Epidemiology, clinical presentation, and evaluation of parapneumonic effusion and empyema in children

INTRODUCTION — Parapneumonic effusion is defined as pleural effusion associated with lung infection, mainly pneumonia. Many parapneumonic effusions are associated with bacterial infections, but they also may be caused by viral, fungal, and mycoplasma pneumonias and tuberculosis. The effusion results from the spread of inflammation and infection to the pleura. Much less commonly, infections in other adjacent areas (eg, retropharyngeal, vertebral, abdominal, and retroperitoneal spaces) may spread to the pleura, resulting in an effusion. Parapneumonic effusions are seen in approximately 2 to 12 percent of children with pneumonia and up to 28 percent of those requiring hospitalization [1,2].

Early in the course of parapneumonic effusion, the pleura becomes inflamed; subsequent leakage of proteins, fluid, and leukocytes into the pleural space forms the effusion. In the early stage, the pleural effusion is usually sterile, with a low leukocyte count. With time, the effusion can become thick and fibrotic and bacteria can invade the fluid, resulting in empyema. The risk for developing a true empyema is determined by a balance between host resistance, bacterial virulence, and timing of presentation for medical treatment [3].

The epidemiology, etiology, pathophysiology, clinical presentation, and evaluation of parapneumonic effusion and empyema in children will be reviewed here. The management of parapneumonic effusion and empyema in children is discussed elsewhere. (See "Management and prognosis of parapneumonic effusion and empyema in children".)

The evaluation and management of parapneumonic effusion in adults are also discussed separately. (See "Pleural fluid analysis in adults with a pleural effusion" and "Imaging of pleural effusions in adults" and "Epidemiology, clinical presentation, and diagnostic evaluation of parapneumonic effusion and empyema in adults".)

DEFINITIONS

●Parapneumonic effusion is defined as an exudative pleural effusion associated with lung infection (pneumonia). Early in the disease course, the effusion usually is free-flowing (also known as a "simple" effusion) and sterile.

●Loculated parapneumonic effusion refers to the presence of septations within the effusion, which interfere with the free flow of fluid. Loculations can be detected by imaging (ultrasonography or computed tomography [CT]). Loculations are caused by accumulation of proteinaceous debris in the fluid as the disease progresses. (See 'Pathophysiology' below.)

●Empyema is defined as the presence of bacterial organisms on Gram stain and/or grossly purulent fluid in the pleural cavity.

●Complicated parapneumonic effusion is often used to refer to either loculated effusion or empyema with changes in the pleural fluid due to bacterial invasion into the pleural space. Because bacteria are cleared rapidly after antibiotic therapy, cultures of fluid from complicated parapneumonic effusions are often negative.

●Complicated pneumonia refers to pneumonia with any complication, including loculated parapneumonic effusion; empyema; or, more rarely, pneumothorax, necrotizing pneumonia, or lung abscess.

This topic will focus on parapneumonic effusions related to bacterial infections, which account for the majority of clinically significant parapneumonic effusions in children. It will not address parapneumonic effusions from other causes (see 'Other pathogens' below). The term "parapneumonic effusion/empyema" will be used when referring to the spectrum of clinical parapneumonic disease associated with bacterial pneumonia.

EPIDEMIOLOGY — Parapneumonic effusion/empyema develops in 2 to 12 percent of children with pneumonia and in up to 28 percent of children requiring hospitalization [1,2]. It is most common among young children; rates are 3.7, 3.9, and 1.3 cases per 100,000 population among children <2 years, 2 to 4 years, and 5 to 17 years, respectively. Males and females are equally affected [1]. The mortality rate is low but may be higher in infants [4,5]. Most deaths are due to the acute pneumonia or sepsis rather than the pleural disease.

In the United States, the incidence of parapneumonic effusion/empyema rose from 2.1 per 100,000 children in 1997 to 1999, peaked at 3.6 cases per 100,000 in 2006 to 2009, then fell to 2.0 per 100,000 by 2011 to 2013 [6]. These trends are partly explained by changes in the incidence of empyema caused by Streptococcus pneumoniae (pneumococcus). After the 7-valent pneumococcal conjugate vaccine (PCV7) was introduced in the United States in 2000, admission rates for pneumococcal pneumonia decreased, but those for pneumococcal empyema continued to increase, probably because PCV7 does not cover several virulent pneumococcal serotypes [7]. After the PCV13 was introduced in the United States in 2010, there were substantial decreases in hospital admissions for pneumococcal pneumonia and pneumococcal empyema [6,8-11]. Similar trends were seen in other countries after the introduction of PCV13 [12-18].

