ED Evaluation and Management of Pediatric Community-Acquired Pneumonia

ED Evaluation and Management of Pediatric Community-Acquired Pneumonia

Respiratory illnesses are a common cause for a visit to the pediatric emergency department (ED). A good number of the patients who present for respiratory infections will have illnesses such as upper respiratory infections (i.e., common colds) and bronchiolitis. However, more serious illnesses may present with respiratory distress. Pneumonia is one such illness, along with bronchiolitis, tracheitis, and laryngotracheobronchitis. This review article will discuss the ED evaluation and management of community-acquired pneumonia (CAP) in children.

— Ann M. Dietrich, MD, Editor

Case Study

An 18-month-old boy presents with poor feeding, fever, and difficulty breathing. He was seen for a cold last week and was diagnosed with an upper respiratory infection. He was put on azithromycin and is on day 3/5. His mother reports that he spiked a fever with some chills today, prompting her to bring him in. He has only fed twice today and has had only one wet diaper. He has required breathing treatments before and has eczema. He is on no home medications and has no allergies. His family history is significant for asthma in his parents and siblings. His vital signs are as follows: T 39.4° C, HR 186, RR 68, BP 83/41, SaO2 92% RA. Your exam is significant for moderate-to-severe respiratory distress, copious clear nasal secretions, grunting, intercostal and suprasternal retractions, nasal flaring, fair air entry bilaterally, but diminished breath sounds on the right base with diffuse rhonchi and wheezes, capillary refill time (CRT) is 3-4 seconds. He has strong pulses centrally and peripherally. He is awake and fussy with exam, but consolable.

What would your next interventions be? What diagnostic testing would you do, if any? Would you continue antibiotics? If so, which one(s) would you use?

Definitions and Pathophysiology

The World Health Organization (WHO) defines pneumonia as "a severe form of acute lower respiratory infection that specifically affects the lungs."¹ There are many other definitions in the literature with slight variations, but all include signs and symptoms as well as radiographic findings in their description of pneumonia.^1-6 In industrialized countries, where most children are fully immunized and where many facilities have easy access to diagnostic testing, pneumonia is defined as a lower respiratory infection with most, if not all, of the following symptoms/findings: fever, cough, rales or rhonchi on lung exam, hypoxia, and an infiltrate on chest X-ray (CXR). In contrast, in developing countries, due to resource limitations and the infrequency of fully immunized status, the definition of pneumonia is based mainly on symptoms and clinical exam findings such as fever and cough. More specifically, CAP is just that — a pneumonia that is acquired outside the hospital setting. This differentiation is made as it has a different set of offending infectious agents than those pneumonias acquired in the hospital setting.

Although there are many etiologies for pneumonia, the most common offenders are viruses and bacteria. In viral and atypical pneumonia, infection of the cells lining the lower respiratory tract results in a humoral and cellular immune response leading to inflammation of the lower airways with migration of mononuclear lymphocytes to the area and sloughing of the epithelial cells into the airspaces, resulting in an obstruction of the small airways.^2,4,5,7,8 In bacterial pneumonia, bacteria migrate down to the lower airways and proliferate, inciting the humoral and cellular immune response and leading to the migration of the polymorphonuclear leukocytes to the area, deposition of fibrin, and finally macrophage migration to the area, which results in the formation of pus in the airways. On occasion, this process can also lead to a fluid collection in the pleural space, either as a sympathetic effusion or as a pus collection in its own right (empyema).^2,4,5,7,8 In both cases, the resultant inflammation and fluid/debris collection in the airways interferes with air movement and gas exchange, and may result in tachypnea, dyspnea, rales/rhonchi/wheezing on physical exam, and hypoxia.

Epidemiology

As with most things in pediatrics, the causes of pneumonia vary with age. In addition, the list of usual suspects has changed dramatically over the years with the widespread use of immunizations against Haemophilus influenzae B (HiB), measles (included in MMR), and Streptococcus pneumoniae (initially PCV7 and now PCV13).^2,3,8,9 Starting with neonates, infections in this age group are usually acquired perinatally; thus, neonatal pneumonia is often due to Group B Streptococcus (GBS), Chlamydia trachomatis, Gram-negative enteric bacteria, or one of the TORCH infections.^8,10 One particular case is "afebrile pneumonia of infancy," a type of pneumonia that usually occurs between 2 weeks and 4 months of age and is typically caused by Chlamydia trachomatis, cytomegalovirus, Mycoplasma hominis, or Ureaplasma urealyticum.^7,11 Pneumonias in infants up to 5 years of age are more often due to viral infections. This is not to say that infants do not get bacterial infections, but that, more often than not, they start out viral. As children get older (i.e., school age and adolescents), atypical bacteria are usually the culprit behind pneumonias.^2,11

