy studies.11 The small
Pakistan, in which investigators reported that the clinical number of identified studies did not allow investigation of
outcome of children with WHO non-severe pneumonia how the tests accuracies varied between studies with their
(and no chest indrawing) did not differ when treated with methodological characteristics.
antibiotics or placebo.6 Because of the rapid spread of The findings of this Article suggest that no one clinical
antibiotic resistance worldwide, the overuse of antibiotics feature is sufficient on its own for diagnosis of
when prescription is based on cough and fast breathing is radiological pneumonia. Indeed, none of the assessed
a matter of concern and should now be addressed in policy clinical features reached the level commonly accepted for
recommendations. clinical significance (positive likelihood ratio >5 to

10 www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4

 Articles

include the diagnosis or negative likelihood ratio <0·2 to Tanzania, funded by a grant from the Swiss National Science Foundation
exclude it). The highest pooled positive likelihood ratio (grant number IZ70Z0–124023). We thank Manuel D Bilkis (Hospital de
Niños Ricardo Gutiérrez, Buenos Aires, Argentina) and Homero
observed was 1·9 (respiratory rate >50 breaths per min) Martinez (Hospital Infantil de México Dr Federico Gómez, Mexico City,
and, besides cough, the lowest pooled negative likelihood Mexico) for providing supplementary data and information or answering
ratio was 0·43 (respiratory rate >40 breaths per min). The our queries about their studies. We thank Isabella Locatelli (Institute for
relatively good pooled negative likelihood ratio (0·30) for Social and Preventive Medicine, University of Lausanne, Switzerland)
for her statistical advice on meta-analysis method. We thank Kristina
cough was probably overestimated because cough was Keitel (Swiss Tropical and Public Health Institute, Basel, Switzerland)
part of the inclusion criteria in all selected studies that for her comments on the manuscript and Amena Briet (Swiss Tropical
assessed it as an index test. Respiratory rate, which is the and Public Health Institute, Basel, Switzerland) for her careful editing of
cornerstone of the present WHO criteria to classify the text.
pneumonia, is thus of poor diagnostic value, even if it References
1 Liu L, Johnson HL, Cousens S, et al. Global, regional, and national
was the best individual clinical predictor in our Article. causes of child mortality: an updated systematic analysis for 2010
However, according to our results, a threshold of with time trends since 2000. Lancet 2012; 379: 2151–61.
50 breaths per min for all age groups would be more 2 Van Hemelrijck MJ, Lindblade KA, Kubaje A, et al. Trends observed
appropriate than would the present recommendation of during a decade of paediatric sick visits to peripheral health
facilities in rural western Kenya, 1997–2006. Trop Med Int Health
50 breaths per min in infants (<12 months) and 40 breaths 2009; 14: 62–69.
per min in children older than 1 year. History of fever 3 Cherian T, Mulholland EK, Carlin JB, et al. Standardized
was the second best predictor in terms of negative interpretation of paediatric chest radiographs for the diagnosis of
pneumonia in epidemiological studies. Bull World Health Organ
likelihood ratio, and adding it to the present WHO 2005; 83: 353–59.
criteria could help to increase specificity. 4 Programme for the Control of Acute Respiratory Infections, WHO.
Combination of the clinical features with the best Technical bases for the WHO recommendations on the
management of pneumonia in children at first-level health
likelihood ratios in a decision tree might indeed improve facilities. Geneva, 1991. http://whqlibdoc.who.int/hq/1991/WHO_
the overall diagnostic performance of symptoms and ARI_91.20.pdf (accessed Oct 3, 2013).
signs. Individual data from selected articles would, 5 World Health Organization, UNICEF. Integrated Management of
Childhood Illness. Chart booklet. 2014. http://apps.who.int/iris/
