Pediatrics
Pediatrics
Pediatrics
The online version of this article, along with updated information and services, is
located on the World Wide Web at:
http://pedsinreview.aappublications.org/content/34/10/438
Pediatrics in Review is the official journal of the American Academy of Pediatrics. A monthly
publication, it has been published continuously since 1979. Pediatrics in Review is owned,
published, and trademarked by the American Academy of Pediatrics, 141 Northwest Point
Boulevard, Elk Grove Village, Illinois, 60007. Copyright © 2013 by the American Academy of
Pediatrics. All rights reserved. Print ISSN: 0191-9601.
Updated Information & including high resolution figures, can be found at:
Services http://pedsinreview.aappublications.org/content/34/10/438
References This article cites 29 articles, 8 of which you can access for free at:
http://pedsinreview.aappublications.org/content/34/10/438#BIBL
Permissions & Licensing Information about reproducing this article in parts (figures, tables) or
in its entirety can be found online at:
http://pedsinreview.aappublications.org/site/misc/Permissions.xhtml
Reprints Information about ordering reprints can be found online:
http://pedsinreview.aappublications.org/site/misc/reprints.xhtml
Pneumonia
Rani S. Gereige, MD, MPH,*
Practice Gap
Pablo Marcelo Laufer, MD†
The epidemiology of pneumonia is changing; chest radiographs and routine laboratory
testing are unnecessary for routine diagnosis of community-acquired pneumonia in chil-
Author Disclosure dren who are candidates for outpatient treatment.
Drs Gereige and Laufer
have disclosed no Objectives The readers of this article are expected to:
financial relationships
1. Know the cause, clinical manifestations, differential diagnosis, and general approach
relevant to this article.
to the diagnosis, treatment, and prevention strategies of the different types of
This commentary does
pneumonia in children of various age groups.
not contain discussion
2. Be aware of the challenges that face the clinician in making an accurate diagnosis of
of unapproved/
pneumonia due to the inaccuracies and shortcomings of the various laboratory and
investigative use of
imaging studies.
a commercial product/
3. Know the complications of pneumonia in children and their appropriate diagnostic
device.
and therapeutic strategies.
Introduction
Pneumonia is commonly encountered by emergency department and primary care clini-
cians. Childhood pneumonia remains a significant cause of morbidity and mortality in de-
veloping countries, whereas mortality rates in the developed world have decreased
secondary to new vaccines, antimicrobials, and advances in diagnostic and monitoring tech-
niques. (1) This review focuses on pneumonia in children: its causes in various age groups,
clinical manifestations, indications for hospitalization, and the challenges that clinicians face
in making an accurate diagnosis despite the new and emerging diagnostic tests.
Epidemiology
The incidence of pneumonia varies by age groups and between
developing and developed countries. Worldwide, the overall
Abbreviations annual incidence of pneumonia in children younger than 5
BAL: bronchoalveolar lavage years is 150 million to 156 million cases, (2)(3) leading to
CAP: community-acquired pneumonia an estimated 2 million deaths per year, most of which occur
CA-MRSA: community-associated methicillin-resistant in developing countries. (4) Forty percent of cases require hos-
Staphylococcus aureus pitalization. (5) In developed countries, the annual incidence of
ELISA: enzyme-linked immunosorbent assay pneumonia is estimated at 33 per 10,000 in children younger
HIV: human immunodeficiency virus than 5 years and 14.5 per 10,000 in children ages 0 to 16 years.
hMPV: human metapneumovirus In the United States, pneumonia is estimated to occur in 2.6%
IGRA: interferon gamma release assay of children younger than 17 years. Fortunately, the mortality
LRTI: lower respiratory tract infection rate in developed countries is less than 1 per 1000 per year. (3)
MRSA: methicillin-resistant Staphylococcus aureus According to the World Health Organization (WHO),
MSSA: methicillin-sensitive Staphylococcus aureus pneumonia is the single largest cause of death in children
PCR: polymerase chain reaction worldwide, leading to an annual death of an estimated 1.2 mil-
RSV: respiratory syncytial virus lion children younger than 5 years. This accounts for 18% of all
VATS: video-assisted thoracoscopic surgery deaths of children younger than 5 years worldwide. (6)
WHO: World Health Organization Cases of pneumonia occur throughout the year; however,
the incidence is increased during the colder months in
*Editorial Board. Department of Medical Education, Miami Children’s Hospital, Miami, FL.