Table 1 lists common microbes that cause pneumonia. The most common viral agents are respiratory syncytial virus (RSV), influenza, parainfluenza, adenovirus, rhinovirus, human metapneumovirus, and coronavirus. The most common bacterial agents are S. pneumoniae, Staphylococcus aureus, H. influenzae (type B and non-typable), and Group A Streptococcus. The two most common atypical bacteria are Mycoplasma pneumoniae and Chlamydophila pneumoniae. Beyond these, depending on the age and the past medical history, other etiologies must be considered such as GBS and Gram-negative bacilli in neonates, tuberculosis, and Pneumocystis carinii, Candida, and Aspergillus infections in the immunocompromised child.^{2-4,7,9,10,11}

One reason pneumonias can be difficult to treat is that, oftentimes, finding the specific organism behind the infection is difficult, if not impossible. In many cases, samples taken for diagnosis in children are actually from the upper respiratory tract, rather than the lower tract, resulting in ambiguity as to whether any bacteria isolated are actually the cause of the pneumonia or if they are just colonizing the upper airway.³ That being said, the most common bacterium isolated has been S. pneumoniae.^2-6,8,9 The other confounding issue is that of co-infections. In many cases (23-33%), particularly the more severe ones, a co-infection of one or more viruses and bacteria can be found.^3,9 As a result of these, identification is difficult, if not impossible, and treatment is frequently empiric, solely based on epidemiologic data, rather than patient-specific data.

The incidence of pneumonia worldwide has been estimated to be about 156 million cases in children younger than 5 years of age, with about 20 million cases requiring admission to a hospital. In industrialized countries, the annual incidence of pneumonia has been estimated to be about 4%, or 33/10,000 in children younger than 5 years of age and 14.5/10,000 in children between 0 and 16 years of age.^1,2,5,12 The WHO noted that there is a 10-fold increase in the incidence of pneumonia when developing countries were compared to industrialized countries.¹ This difference is thought to be due in large part to the routine immunization of children against H. influenza type B (HiB series), measles (MMR series), and S. pneumoniae (PCV7 and PCV13 series), as mentioned above.¹ The mortality rate is equally discordant. In industrialized countries, the WHO reports that 2% of children younger than 5 years of age die from pneumonia. In contrast, this mortality rate is 20% in developing countries. There is speculation that factors such as late recognition of symptoms by health care providers and caregivers, lack of education on the part of caregivers, and lack of resources/treatments are behind such a high mortality rate.¹

Although vaccines have done wonders in terms of reducing the morbidity and mortality of pneumonias, several risk factors are thought to increase a child’s susceptibility. Underlying medical problems affecting the heart and lungs such as cystic fibrosis, asthma, sickle cell disease, and congenital heart disease can predispose a child to developing pneumonia. An association with lower socioeconomic groups has been noted, presumably due to crowded conditions, leading to increased spread of infectious droplets. Similarly, school-age children bring home a variety of viral respiratory illnesses they acquire at school and daycare, subsequently exposing the rest of the family. As with any respiratory illness, smoke (e.g., from pipes, cigarettes, etc.) interferes with the normal function of the airway defenses, resulting in poor mucociliary clearance and diminished macrophage activity and thus a higher risk for infections such as pneumonia.^2,3

Pediatric CAP accounts for more than 500,000 ED visits per year and about 7% of pediatric hospitalizations in the United States.¹² The cost for treatment of pneumonia varies, depending on whether the treatment is done as an outpatient or an inpatient. In 1997, the typical cost for outpatient treatment of CAP was $1464/episode, while the cost for inpatient management was $12,000/episode.³ In the United Kingdom in 2007, an estimated £13-20 million (about $26-40 million)/year was spent on inpatient care for pneumonia in children 0-16 years of age.⁹ The cost for outpatient treatment includes an antibiotic course and follow-up visits. The cost of inpatient treatment is much higher, and includes the use of the hospital facilities, labor (in terms of antibiotic administration and IV fluids), and, more often than not, broad-spectrum (and thus more expensive) antibiotics. In addition, these numbers do not reflect the total cost of pediatric CAP in that they do not include lost wages and lost productivity from the parents having to stay home from work to care for their child. As with many acquired pediatric illnesses, children frequently recover without sequelae, but the cost of treatment can impose a significant financial and emotional burden on the family.

Physical Exam

Unlike adults, the signs for pneumonia in children can be very subtle, particularly in younger age groups. Infants can present with only vague symptoms such as fever, fussiness, and poor feeding. In older age groups, the typical signs are more common: fever, cough, difficulty breathing, and wheezing.^3-5,9,11 On occasion, children may complain of chest pain that can be from tracheobronchial irritation, costochondral strain, or from pleuritis.^4,7,11,13 In addition, other types of pain can be associated with pneumonia. For example, lower lobe pneumonia can be associated with a complaint of abdominal pain, and upper lobe pneumonia can be associated with a complaint of neck pain.^3,7,8,11