however, have been necessary to appropriately assess bitstream/10665/104772/16/9789241506823_Chartbook_eng.
combination of clinical features. This was out of the pdf?ua=1 (accessed March 25, 2014).
scope of our Article, but it would be relevant to embark 6 Hazir T, Nisar YB, Abbasi S, et al. Comparison of oral amoxicillin
with placebo for the treatment of world health organization-defined
on such meta-analyses prospectively, provided the nonsevere pneumonia in children aged 2–59 months: a multicenter,
heterogeneity in methodological quality of the included double-blind, randomized, placebo-controlled trial in Pakistan.
studies is mitigated. Clin Infect Dis 2011; 52: 293–300.
7 Scott JAG, Wonodi C, Moisi JC, et al. The Definition of Pneumonia,
The development of innovative point-of-care tests for the Assessment of Severity, and Clinical Standardization in the
diagnosis of bacterial pneumonia is crucial to assist Pneumonia Etiology Research for Child Health Study.
clinicians in decision making. The assessment of the Clin Infect Dis 2012; 54: S109–16.
8 Whiting PF, Rutjes AW, Westwood ME, et al. QUADAS-2: a revised
diagnostic value of combinations of clinical features and tool for the quality assessment of diagnostic accuracy studies.
point-of-care tests to measure host biomarkers or specific Ann Intern Med 2011; 155: 529–36.
pathogens needs appropriately designed prospective 9 Lynch T, Bialy L, Kellner JD, et al. A systematic review on the
studies to propose new evidence-based diagnostic diagnosis of pediatric bacterial pneumonia: when gold is bronze.
PLoS One 2010; 5: e11989.
procedures. Because no gold standard test exists for 10 Harbord RM, Whiting P. metandi: meta-analysis of diagnostic
bacterial pneumonia, investigation is important of the accuracy using hierarchical logistic regression. Stata J 2009; 9: 211.
clinical outcome of children with acute respiratory 11 Macaskill P, Gatsonis C, Deeks J, Harbord R, Takwoingi Y.
Chapter 10: Analysing and Presenting Results. In: Deeks JJ,
infection when treated (or not) with antibiotics33 to Bossuyt PM, Gatsonis C (eds). Cochrane handbook for systematic
accurately identify patients truly in need of antibiotics. reviews of diagnostic test accuracy version 1.0, The Cochrane
Potentially severe forms of acute respiratory infections Collaboration. 2010. http://srdta.cochrane.org/ (accessed Jan 6,
2014).
also include viral pneumonia and bronchiolitis. 12 Wafula EM, Muruka FJ. Chest x-rays in children with acute
Therefore improvements in access to oxygen therapy, respiratory infections or bronchospasm at Kenyatta National
rather than provision of antibiotics, will often be the life- Hospital. East Afr Med J 1984; 61: 900–06.
13 Campbell H, Byass P, Lamont AC, et al. Assessment of clinical
saving treatment. criteria for identification of severe acute lower respiratory tract
Contributors infections in children. Lancet 1989; 1: 297–99.
CR-A, FA, and VD’A conceived the study. CR-A and VD’A did the 14 Wafula EM, Tindyebwa DB, Onyango FE. The diagnostic value of
literature search and analyses. BG and VD’A oversaw the study and various features for acute lower respiratory infection among under
provided important scientific input. All authors contributed to the fives. East Afr Med J 1989; 66: 678–84.
interpretation of the data and the drafting of the Article. All authors 15 Lucero MG, Tupasi TE, Gomez ML, et al. Respiratory rate greater
revised the Article critically and approved the version to be published. than 50 per minute as a clinical indicator of pneumonia in Filipino
children with cough. Rev Infect Dis 1990; 12: S1081–83.
Declaration of interests 16 Harari M, Shann F, Spooner V, Meisner S, Carney M, de Campo J.
We declare no competing interests. Clinical signs of pneumonia in children. Lancet 1991; 338: 928–30.
Acknowledgments 17 Lozano JM, Steinhoff M, Ruiz JG, Mesa ML, Martinez N, Dussan B.
The study was part of a larger project (PeDiAtrick) that aimed to improve Clinical predictors of acute radiological pneumonia and hypoxaemia
the quality of health care and rational use of drugs for children in at high altitude. Arch Dis Child 1994; 71: 323–27.