†
Division of Pediatric Infectious Diseases, Miami Children’s Hospital, Miami, FL.
temperate climates for unknown reasons. It is presumed inflammatory process of the lungs, including airways, al-
that person-to-person transmission of respiratory drop- veoli, connective tissue, visceral pleura, and vascular
lets enhanced by indoor crowding, impaired mucociliary structures. Radiologically, pneumonia is defined as an in-
clearance, and the peak of viral infections that led to viral filtrate on chest radiograph in a child with symptoms of
pneumonias with secondary bacterial pneumonias are the an acute respiratory illness. (1)(7)
cause of this peak. In tropical climates, peaks of respira-
tory infections are seen sporadically throughout the year. Walking Pneumonia
(4) Table 1 highlights the risk factors of pneumonia in Walking pneumonia is a term typically used in school-aged
neonates and older children and teens. children and young adults with clinical and radiographic ev-
idence of pneumonia but with mild symptoms in which the
Definitions respiratory symptoms do not interfere with normal activity.
Before further discussion of this topic, it is important to dis- Typically, Mycoplasma pneumoniae has been implicated as
cuss the definitions of the various terms related to pneumonia. the organism presumably responsible for walking pneumonia.
pneumonia. (9) Pneumonia that affects those individuals infection. This mechanism is still not clear, but animal
living in chronic care facilities and those who were re- models suggest that influenza A enhances transmission
cently hospitalized fall in this category as well. of bacteria such as S aureus. (13)
New viruses emerged in the past few years. The hu-
man metapneumovirus (hMPV) was described in 2001.
Etiology Often considered to be a pathogen associated with bron-
A large number of microorganisms cause pneumonia, chiolitis, it is described in association with pneumonia.
ranging from viruses to bacteria and fungi (Table 2). Children younger than 5 years are susceptible to hMPV
The etiologic agents of pneumonia depend on the pa- infection, and infants younger than 2 years with primary
tient’s age. In neonates (0-3 months of age), maternal infection are particularly at risk of severe infection.
flora, such as group B streptococcus and gram-negative bac- Seroepidemiologic studies indicate that virtually all chil-
teria, are common causes that are vertically transmitted. dren are infected with hMPV by 5 to 10 years of age.
Overall, Streptococcus pneumoniae remains the most com- In one series, hMPV was isolated in 8.3% (second only
mon bacterial cause of pneumonia in children older than 1 to respiratory syncytial virus [RSV]) of cases of radio-
week, whereas viruses account for 14% to 35% of cases. (7) logically diagnosed CAP. Children with hMPV were
(10) In children ages 3 months to 5 years, 50% to 60% of older than those with RSV (mean age of 19 vs 9 months)
cases are associated with viral respiratory infections. (11) In and had a higher incidence of gastrointestinal symptoms
school-aged children (>5 years), atypical organisms, such and wheeze. Indicators of severity (such as saturations
as M pneumoniae and Chlamydophila (previously known on admission, respiratory rate, and duration of stay) were
as Chlamydia) pneumoniae, are more common. (12) no different in hMPV compared with other viruses. (13)
Mycoplasma pneumoniae remains the leading cause of (14)
pneumonia in school-age children and young adults. The human bocavirus is in the parvovirus family. Al-
New vaccines and emerging antibiotic resistance led to though it has not been cultured yet, it can be identified
a change in the pathogens implicated in pneumonia. The by electron microscopy. Initially, its role in pneumonia
first vaccine that affected the epidemiology of pneumonia was unclear. Preliminary evidence suggests that nearly
in the United States was the conjugated Haemophilus in- all children have produced antibodies to human bocavirus
fluenzae type b vaccine (1990). It drastically reduced in- by school age, and most newborns receive antibodies from
vasive disease by this organism. In 2000, the their mothers. (13)(14)
pneumococcal conjugated 7-valent vaccine not only de-
creased the rates of invasive disease significantly (98.7 Clinical Manifestations
cases per 10,000 in 1998–1999 vs 23.4 cases per Pneumonia in children is a challenging diagnosis because
10,000 in 2005) but also decreased the incidence of the presenting signs and symptoms are nonspecific, might
pneumonia that required hospitalization and ambulatory be subtle (particularly in infants and young children), and
visits in children younger than 2 years. (10)(12)(13) The vary, depending on the patient’s age, responsible patho-
rates for children ages 1 to 18 years, however, remained gen, and severity of the infection. (1)(4)(7)(13)
stable. Conjugated vaccines reduce nasopharyngeal colo- In all age groups, fever and cough are the hallmark of
nization. This effect benefited nonimmunized adults pneumonia. (4) Other findings, such as tachypnea, in-
older than 65 years through herd immunity. As expected, creased work of breathing (eg, nasal flaring in infants),
the pneumococcal conjugated 7-valent vaccine led to and hypoxia, may precede the cough. The WHO uses ta-
a shift of the most common serotypes that cause disease chypnea and retractions to effectively diagnose pneumo-
in children, and the 13-valent pneumococcal conjugate nia in children younger than 5 years but tachypnea
vaccine introduced in 2010 provides additional coverage becomes less sensitive and specific as age increases (in
against common pneumococcal serotypes 1, 3, 5, 6A, 7F, children >5 years). (4) Most of the clinical signs and
and 19A, further decreasing the incidence of pneumonia symptoms have a low sensitivity and specificity except
that requires hospitalization. (10)(12)(13) for cough, crackles (rales), retractions, rhonchi, and nasal
Community-associated methicillin-resistant Staphylococcus flaring (in young infants), which are highly specific but
aureus (CA-MRSA) should be considered in cases of not sensitive, meaning that their absence might help rule
complicated pneumonia with empyema and necrosis. out the disease. (1) The rate of diagnosed pneumonia in
The latter can be severe when associated with influenza patients with fever but no cough or tachypnea is 0.28%.