As mentioned above, it is difficult to differentiate between the different etiologies of pneumonia. Other than age group and epidemiology, there can be clues in the history that give some indication of the infecting agent. Viral pneumonias usually have a history of persistent cough, coryza, and lower grade fevers (usually < 39° C) over a period of time, usually 2-4 days; children with viral pneumonia are not usually toxic appearing. In contrast, bacterial pneumonias tend to have a history of sudden onset of a high fever after several days of upper respiratory tract infection symptoms (i.e., cough, rhinorrhea), chills, ill appearance, and occasionally pleuritic pain. Neonates with afebrile pneumonia of infancy usually present between 2 weeks and 4 months of age with a history of rhinorrhea, tachypnea, and a staccato cough (separated by inspirations, not coughing fits).^2,4,10 Atypical pneumonias often present very similarly to viral pneumonias starting with nonspecific systemic symptoms such as fever, myalgias, malaise, and headache. After about 3-5 days, a nonproductive cough starts, and hoarseness, sore throat, and in some cases chest pain are present. Occasionally, a rash (usually maculopapular) or arthralgias can be present.^3,4,6,8,9,11

Likewise, the presentation of pneumonia on physical exam also can be very subtle. Sometimes, the only indication of pneumonia may be vital sign abnormalities such as fever, tachycardia, tachypnea, or hypoxia. Obviously, such changes can be seen with a number of illnesses, but their presence in the setting of a history concerning for a lower airway infection should raise the suspicion for pneumonia. The most concerning issues that would require more immediate interventions would be those of severe respiratory distress. Table 2 lists signs and symptoms of respiratory distress.^3,7,9,11,13 Using the ABC mnemonic, the airway can be clear or can have secretions (in the setting of rhinorrhea or nasal congestion with such illnesses as bronchiolitis). In terms of breathing, one can see poor air entry, increased work of breathing (retractions, nasal flaring, grunting), and abnormal adventitial sounds (rales, rhonchi, wheezes). More often than not, there will be tachycardia as part of circulation assessment. This can be due to a fever or dehydration. The capillary refill and pulse quality also may be abnormal. As for disability, the child may have altered mental status (lethargy, somnolence) due to dehydration, hypoglycemia from poor intake, or hypoxemia. On exposure, there may be a rash, again depending on the infectious etiology.

In addition to the signs of respiratory distress, the ED physician specifically should look for abnormal adventitial sounds (rales, rhonchi, wheezing), egophany (a sign of consolidation), and dullness to percussion (a sign of effusion).^2,4,7,11 If a child is being treated for pneumonia and does not seem to be improving, special attention must be paid to look for signs of complications, such as splinting and dullness to percussion that may be indicative of an effusion or empyema or asymmetric breast sounds/chest rise that may be indicative of a pneumothorax from ruptured pneumatocoeles due to S. aureus pneumonia. Other physical exam findings, such as pathognomonic rashes, might help shed some light on the infecting agent.

Diagnostic Testing: Labs

The diagnosis of pneumonia is primarily clinical. Lab and radiologic testing are more useful in assessing the severity of pneumonias and monitoring for progression of disease. Typical blood tests that are sent for pneumonia are a complete blood count with a differential (CBC) and blood culture, in the hopes that the white blood cell count (WBC) or blood culture will give some indication as to whether one is dealing with a viral or bacterial pneumonia. Unfortunately, the WBC is not a great test to differentiate between viral and bacterial pneumonia since it is not reliably higher in cases of bacterial pneumonia.^3,9,11,13 Blood cultures are only rarely positive and thus not very helpful in most cases.^3,4,9,11,14 For those children being treated as outpatients, the chance of a blood culture being positive is < 3%.^3,11 This makes sense given that a good number of pneumonias are viral, rather than bacterial, and that by the time blood cultures are drawn, many of the children already have been on antibiotics. The positive blood culture rate understandably increases in children with pneumonia with any complication to about 8%.³ In about 5% of cases of positive blood cultures drawn in the setting of CAP, the bacteria present was S. pneumoniae.^9,13,15

Because of the lack of consistent utility of CBC with differential and blood culture, there were high hopes that other inflammatory markers might be more helpful in differentiating bacterial from viral infection. Sadly, both C reactive protein (CRP) and erythrocyte sedimentation rate (ESR) also have not been shown to be reliable for this purpose.^3,9,11,13 That being said, these tests (CBC, CRP, and ESR) are often used to show improvement on a given treatment regimen in inpatients (i.e., downward trends in WBC count, CRP, and ESR as the child improves).³ Blood cultures are frequently sent on children who are being admitted, as they tend to be much sicker or have already failed outpatient therapy and thus have a higher risk of being bacteremic. However, whether those children who are admitted and who are not "moderately to severely ill" also need blood cultures is being called into question currently.¹⁶ More recently, there had been much interest in procalcitonin as a test that would be specific for bacterial sepsis and thus give some indication whether a pneumonia was being caused by a bacterium. However, once again, there has not been great evidence to use procalcitonin in this capacity for pediatric pneumonia.³