www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4 11

 Articles

18 Dai Y, Foy HM, Zhu Z, Chen B, Tong F. Respiratory rate and signs 27 Sigaúque B, Roca A, Bassat Q, et al. Severe pneumonia in
in roentgenographically confirmed pneumonia among children in Mozambican young children: clinical and radiological
China. Pediatr Infect Dis J 1995; 14: 48–50. characteristics and risk factors. J Trop Pediatr 2009; 55: 379–87.
19 Palafox M, Guiscafré H, Reyes H, Munoz O, Martínez H. 28 Bilkis MD, Gorgal N, Carbone M, et al. Validation and development
Diagnostic value of tachypnoea in pneumonia defined of a clinical prediction rule in clinically suspected community-
radiologically. Arch Dis Child 2000; 82: 41–45. acquired pneumonia. Pediatr Emerg Care 2010; 26: 399–405.
20 Rothrock SG, Green SM, Fanelli JM, Cruzen E, Costanzo KA, 29 Wingerter SL, Bachur RG, Monuteaux MC, Neuman MI.
Pagane J. Do published guidelines predict pneumonia in children Application of the world health organization criteria to predict
presenting to an urban ED? Pediatr Emerg Care 2001; 17: 240–43. radiographic pneumonia in a US-based pediatric emergency
21 Shamo’on H, Hawamdah A, Haddadin R, Jmeian S. Detection of department. Pediatr Infect Dis J 2012; 31: 561–64.
pneumonia among children under six years by clinical evaluation. 30 Simoes EAF. Recognizing and diagnosing pneumonia in
East Mediterr Health J 2004; 10: 482–87. developing countries. Curr Opin Infect Dis 1994; 7: 358–63.
22 Mahabee-Gittens EM, Grupp-Phelan J, Brody AS, et al. Identifying 31 Shann F, Hart K, Thomas D. Acute lower respiratory tract infections
children with pneumonia in the emergency department. in children: possible criteria for selection of patients for antibiotic
Clin Pediatr (Phila) 2005; 44: 427–35. therapy and hospital admission. Bull World Health Organ 1984;
23 Hazir T, Nisar YB, Qazi SA, et al. Chest radiography in children 62: 749–53.
aged 2–59 months diagnosed with non-severe pneumonia as 32 Cherian T, John TJ, Simoes E, Steinhoff MC, John M. Evaluation of
defined by World Health Organization: descriptive multicentre simple clinical signs for the diagnosis of acute lower respiratory
study in Pakistan. BMJ 2006; 333: 629. tract infection. Lancet 1988; 2: 125–28.
24 Enwere G, Cheung YB, Zaman SMA, et al. Epidemiology and 33 Mulholland EK, Simoes EA, Costales MO, McGrath EJ,
clinical features of pneumonia according to radiographic findings Manalac EM, Gove S. Standardized diagnosis of pneumonia in
in Gambian children. Trop Med Int Health 2007; 12: 1377–85. developing countries. Pediatr Infect Dis J 1992; 11: 77–81.
25 Puumalainen T, Quiambao B, Abucejo-Ladesma E, et al. Clinical 34 Ayieko P, English M. Case management of childhood pneumonia
case review: a method to improve identification of true clinical and in developing countries. Pediatr Infect Dis J 2007; 26: 432–40.
radiographic pneumonia in children meeting the World Health 35 Pio A. Standard case management of pneumonia in children in
Organization definition for pneumonia. BMC Infect Dis 2008; 8: 95. developing countries: the cornerstone of the acute respiratory
26 Muangchana C. Factors associated with diagnosis of bacterial infection programme. Bull World Health Organ 2003; 81: 298–300.
pneumonia in children of Northern Thailand.
Southeast Asian J Trop Med Public Health 2009; 40: 563–69.

12 www.thelancet.com/infection Published online March 11, 2015 http://dx.doi.org/10.1016/S1473-3099(15)70017-4