infection. In the last few years, clinicians have encoun- Upper lobe pneumonias may present with a clinical pic-
tered severe secondary bacterial infections after influenza ture suggestive of meningitis due to radiating neck pain.
complicated pneumonia
• Staccato cough
despite treatment
onset pneumonia
are effective against LRTI caused by Mycoplasma in chil- and cyanosis. Young infants may present with lethargy, poor
dren. (15) Mycoplasma pneumoniae may be also associated feeding, or irritability. Although the presence of fever is not
with a variety of extrapulmonary manifestations (Table 3). specific for pneumonia, it may be the only sign in occult
Chlamydia pneumoniae is indistinguishable from pneu- pneumonia. In a systematic review, tachypnea was twice
monia caused by other factors. Extrapulmonary manifes- as frequent in children with vs without radiographic pneu-
tations of C pneumoniae infections may include the monia, and its absence was the most valuable sign for ruling
following: out the diagnosis, making it the most sensitive sign. Other
signs of respiratory distress include increased work of
• Meningoencephalitis
breathing (intercostal, subcostal, or suprasternal retractions;
• Guillain-Barré syndrome
nasal flaring; grunting, head bobbing; use of accessory
• Reactive arthritis
muscles), apnea, and altered mental status. Signs of respira-
• Myocarditis
tory distress are more specific than fever or cough for LRTI
Viral pneumonia has a gradual insidious onset. The pa- but not as sensitive.
tient usually experiences nontoxic effects, with upper re- Lung examination is a key part of the assessment. Aus-
spiratory tract symptoms, and auscultatory findings are cultation of all lung fields should be performed, listening
more likely to be diffuse. Wheezing is more frequent in for crackles (rales) or crepitations, the presence of which
viral than bacterial pneumonia. correlates with pneumonia. The absence of these findings
does not rule out pneumonia. Other findings consistent
General Approach with consolidated lung parenchyma might include de-
History and Physical Examination creased breath sounds, egophony, bronchophony, tactile
The approach to the child with suspected pneumonia be- fremitus, or dullness to percussion. Again, wheezing is
gins with a detailed history and careful physical examination. more likely in viral or atypical pneumonia, but when
History is more likely to reveal fever, with associated respi- wheezing is only heard unilaterally and is associated with
ratory symptoms, including cough and tachypnea. On phys- fever, bronchial obstruction (intrinsic or extrinsic) and
ical examination, the clinician must pay special attention to associated bacterial infection should be suspected.
the general appearance of the patient and assess for hypoxia Splinting, dullness to percussion, distant breath sounds,
and friction rub are suggestive of pleural effusion and
must be confirmed by imaging. It is important to men-
tion that many patients clinically suspected of having
pneumonia are treated empirically with no need for im-
Clinical Manifestations of
Table 3.
aging or laboratory workup except for severely ill chil-
Mycoplasma pneumoniae dren with hypoxia, respiratory distress, and inability to
Infections (7) eat or drink.
1. Severe disease, hypoxemia, or significant respiratory origins or dictate management, particularly in the out-
distress that requires hospitalization. patient setting. (7)(8) A complete blood cell count with
2. Inconclusive clinical findings. differential is typically performed in children who are
3. To rule out other causes of respiratory distress (eg, candidates for hospitalization (Table 4). Peripheral eo-
foreign body, heart disease, underlying cardiopulmo- sinophilia suggests Chlamydia trachomatis in infants
nary conditions). with afebrile pneumonia of infancy. Acute phase reac-
4. Prolonged fever and worsening symptoms despite ad- tants, such as erythrocyte sedimentation rate, C-reactive
equate antibiotic coverage to rule out complications protein, and serum procalcitonin, should not be routinely
(parapneumonic effusion, pneumothorax). measured in fully immunized children with mild disease
5. As part of the workup of a young infant with fever but may be useful in monitoring response to treatment
without a source and leukocytosis. in children hospitalized with severe or complicated pneu-
monia. (11)(16) Other blood tests might include serum
Follow-up chest radiographs are not routinely indi-
electrolytes to assess for degree of dehydration and to rule
cated in children who are adequately treated and recov-
out hyponatremia secondary to syndrome of inappropriate
ered. Follow-up radiographs are indicated in complicated
antidiuretic hormone secretion.