Several blood tests can help the ED physician assess severity of disease and complications, rather than determining etiology of a pneumonia. A basic metabolic panel is helpful in assessing hydration status, as well as evaluating for other infrequent complications of pneumonia such as syndrome of inappropriate anti-diuretic hormone and hemolytic uremic syndrome.^7,9,11 Perhaps one of the more useful tests in the ED in terms of management of pneumonia is the blood gas, looking for respiratory insufficiency as evidenced by hypercapnea and hypoxia in the setting of tachypnea and the subsequent need for more aggressive ventilatory management with either non-invasive or invasive respiratory support.^3,7,17

Other lab tests that can be used — again usually more for the inpatient management of pneumonia in the hopes of finding a specific infectious agent so that the treatment plan can be further narrowed — are a urine screen for evidence of S. pneumoniae, cold agglutinins, acute and convalescent titers, a respiratory viral PCR, and testing on pleural fluid or lung tissue. Although useful in the treatment of adults, urine screens for S. pneumonia have yet to be shown to be useful in pediatrics.^3,9 Cold agglutinins traditionally have been used to support the diagnosis of infection with M. pneumoniae (i.e., atypical pneumonia). Unfortunately, there are so many other infections and illnesses that can have a positive cold agglutinin test that the test is not particularly useful.^3,5,8 Acute and convalescent titers for different infections can be very useful for those cases that do not improve, but have no utility in the ED setting due to the length of time needed to get results.^3,8,9,11

So if blood and urine tests are not helpful, perhaps samples directly from the respiratory tract and system would be. While this is a good thought, unfortunately, the tests that have been looked at are also of limited utility. A respiratory viral PCR panel can test for evidence of typical viruses causing a respiratory infection.³ These tests can be helpful in showing the presence of a virus in the respiratory tract, but they are often expensive and do not exclude bacterial co-infection.^3,5 In one small study, by Tsolia et al, 35% of the patients who had been hospitalized with CAP were found to have mixed infections, of which 81% were a combination of viruses and bacteria.¹⁸ That being said, the influenza viral test can be useful. In the cases of pneumonia with a flu infection, treatment with an antiviral may help reduce the risk of complications such as bacterial superinfection (usually S. aureus), even if the child is outside the 2-day window recommended to start antiviral therapy.^3,19

Even with samples directly from the lungs and pleura, the utility of testing is limited. Pleural fluid from an effusion or empyema unfortunately rarely grows bacteria.^3,9,11 Lung biopsy and bronchoalveolar lavage would most likely be the most useful in terms of identifying the infecting organism; however, these procedures are so invasive that unless the child is doing poorly, these procedures are not performed.^3,11 One test that has been useful is the gram stain and culture of tracheal aspirates when a child is intubated, as opposed to a sputum sample, which usually shows what organisms have colonized the airway rather than which organism is causing a problem.^3,11

Diagnostic Testing: Radiology

Many physicians feel obliged to obtain a CXR to prove a diagnosis of pneumonia. However, the diagnosis of pneumonia is usually a clinical diagnosis. CXRs often lag behind the clinical picture and thus are not always helpful in most cases of CAP in the ED. Thus, obtaining a CXR in a child without respiratory distress who is being treated as an outpatient for CAP will most likely be low-yield and is not necessary.^3,9,20 However, if a child is admitted, a two-view CXR (PA and lateral) can be useful in assessing severity of illness, particularly since these children have often failed outpatient therapy or are sicker.³

Common findings on CXR for a viral pneumonia are increased interstitial markings in a bronchopneumonic pattern, peribronchial cuffing, hyperinflation, and atelectasis. Atypical pneumonias have a similar appearance on CXR.^4-8,11 In contrast, the CXR of bacterial pneumonia tends to have focal (segmental) infiltrates with air bronchograms.^4-7 One specific kind of pneumonia is the round pneumonia, named for its appearance on CXR of a spherical consolidation that is usually > 3 cm in diameter, solitary, and posterior. Like most other pneumonias, these pneumonias commonly are caused by S. pneumoniae. They also can be caused by other streptococcal bacteria, H. influenzae, S. aureus, and M. pneumoniae.¹¹ Despite the "common" appearances of various types of pneumonia on CXR, they are anything but specific. And to make things even more confusing, there is much inter-rater variation in terms of interpretations of CXR.⁹

However, there may be a role for CXRs in confirming clinical improvement. After a treatment course, a repeat CXR is not necessary unless the child is not improved or worsens.