pneumonias that are clinically unstable, in patients receiv-
ing adequate antibiotic coverage for 48 to 72 hours with
poor clinical improvement or worsening, and in recurrent Microbiologic Tests
pneumonias that involve the same lobe to rule out a sus- BLOOD CULTURES
pected anomaly, chest mass, or foreign body. Children • Not routinely indicated in the outpatient setting in
with complicated pneumonia treated with chest tube children who have nontoxic effects and fully immu-
placement or video-assisted thoracoscopic surgery (VATS) nized due to low yield (only positive in 10%-12% of
do not require routine daily chest radiography if they are children). (8)
clinically stable and improving. • In patients with parapneumonic effusion or empyema
When indicated, chest radiographs should be obtained the yield increases to 30% to 40%.
in the posteroanterior upright position in children younger • Should be obtained in children hospitalized with se-
than 4 years and in the supine anteroposterior position in vere disease, who fail to demonstrate response despite
younger children. A lateral view is preferred, and a lateral adequate antibiotic coverage, or in children with com-
decubitus view (with affected side down) should be ob- plicated pneumonia. (16)
tained when a pleural effusion is suspected. • Follow-up blood cultures are not necessary in patients
Bedside ultrasonography of the chest was studied and with clear improvement.
compared with chest radiographs. In one prospective cohort
study of 200 patients, ultrasonography had an overall sensi- NASOPHARYNGEAL SAMPLES. Nasopharyngeal cul-
tivity of 86% (95% CI, 71%-94%) and a specificity of 89% tures do not provide useful information because the bac-
(95% CI, 83%-93%). Specificity increased to 97% in children teria recovered are usually normal upper respiratory tract
with consolidation greater than 1 cm by chest radiographs. flora and do not necessarily correlate with the cause of
The authors concluded that bedside ultrasonography was pneumonia. Polymerase chain reaction (PCR) is now
found to be a highly specific, noninvasive, radiation-free test available for the detection of several pathogens in naso-
that can be used by clinicians to diagnose pneumonia. (17) pharyngeal samples as discussed below. The identification
of bacteria by PCR in nasopharyngeal samples is not as
Laboratory Testing useful for the same reason expressed above.
Routine laboratory testing is not indicated to diagnose pneu- SPUTUM CULTURES
monia, particularly in children who are stable, are nonhy- • Difficult to obtain and induce in young children (<5
poxic, and have suspected CAP and are candidates for years) and in outpatient setting.
outpatient treatment. Patients with hypoxemia, severe respi- • Should be obtained in older hospitalized children,
ratory distress, possible complicated pneumonia, or associ- children who are in intensive care, those who have
ated comorbid conditions may need further workup. complicated pneumonia, or those who do not respond
to empiric therapy; good-quality sputum samples can be
Blood Tests obtained.
A complete blood cell count with differential does not • An adequate sputum specimen for examination is one
allow differentiation among bacterial, atypical, or viral with:
PLEURAL FLUID
Indications for
Table 4. • When pleural fluid is more than minimal in amount, it
should be obtained through a diagnostic (and possi-
Hospitalization (8)(16) bly therapeutic) thoracentesis and sent for Gram
• Hypoxia (oxygen saturations <90%-92%)
stain and culture ideally before administration of
• Infants <3-6 months with suspected bacterial antibiotics.
infection (unless a viral cause or Chlamydia • Because most children have already received antibiotics
trachomatis is suspected and they are normoxemic and by the time the pleural fluid is sampled, thereby signif-
relatively asymptomatic) icantly reducing the yield of conventional cultures, an-
• Tachypnea:
B Infants <12 months of age: respiratory rate >70
tigen testing and PCR may be helpful in identifying the
breaths/min causative agent.
B Children: respiratory rate >50 breaths/min • Studies such as pH, glucose, protein, and lactate
• Respiratory distress: apnea, grunting, difficulty dehydrogenase rarely change management and are
breathing, and poor feeding not recommended, except for white blood cell
• Signs of dehydration
• Inability to maintain hydration or oral intake
count with differential to differentiate bacterial
• Capillary refill time >2 seconds from mycobacterial causes and from malignancy.
• Infants and children with toxic appearance or (16) Table 5 highlights the laboratory findings in
suspected or confirmed to have infection with empyema.
a virulent organism (CA-MRSA or group A
Streptococcus)
• Underlying conditions comorbidities that:
B May predispose patients to a more serious course Rapid Tests
(eg, cardiopulmonary disease, genetic syndromes, Nasopharyngeal swab specimen for rapid testing by PCR
neurocognitive disorders) or immunofluorescence may be useful. (8) A positive
B May be worsened by pneumonia (eg, metabolic
rapid test result for viruses in inpatient and outpatient set-
disorder) tings might decrease the need for further testing or for
B May adversely affect response to treatment (eg,
immunocompromised host, sickle cell disease) starting antibiotic therapy; it may also give the opportu-
• Complications (eg, effusion/empyema) nity for starting antiviral therapy early. (16) Rapid tests
• Failure of outpatient therapy (48-72 hours with no exist for the following microorganisms:
response)
• Caretaker unable to provide appropriate observation or • RSV
to comply with prescribed home therapy • Influenza viruses
Indications for intensive care admission include: • Parainfluenza viruses
• Severe respiratory distress or impending respiratory • Adenovirus
failure requiring • Mycoplasma pneumoniae
B Intubation and mechanical ventilation
• Chlamydophila pneumonia
B Positive pressure ventilation
• Coronaviruses
• Recurrent apnea or slow irregular respirations
• Cardiopulmonary monitoring due to cardiovascular • Bordetella pertussis
compromise secondary to: • Picornavirus (rhinovirus and enterovirus)
B Sustained tachycardia
• hMPV (can only be identified by PCR)
B Inadequate blood pressure
B Requires pharmacologic support of blood pressure or The PCR tests for pneumococcus in sputum and
perfusion blood are not recommended because their sensitivity
B Altered mental status due to hypercarbia or hypoxemia
and specificity in children have not been conclusively
• Pediatric Early Warning Score >6 established.
CA-MRSA¼community-acquired methicillin-resistant Staphylococcus
aureus.
Antigen Detection and Serologic Testing
Urinary antigen detection tests have low sensitivity and
B 10 or fewer epithelial cells high false-positive rates. (8)(16) Hence, they are not rec-
B And 25 or more polymorphonuclear leukocytes ommended for the diagnosis of pneumococcal pneumo-
under low power (100). nia in children. (16)
B A predominant microorganism and/or intracellu- Pleural fluid antigen detection: In children with para-
lar organisms suggest the etiologic agent. pneumonic effusion or empyema whose pleural fluid
Laboratory Findings in
Table 5.
a. Quantiferon Gold: Measures interferon gamma
produced by lymphocytes
Empyema (2)(4)(16) b. ELISA spot: Measures the number of lymphocytes
producing interferon gamma both in response to
Studies Empyema
specific M tuberculosis antigens.
pH <7.1
Glucose <40 mg/dL IGRAs measure response to antigens not present in
Lactate dehydrogenase >1000 IU/mL BCG or Mycobacterium avium; therefore, it has better
Gram stain and culture with or Bacteria specificity than tuberculin skin testing, especially in
without polymerase chain reaction children who had received BCG vaccine in whom fre-
Gross appearance Purulent
quent purified protein derivatives can cause a boosting
effect.
2. Urine antigen testing for legionellosis due to serogroup 1.
culture was obtained after antibiotic therapy, a positive 3. Serum and urine antigen testing for histoplasmosis.
pneumococcal antigen in the pleural fluid can be helpful 4. Histoplasmosis serologic testing (immunodiffusion
in confirming the cause. and complement fixation).
Routine serologic testing for specific pathogens (eg, 5. Cryptococcus antigen detection in serum.
S pneumoniae, M pneumoniae, C pneumoniae) is not in- 6. The following tests can be used as part of the workup
dicated because results do not usually influence manage- of the immunocompromised patient with suspected
ment. Viral serologic testing is not practical because acute pneumonia:
and convalescent specimens are needed. Serologic testing
a. b-D-Glucan levels: b-D-Glucan is part of the cell wall
for Chlamydophila species is not readily available.
of yeast and fungi and even Pneumocystis jirovecci
Mycoplasma pneumoniae, when suspected in an older
and can be elevated in fungal pneumonias. (18)
child, is often treated empirically. However, serologic and
b. Galactomannan levels: Galactomannan is part of the
PCR testing can be helpful in evaluating the younger
cell wall of molds, such as aspergillus. Antigen levels
child or in establishing the diagnosis in patients with ex-
in bronchoalveolar lavage (BAL) or serum are pos-
trapulmonary (particularly central nervous system) man-
itive in suspected pneumonia due to aspergillus.