If a child develops an effusion or empyema, more information can be obtained to help with the treatment of the child with special X-ray views of the chest (e.g., lateral decubitus views to ascertain if a pleural effusion layers or is loculated, inspiratory/expiratory views to look for a pneumothorax from a pneumatocele, or apical views to look for cavitary lesions or tuberculomas in tuberculosis) and other modalities such as chest ultrasound or chest CT, which can help identify rarer complications such as lung abscesses and bronchopulmonary fistulas.^3,11

In terms of diagnostic evaluation, the diagnosis of pneumonia is primarily a clinical one. There is no need, nor is there much use for, blood work or X-rays, particularly in those children who are discharged from the ED.^3,6,13 However, if the child has already failed an outpatient treatment course or is getting worse, blood work (e.g., CBC, blood cultures, CRP, or ESR), respiratory viral PCR, tracheal aspirate (if intubated), and a CXR can be very useful to the inpatient service in terms of assessing the severity of the illness, selection of treatment options, and monitoring for improvement.^3,13

There are two special cases that require more evaluation in the ED. One case is that of neonates, where the over-riding concern should be for sepsis, with pneumonia being only one possible etiology. Thus, neonates should undergo the full septic evaluation with examination of their blood, urine, and CSF, as well as a CXR if there is any concern for pulmonary pathology such as pneumonia. The second special case is that of immunocompromised children. Due to their increased risk of pervasive and invasive infectious processes, these children should be screened with blood work, CXRs, and a respiratory viral panel (or at least a flu test in the appropriate season), depending on their medical problem, as well as their history (e.g., whether or not they have had surgeries or foreign devices such as a port, graft, or stent).^10,17

Treatment: Respiratory and Circulation

As with anything in the ED, treatment is focused on the initial stabilization of a child. As mentioned before, following the ABC mnemonic, the patient’s airway and breathing should be evaluated and treated rapidly. The first step is to open and clear the airway. This is usually going to be more of an issue for infants and those children with cerebral palsy without the ability to handle their secretions. Suctioning (bulb or deep, with or without vacuum assistance) will allow for a better evaluation of the lower airway. As a reminder, aggressive deep suctioning may cause a transient bradycardia in younger infants that usually resolves with cessation of the suctioning and provision of oxygen.

Once the airway is clear, if the child is still in respiratory distress and/or hypoxic, particularly if he continues to be poorly responsive, provision of supplemental O₂ is needed. The goal O₂ saturation (SaO₂) in pneumonia is > 90% or 92%, depending on the consensus statement: the Infectious Disease Society of America (IDSA) or British Thoracic Society (BTS), respectively.^3,9 If the only issue is hypoxia (i.e., no respiratory distress or altered mental status), oxygen can be provided by face mask (with a wall blender unit or venturi mask where the percentage of inspired O₂ can be controlled) or nasal cannula.

Two things should be considered when using these two methods of providing supplemental O₂. First, some children are natural mouth breathers or have so much nasal secretions or congestion that the utility of a nasal cannula is limited. Second, as the flow (number of liters) of oxygen via nasal cannula is increased, at some point the amount of supplemental flow being provided can equal or surpass the tidal volume of the child. Once this happens, the child is receiving high-flow nasal cannula (HFNC), which is similar to CPAP. In other words, in such a situation:

NC flow > (tidal volume)(weight)(respiratory rate)

The average tidal volume is 5-7 mL/kg/breath. For example, a 4 kg infant’s tidal volume will be 20-28 mL/breath. If this infant breathes at 60 breaths/minute, then he would be moving 1200-1680 mL/minute. If 2L by NC is provided, then the infant is receiving more than enough flow (2000 mL/minute) to replace his tidal volume and, thus, is getting high flow by nasal cannula. Knowing this is important when it comes time to disposition the child from the ED.

If the child is hypoxic and in respiratory distress, as evidenced by his work of breathing or need for increasing FiO₂, the next step is to consider non-invasive ventilatory support such as HFNC, CPAP, or BiPAP with humidified O₂. In this situation, the child is spending a lot of time and energy keeping his airways open and dealing with the atelectasis that comes with lower airways disease processes. Oftentimes, one can verify that this is happening and that the child is tiring out by checking a blood gas to look for acidosis and hypercarbia in the setting of severe tachypnea (i.e., CO₂ retention despite tachypnea indicating ineffective gas exchange). All of these methods of ventilatory support allow the ED physician to increase the mean airway pressure and take away some of the work from the child without having to intubate him. If these methods do not work and the child gets worse with increasing respiratory distress or becomes unresponsive, one must always be prepared to intubate. When intubating these children, it would be helpful to get a tracheal aspirate to attempt to identify a specific infectious agent to direct later management.^3,9,11

Another adjunct therapy for these children, particularly if there is a history of asthma or reactive airway disease, is ß-agonist nebulizer treatments (e.g., albuterol) if there is bronchospasm or wheezing on exam. These treatments aid with mucociliary clearance. Pain control, particularly with an anti-inflammatory medication such as an NSAID, can help with chest pain that can accompany pneumonia, allowing the child to stop splinting and take deeper breaths, aiding in pulmonary toilet.^13,19 Routine use of cough suppressants should be avoided, as these will blunt the cough reflex needed for pulmonary toilet.^13,19 Maintenance of hydration is also important, especially as these children usually are not drinking since they do not feel well, despite having increased insensible losses due to their tachypnea and fever. Usually, rehydration will most likely have to occur via IV fluids rather than enterally due to respiratory distress.¹³ Careful consideration should be given before using an NG tube for enteral rehydration (since most of these children will not be able to take oral fluids without problems) as they may cause a physical obstruction in the airway.⁹ One thing to bear in mind in those children who get IV hydration is that as they become more hydrated, they might take a turn for the worse from the respiratory standpoint as the pneumonia "blossoms," resulting in more swelling and congestion in the lower airways.¹⁷

Treatment: Antibiotics and Antivirals

When the child’s ventilation is stabilized, the next step is determining which antibiotics should be given. Many infants will have a viral etiology of their symptoms, so antibiotics may not be necessary. If an infant is well appearing, nontoxic, no respiratory distress, and not hypoxic, consider holding off starting antibiotics and having the family return to the pediatrician for close follow-up.^3,7-9,17,19 However, if the decision is made to treat with antibiotics, which is very common especially since it is difficult to differentiate between viral and bacterial pneumonia, antibiotic choice should be based on the usual bacterial agents from an epidemiologic standpoint.⁴ Every effort should be made to use as narrow spectrum of an antibiotic as possible, in hopes of not creating more antibiotic resistance in the community.³

For those younger children who are discharged, amoxicillin is recommended at 90 mg/kg/day divided BID (max 4 g/day) to target S. pneumonia, the most common offender for CAP.^3,7-9,17,19 If local antibiograms show more resistant strains of S. pneumoniae, amoxicillin should be dosed at 90 mg/kg/day divided TID, which will allow for more time above the minimum inhibitory concentration.^3,6-8,19 If there is a concern of specific bacteria other than S. pneumoniae, other antibiotics can be used (See Table 3). If the child is penicillin allergic, other options include second- or third-generation cephalosporins, clindamycin, levofloxacin, and linezolid.^3,8,9,17,19 The optimal duration of treatment is arbitrary. Most physicians treat uncomplicated CAP for 7-10 days, although no studies with generalizable (i.e., applicable to both developing and industrialized countries) results have been conducted to determine the optimal treatment course.^3,17-19,21 For complicated pneumonias with parapneumonic effusions or abscesses, treatment duration should be > 10 days; however, the exact length of time is a matter of debate. Some advocate for continuing antibiotics 7-10 days after the resolution of fever, while others urge 4-6 weeks of treatment.³ Unfortunately, there is not much in the way of randomized, prospective trials to provide further guidance on this matter. Older children and teenagers tend to have atypical pneumonia. In this age group, a course of a macrolide such as azithromycin (10 mg/kg/day for day 1, then 5 mg/kg/day for days 2-5, max 500 mg/dose) or clarithromycin would be more appropriate. In areas where M. pneumoniae is becoming more macrolide resistant, other options include respiratory fluoroquinolones (such as levofloxacin or moxifloxacin) and doxycycline.^3,5-8,19 (See Table 3.)

For those children who are admitted, a similar rationale for a narrow-spectrum antibiotic, such as ampicillin, initially is recommended unless the child has already been on a narrow-spectrum antibiotic or is not fully immunized.^3,9,17 Review of the use of narrow-spectrum antibiotics has shown that they are as effective as broader coverage at treating children hospitalized for pneumonia, and, in the long run, these children hopefully will be spared a higher risk of subsequent infections with resistant organisms.^22,23 Exceptions to this recommendation are if the child is not fully immunized, is < 12 months old, does not show any improvement within a couple of days, or has already failed a course of antibiotics. In these cases, consideration should be given to either broadening the coverage by changing to a third-generation cephalosporin, such as ceftriaxone or cefotaxime, to provide better coverage for the other possible bacterial causes of pneumonia (other than S. pneumonia) or by adding a macrolide (e.g., azithromycin) to cover for atypical bacteria.^13,17,19 If there is any concern for methicillin-resistant S. aureus (MRSA) with persistent deterioration or development of a parapneumonic effusion or empyema, addition of clindamycin or vancomycin is warranted.³ As soon as the child shows sustained improvement, he can be switched to an oral antibiotic, observed for continued improvement after the switch, and then discharged.^3,9,17,19 (See Table 3.)

Since many of these cases are viral in nature, would antiviral medications be helpful? For the most part, antivirals would not help. The one exception is if there is a proven infection with influenza. In these cases, it has been shown that initiation of antiviral therapy (e.g., Oseltamivir), even if the child is already outside of the first 48 hours of therapy, may provide some benefit in terms of the pneumonia not getting worse. In addition, it is recommended that if antibiotics are started as treatment for bacterial superinfection, some consideration should be given to providing coverage for S. aureus, as there is a high co-infection rate of influenza and S. aureus.^3,9 (See Table 3.)