ifestations. The most widely used serodiagnostic test is
(19)
enzyme-linked immunosorbent assay (ELISA); however,
the complement fixation test has better specificity. It The clinician must be aware that certain antibiotics,
measures early IgM (predominantly) and IgG antibodies such as piperacillin-tazobactam or transfusion with blood
(to a lesser extent) to M pneumoniae. A positive result is or blood-derived products such as intravenous immuno-
defined as follows: globulin, may induce false-positive test results. (20)
• A 4-fold or greater increase in titer in paired sera OR
• A single titer of greater than or equal to 1:32
Invasive Studies
Antibody titers rise 7 to 9 days after infection and peak Invasive studies to establish the cause of pneumonia in
at 3 to 4 weeks. A 4-fold decline in titer also is diagnostic children are reserved for the critically ill child or the child
if late samples are obtained. The presence of antibodies with significant comorbidity whose initial diagnostic
either by enzyme immunoassay or complement fixation workup is inconclusive and in whom the risk of establish-
is highly sensitive for the detection of M pneumoniae ing the diagnosis outweighs the risk of the invasive pro-
infection. A major disadvantage of these tests is their cedure. (16) Invasive studies are rarely needed. Invasive
false-positive results, particularly during inflammatory re- studies include the following:
actions, such as neurologic syndromes, bacterial meningitis,
• Bronchoscopy with BAL - Quantitative culture techni-
and acute pancreatitis.
ques differentiate true infection from upper airway
Less commonly used diagnostic tests are as follows:
contamination.
1. Tuberculin skin testing or Quantiferon gold (children • Morning gastric lavage through a nasogastric tube for
>5 years old): If pulmonary tuberculosis is suspected, acid fast bacilli stain and culture is used in the diagnosis
either tuberculin skin testing (purified protein deriva- of tuberculosis.
tive) or interferon gamma release assays (IGRAs) can • The BAL technique for obtaining cultures in intubated
be used. There are 2 available IGRAs: patients uses a catheter inside a catheter, avoiding
sampling the upper airway and directly obtaining cul- no role in the treatment of pneumonia. (21)(22) Zinc sup-
tures from the alveoli. Because of the anatomy of the plementation has been studied and found to be an effective
lungs, samples are obtained from the right lower lobe. adjunct to decreasing the incidence and prevalence of
• Computed tomography or ultrasonography-guided pneumonia in children 2 to 59 months. (23)(24) In most
percutaneous needle aspiration of the affected lung cases of CAP, the chances of having a specific etiologic di-
tissue. agnosis are low, leading the clinician to treat empirically.
• Lung biopsy either by a thoracoscopic or thoracotomy The Figure gives highlights of the decision tree of the ap-
approach is rarely used in United States, but open bi- proach to the child with suspected pneumonia.
opsy yields diagnostic information that may affect
medical management in up to 90% of patients. In Outpatient Management
one study, open lung biopsy confirmed the infectious EMPIRIC THERAPY. Antimicrobial therapy is not rou-
cause in 10 of 33 patients, 8 of whom had a prior non- tinely recommended in preschool children with pneumonia
diagnostic BAL. Lung biopsy is commonly used in im- (viruses are more common). (21) Because S pneumoniae
munocompromised patients. remains the most commonly implicated pathogen, amoxi-
cillin or amoxicillin-clavulanate remains the most appropri-
ate first-line antimicrobial agent used empirically for CAP in
Differential Diagnosis fully immunized, healthy, young preschool children with
When the clinician is faced with a child presenting with fe- mild to moderate symptoms. (25) Clavulanate adds the
ver, tachypnea, cough, respiratory distress, and infiltrates benefit of action against b-lactamase–producing organ-
on chest radiograph, the diagnosis of pneumonia is highly isms (H influenza and Moraxella catarrhalis). S pneu-
likely. (7) Other diagnoses, however, must be considered. moniae resistance to penicillin is due to a penicillin-
In a neonate with respiratory distress, congenital anatom- binding protein (PBP2x) that has decreased affinity
ical cardiopulmonary anomalies must be ruled out, such as to b-lactams. Increasing the dose of amoxicillin (90-
tracheoesophageal fistula, congenital heart disease, and 100 mg/kg daily) may overcome this mechanism of re-
sepsis. In infants and young children, foreign body aspira- sistance and should be prescribed if the clinician suspects
tion (even if no history of any witnessed aspiration), bron- resistance (eg, children in day care or siblings in day
chiolitis, heart failure, sepsis, and metabolic acidosis may all care, history of frequent infections). Amoxicillin-clavu-
cause tachypnea. In these cases, a careful history and phys- lanate is dispensed in 2 different amoxicillin-clavulanate
ical examination and a supportive imaging study can dis- ratios: 7:1 and 14:1. The 14:1 ratio should be used when
tinguish pneumonia from other conditions. high-dose amoxicillin is required to reduce the possibility
In adolescents and young adults, Lemierre syndrome of antibiotic-associated diarrhea.