In terms of other special cases such as neonates and immunocompromised children, the philosophy of narrow coverage is difficult to follow given the number and type of infections that need to be covered is much larger. These cases usually start with a broader coverage and then narrow down once an offending agent is identified. Knowing the past history of these children is important in that coverage must be provided for prior exposures and infections, e.g., the child with a tracheostomy whose trach cultures have grown Stenotrophomonas maltophilia and Serratia marascens or the neonate whose mother was GBS positive and had a history of untreated HSV.

Disposition

Once the child is stabilized and treatment has started, where should they go? The answer to this question goes back to the original question when you were first examining the child: Is the child in respiratory distress? If the child is not in distress, is not requiring any supplemental O₂ or other interventions, is able to tolerate oral intake, and the family seems capable of taking care of the child, then outpatient treatment with close follow up with the pediatrician would most likely be fine (see Table 4). However, if the child or the family falls short on any one of these criteria or has already failed a trial of outpatient treatment, he should be admitted for treatment and observation.^3,19 That being said, the IDSA recommends admitting all infants less than 3-6 months with suspected bacterial pneumonia as well as any child with pneumonia due to a pathogen with suspected increased virulence, such as MRSA.³ As to whether the child should be on a general ward or in the intensive care unit (ICU), each institution has its own criteria as to what can and cannot be managed on the general ward. Of course, if the child is intubated, in shock with hypotension, or has altered mental status, he or she should be admitted to the ICU.

If the child does not respond to therapy (inpatient or outpatient) within a couple of days, the child should be reassessed for any evidence of respiratory or circulatory compromise. Typically, these children will need further diagnostic workup (e.g., blood work as described above, viral testing), as well as a CXR, specifically looking for progression of disease if there was a prior film.³ If a parapneumonic effusion is noted, the child might need further imaging with a CT or ultrasound and, ultimately, a pleural tap or video-assisted thoracoscopy (VATS) procedure to drain off the fluid. If the child continues not to improve, he or she ultimately may need a bronchoalveolar lavage (BAL) or lung biopsy to get a definitive diagnosis to direct therapy.

Although the general concepts behind the two major consensus papers from the IDSA and BTS are not necessarily earth-shattering, they do represent a slight departure from the prior teachings in terms of how to care for these children, particularly the recommendations on limiting diagnostic testing and use of narrow-spectrum antibiotics. Clinical practice guidelines can aid front-line providers by summarizing these recommendations and providing prompts that can be used during the daily care of these children.²⁴ Interestingly, those children who had undergone more of a workup tended to have a longer hospital stay. And while they did not return as often in the 14 days after discharge, they did not necessarily have a lower readmission rate.¹³

Summary

Pneumonia is a lower respiratory infection defined by clinical signs and symptoms, making it a clinical diagnosis. Many of the guidelines include radiographic elements, but such testing modalities are not always readily available. The etiology of pneumonia depends on age and geographic location. What makes pneumonia particularly difficult is that, more often than not, in a given case, one may never definitively identify the causative agent. As with much of ED care, exam and management of these patients are geared toward assessing and stabilizing the patient, specifically looking for and addressing respiratory distress with supplemental oxygen and airway adjuncts. Treatment with antimicrobials depends on age and local epidemiology. In industrialized countries, a high-dose amoxicillin course in younger children should be adequate for outpatient therapy if there is concern that the pneumonia is not viral in nature. A macrolide course (e.g., azithromycin) should be adequate for older children and adolescents who more frequently have atypical pneumonia. Similarly, initiation of inpatient therapy with a narrow-spectrum antibiotic such as ampicillin is a good place to start. If the child is critical or does not rapidly improve within 48 hours or gets worse, coverage should be broadened (e.g., a cephalosporin) or by adding other antibiotics (e.g., macrolides, clindamycin, vancomycin) or antivirals. Disposition depends on how much distress the child is in, the ability of the family to care of the child, the ability of the child to tolerate oral intake, and how much support/treatment the child is requiring.

Case Discussion

This infant is in severe respiratory distress with a history of cold symptoms that have progressed to difficulty breathing, poor feeding, and fever with chills. He has worsened despite starting a course of azithromycin. In this case, there is high concern for bacterial pneumonia, most likely superimposed on the original viral illness. The stabilization of the child should include clearing the airway via suctioning and reassessment. An albuterol treatment can be considered given the history of prior use and the family history of asthma. Ultimately, the child might need supplemental oxygen and possibly ventilatory support such as HFNC, CPAP/BiPAP, or intubation. He should be given IV fluids because of his dehydration, as evidenced by his history of poor intake, tachycardia, and prolonged capillary refill. An antipyretic also would be helpful, as his fever is most likely contributing to some of his clinical picture. Given that he has failed outpatient therapy and has gotten worse, a workup is indicated. Blood work (CBC with diff, CRP, blood culture, and basic metabolic panel) should be obtained, as well as a CXR. A respiratory viral panel can also be considered, if available. In terms of antibiotics, either ampicillin or a cephalosporin such as cefotaxime or ceftriaxone can be given since he failed an azithromycin course; these would be better choices to treat CAP, most likely caused by S. pneumoniae. He should be admitted to the general ward or ICU, depending on how he responds to the initial therapy.