(jugular vein suppurative thrombophlebitis) must be In school-aged children and teens with a clinical pic-
considered. Lemierre syndrome is typically caused by ture compatible with atypical CAP, coverage using a mac-
Fusobacterium species that infect the carotid sheath and rolide (azithromycin or clarithromycin) should be
spread to lungs and mediastinum. considered. A systematic review of studies in developing
Children who present with respiratory distress and countries found no significant difference in the treatment
wheezing may have CAP; however, first-time wheezing failures or relapse rates between 3- and 5-day courses of
of asthma with or without bronchiolitis can be the true antibiotics in children ages 2 to 59 months with outpa-
diagnosis. A patient with asthma or bronchiolitis may tient management of CAP. (26)
have a radiographic picture that is normal or has infiltrates In children with moderate to severe CAP suspected of
that could potentially be due to atelectasis. having influenza infection and because early antiviral
Other entities that may mimic pneumonia on clinical therapy provides the maximum benefit, treatment with
examination or on radiographs in children are listed in antiviral therapy should not be delayed until confirmation
Table 6 of a positive influenza test result. It is also worth noting
that treatment after 48 hours of symptoms might still
provide clinical benefits in severe cases of influenza. (16)
Treatment
Treatment of pneumonia varies between inpatient and
outpatient settings. In either setting, supportive care in- Inpatient Management
cludes the use of antipyretics, suctioning, and hydration Table 4 highlights the indications for hospitalizations and
when needed. Mucolytics and cough suppressants have intensive care admission.
B Bronchopleural fistula
where from every 8, 12, or 24 hours B Septicemia
later. On the basis of the currently B Cerebral abscess
Sickle Cell
Pneumonia Discharge
Table 10. In patients with sickle cell anemia who present with fever,
hypoxia, and respiratory distress due to acute chest syn-
Criteria (15) drome, atypical bacterial pathogens are primarily the cul-
• Clinical improvement (activity level, appetite) prits. Other causes include S pneumoniae, S aureus, and
• Afebrile for 12-24 hours H influenzae.
• Sustained pulse oximetry >90% on room air for 12-24 Other special considerations for therapy include the
hours following:
• Baseline and stable cardiorespiratory and mental status
• Ability to tolerate oral anti-infective therapy and • Residence or travel to certain geographic areas that are
ability of caretaker to administer it endemic for specific pathogens, such as tuberculosis
• Ability to tolerate oral intake of food and fluids (Asia, Africa, Latin America, and Eastern Europe),
• For children who had a chest tube, the tube must have
been discontinued 12-24 hours before discharge with or exposure to individuals at high risk for tuberculosis,
no clinical signs of deterioration including homeless, incarcerated individuals, and
• Availability of a follow-up plan before discharge HIV-infected patients.
• Exposure to certain animals such as the deer mouse
(hantavirus), bird droppings (Histoplasmosis), birds
(Chlamydophila psittaci), sheep, goats, or cattle (Coxiella
• Other opportunistic pathogens include Fusarium spe-
burnetii – Q fever)
cies and Pneumocystis jirovecii (formerly known as
Pneumocystis carinii).
• Viral pathogens to be considered include rubeola, cy- Prevention and/or Control
tomegalovirus, varicella zoster virus, and Epstein-Barr The most effective prevention method based on strong
virus. evidence is active immunization of children against
• Atypical mycobacteria are a significant pathogen in H influenzae type b, S pneumoniae, influenza, and per-
children infected with human immunodeficiency virus tussis. Influenza virus vaccine should be administered an-
(HIV). nually to all infants 6 months or older and to adult
• HIV-positive patients or patients receiving immuno- caretakers of infants younger than 6 months. The latter
suppressive or chronic steroid therapy must be treated should also receive the pertussis vaccine. High-risk infants
for latent tuberculosis. (21) should receive the RSV-specific monoclonal antibody–
The treatment of HIV-infected children depends based on the American Academy of Pediatrics recom-
on their CD4 cell count. Most children in the United mendation. (4)(16) Several measures can be adopted
States benefit now from antiretrovirals and have nor- to prevent or decrease transmission. Because transmission
mal immune status so their treatment parallels those occurs by droplet or contact, good hand washing and
without HIV infection. Those children whose CD4 good personal hygiene are the most important measures.
cell count is low are at risk of unusual pathogens, such Standard isolation precaution is required in hospitalized
as Pneumocystis jirovecii or cryptococcus; consulting patients with pneumococcal pneumonia and negative iso-
with an infectious disease specialist is recommended. lation in patients with TB. Other measures include limit-
(28) ing exposure to infected individuals and to cigarette
smoke. Additional infection control measures based on
cause include the following:
Cystic Fibrosis
Pneumonia in patients with cystic fibrosis is caused by in- • Respiratory syncytial and parainfluenza viruses – gown
fection by S aureus, P aeruginosa, and H influenzae and gloves (ie, contact precautions).
(mostly nontypable strains) early in their disease. Older • Influenza virus, group A Streptococcus (for the first 24
children with cystic fibrosis have multiple drug-resistant hours of treatment), MSSA, Bordetella pertussis (until pa-
gram-negative organisms, such as Burkholderia cepacia, tient has received 5 days of effective therapy), and M
Stenotrophomonas maltophilia, and Achromobacter xylo- pneumoniae – mask within 3 ft (ie, droplet precautions).
soxidans. Aspergillus species and nontuberculous myco- • Adenovirus – contact and droplet precautions.
bacteria also may also cause disease in this population. • Methicillin-resistant S aureus – special organism pre-
These patients rarely get rid of their bacteria, so reviewing cautions; contact and droplet precautions and dedi-
previous cultures is very important. cated patient equipment.