References

1. Wardlaw T, Johanssen EW, Hodge M. Pneumonia: The Forgotten Killer of Children. New York: UNICEF/WHO. 2006.
2. Barson W. Pneumonia in children: Epidemiology, pathogenesis, and etiology. Kaplan SL, Torchia MM, eds. UpToDate Available at: www.uptodate.com. Jan 2014.
3. Bradley JS, Byington CL, Shah SS, et al. The management of community-acquired pneumonia in infants and children older than 3 months of age: Clinical practice guidelines by the Pediatric Infectious Diseases Society and Infectious Diseases Society of America. Clin Infect Dis 2011;53:e25-e76.
4. Lichenstein R, AH Suggs, J Campbell. Pediatric pneumonia. Emerg Med Clin North Am 2003;21:437-451.
5. Rafei K, Lichenstein R. Airway infectious disease emergencies. Pediatr Clin North Am 2006;53:215-242.
6. Shah S, Sharieff GQ. Pediatric respiratory infections. Emerg Med Clin North Am 2007;25:961-979.
7. Harper MB, Fleisher GR. Chapter 92: Infectious Disease Emergencies. In: Textbook of Pediatric Emergency Medicine, 6th ed. Fleisher GR, Ludwig S, eds. Philadelphia: Lippincott Wiliams & Wilkins; 2010:912-916.
8. Sectish TC, Prober CG. Chapter 397: Pneumonia. In: Nelson Textbook of Pediatrics, 18th ed. Kliegman RM, et al, eds. Philadelphia: Saunders Elsevier; 2007:1795-1800.
9. Harris M, Clark J, Coote N, et al. British Thoracic Society guidelines for the management of community acquired pneumonia in childrenp: Update 2011. Thorax 2011;66(Suppl 2):ii1-ii23.
10. Speer ME. Neonatal pneumonia. Garcia-Prats JA, Edwards MS, Kim MS, eds. UpToDate Available at: www.uptodate.com. Mar 2013.
11. Barson W. Pneumonia in children: Epidemiology, pathogenesis, and etiology. Kaplan SL, Torchia MM, eds. UpToDate Available at: www.uptodate.com. Jan 2014.
12. Ross RK, Hersh AL, Kronman MP, et al. Impact of Infectious Diseases Society of America/Pediatric Infectious Diseases Society Guidelines on treatment of community-acquired pneumonia in hospitalized children. Clin Infect Dis 2014;58:834-838.
13. Brogan TV, Hall M, Williams DJ, et al. Variability in processes of care and outcomes among children hospitalized with community-acquired pneumonia. Pediat Infect Dis J 2012;31:1036-1041.
14. Williams DJ. Commentary: Do all children hospitalized with community-acquired pneumonia require blood cultures? Hosp Pediatr 2013;3:177-179.
15. Myers AL, Hall M, Williams DJ, et al. Prevalence of bacteremia in hospitalized pediatric patients with community-acquired pneumonia. Pediatr Infect Dis J 2013;32:736-740.
16. Heine D, Cochran C, Moore M, et al. The prevalence of bacteremia in pediatric patients with community-acquired pneumonia: Guidelines to reduce the frequency of obtaining blood cultures. Hosp Pediatr 2013;3:92-96.
17. Barson W. Community-acquired pneumonia in children: Inpatient treatment. Edwards MS, Mallory GB, Torchia MM, eds. UpToDate Available at: www.uptodate.com: Oct 2013.
18. Tsolia MN, Psarras S, Bossios A, et al. Etiology of community-acquired pneumonia in hospitalized school-age children: Evidence for high prevalence of viral infections. Clin Infect Dis 2004; 39: 681-686.
19. Barson W. Community-acquired pneumonia in children: Outpatient treatment. Edwards MS, Mallory GB, Torchia MM, eds. UpToDate Available at: www.uptodate.com: Sep 2013.
20. Neuman MI, Monuteaux MC, Scully KJ, Bachur RG. Prediction of pneumonia in a pediatric emergency department. Pediatrics 2011;128:246-253.
21. Sutijono D, Hom J, Zehtabchi S. Efficacy of 3-day versus 5-day antibiotic therapy for clinically diagnosed nonsevere pneumonia in children from developing countries. Eur J Emerg Med 2011;18:244-250.
22. Queen MA, Myers AL, Hall M, et al. Comparative effectiveness of empiric antibiotics for community-acquired pneumonia. Pediatrics 2014;133:e23-e29.
23. Williams DJ, Hall M, Shah SS, et al. Narrow vs broad-spectrum antimicrobial therapy for children hospitalized with pneumonia. Pediatrics 2013;132:e1141-e1148.
24. Neuman MI, Hall M, Hersh AL, et al. Influence of hospital guidelines on management of children hospitalized with pneumonia. Pediatrics 2012;130:823-830.