26. Sutijono D, Hom J, Zehtabchi S. Efficacy of 3-day versus 5-day 29. Wonodi CB, Deloria-Knoll M, Feikin DR, et al; Pneumonia
antibiotic therapy for clinically diagnosed nonsevere pneumonia in Methods Working Group and PERCH Site Investigators. Evalua-
children from developing countries. Eur J Emerg Med. 2011;18(5): tion of risk factors for severe pneumonia in children: the Pneumonia
244–250 Etiology Research for Child Health study. Clin Infect Dis. 2012;54
27. Gilchrist FJ. Is the use of chest physiotherapy beneficial in (Suppl 2):S124–S131
children with community acquired pneumonia? Arch Dis Child. 30. Chang AB, Chang CC, O’Grady K, Torzillo PJ. Lower
2008;93(2):176–178 respiratory tract infections. Pediatr Clin North Am. 2009;56(6):
28. Punpanich W, Groome M, Muhe L, Qazi SA, Madhi SA. 1303–1321
Systematic review on the etiology and antibiotic treatment of 31. Esposito S, Cohen R, Domingo JD, et al. Do we know when,
pneumonia in human immunodeficiency virus-infected children. what, and for how long to treat? antibiotic therapy for community-
Pediatr Infect Dis J. 2011;30(10):e192–e202 acquired pneumonia. Pediatr Infect Dis J. 2012;31(6):e78–e85
PIR Quiz
This quiz is available online at http://pedsinreview.org. NOTE: Learners can take Pediatrics in Review quizzes and claim credit online only. No paper
answer form will be printed in the journal.
1. Anticipatory guidance to parents of children recovered from pneumonia includes discussion about the length
of time that cough may persist. What is the length of time that cough may normally persist after a community-
acquired pneumonia?
A. 2 weeks.
B. 4 weeks.
C. 2 months.
D. 3 months.
E. 4 months.
2. Prevention of pneumonia in children includes active immunization of adult caretakers of infants younger than
6 months against which of the following pathogens?
A. Bordetella pertussis.
B. Haemophilus influenzae type b.
C. Neisseria meningitidis.
D. Respiratory syncytial virus.
E. Tuberculosis.
3. A 10-year-old boy presents with a history of fever, headache, malaise, mild sore throat, and worsening
nonproductive cough. Lung examination reveals diffuse crackles. Chest radiographs reveal bilateral diffuse
patchy infiltrates. The next step in management is to prescribe:
A. Amoxicillin.
B. Amoxicillin-clavulanate.
C. Azithromycin.
D. Cephalexin.
E. Penicillin.
4. A previously healthy, fully immunized 3-year-old girl presents with a 4-day history of gradual onset of runny
nose, cough, fatigue, and slightly fast breathing. She continues to drink liquids but refuses solid foods.
Temperature is 38.3˚C. Physical examination reveals a tired-appearing child with mild tachypnea and subtle
intercostal retractions. Lung examination reveals adequate aeration and crackles over all lung fields bilaterally.
The next step in management is:
A. Amoxicillin (50 mg/kg daily).
B. Blood culture.
C. Chest radiographs.
D. Close observation and follow-up.
E. Complete blood cell count.
5. One week ago, a previously healthy 6-year-old boy was seen in the outpatient clinic with a 5-day history of
fever, chills, fatigue, muscle aches, and cough. Laboratory testing revealed a diagnosis of influenza A for which
he was not treated because of delay in diagnosis. Today he returned to the clinic looking significantly worse,
with tachypnea, dyspnea, and retractions. Chest radiography suggests a small empyema in the right lower lobe.
The next step in management is to prescribe:
A. Ampicillin.
B. Amoxicillin-clavulanate.
C. Azithromycin.
D. Ceftriaxone and clindamycin.
E. Penicillin.
Corrections
In the April 2013 article “Pelvic Inflammatory Disease” (Trent M. Pediatr Rev. 2013;34(4):163–172), the video link at the
end of the second-to-the-last paragraph in the section titled “Does Outpatient Treatment Work for Adolescents?” should
read as follows: http://www.youtube.com/watch?v=lGuXF8vpujQ.
Readers can also search PID and Johns Hopkins on the general YouTube web page to locate the video.
Also, the beginning of the second sentence in the following paragraph should read, “One Brazilian trial has demonstrated
that ceftriaxone 250 mg intramuscularly (IM) plus 1 g of azithromycin given orally at baseline each week for 2 weeks…”.
The journal regrets these errors.