However, in a German surveillance data report, the incidence of parapneumonic/empyema decreased until 2013, then increased again through 2018, with increased proportion of pneumococcal serotype 3, which might not be well covered by the PCV13 vaccine [19].

Other important pathogens are Staphylococcus aureus and streptococcal species. (See 'Bacterial infections' below.)

Parapneumonic effusion/empyema is more prevalent in the spring and winter than in the summer and fall [4,20], perhaps because of an association between seasonal influenza and bacterial pneumonias [21,22]. (See "Seasonal influenza in children: Clinical features and diagnosis", section on 'S. pneumoniae or S. aureus coinfection'.)

Certain underlying diseases may increase the risk of parapneumonic effusion/empyema in children. In one review of 61 children with parapneumonic effusion/empyema, 11 percent had an underlying illness or condition [5]. Predisposing problems include immunodeficiencies [4,5,20,23], influenza [24], malignancy [5], Down syndrome [4,20], congenital thrombocytopenia [5], cerebral palsy [4,5,20], prior surgery [5,25,26], tuberculosis [23], congenital heart disease [4], prematurity [4], history of esophageal stricture [4,5], and cystic fibrosis [4].

ETIOLOGY — Parapneumonic effusion/empyema in children occurs primarily in association with an underlying bacterial pneumonia [4,27]. The predominant causative organisms have changed over time with the advent of antibiotic therapy, the development of antibiotic resistance, and the development and widespread use of polysaccharide and conjugate vaccines for Haemophilus influenzae type b and S. pneumoniae [28].

Bacterial infections — Common bacterial pathogens associated with parapneumonic effusion/empyema are S. pneumoniae (pneumococcus) and S. aureus, including methicillin-resistant S. aureus (MRSA). Information regarding the predominant bacteria must be interpreted with caution [29] since the range of reported rates of identification of an infectious organism varies broadly with differences in definitions, inclusion and exclusion criteria, rates of pleural fluid sampling, culture and other microbiologic identification techniques, and pretreatment with antibiotics [4,20,29-31].

●S. pneumoniae (pneumococcus) has been the most common pathogen causing parapneumonic effusions and empyema, but its prevalence is decreasing with changes in pneumococcal vaccines [6,18]. Common serotypes that cause invasive pneumococcal disease are 1, 3, 7F, or 19A [32-36]. These serotypes were not included in the 7-valent pneumococcal conjugate vaccine (PCV7) but are in the PCV13, which was introduced in the United States in 2010. They are also included in the 23-valent polysaccharide vaccine (PPSV23) that is used for high-risk children. Accordingly, the burden of complicated pneumococcal pneumonia in children declined after the PCV7 was replaced with the PCV13 vaccine [8]. In particular, the burden of disease caused by serotypes 1 and 19A decreased (although, the latter was still responsible for a substantial number of cases). Of note, serotype 3 has not decreased, suggesting that PCV13 may not be optimally effective against this serotype [8,37].

More than 90 percent of S. pneumoniae causing complicated pneumonia are penicillin-sensitive, and the proportion of isolates that are penicillin-sensitive has been increasing since 2000. Previously, penicillin- and cephalosporin-resistant S. pneumoniae caused a substantial proportion of cases of necrotizing pneumonia and empyema in some communities [20,27,33,38].

●Streptococcus pyogenes (group A Streptococcus) is increasingly common, accounting for as much as 45 percent of complicated pneumonias; it may be associated with more severe disease than S. pneumoniae [18,39-43].

●Community-associated S. aureus is also a common cause of parapneumonic effusion/empyema, accounting for a majority of cases in some series, most of which were MRSA [44-47]. (See "Methicillin-resistant Staphylococcus aureus infections in children: Epidemiology and clinical spectrum", section on 'Clinical spectrum'.)

●Other bacteria that have been identified in parapneumonic effusion/empyema include Streptococcus viridans [48] and Actinomyces species [49]. including neonates [50]. During the 1980s, H. influenzae type b was a common etiologic organism for empyema in children [4] but has almost completely disappeared as a major pathogen in children in the United States, largely because of widespread immunization of infants [49].

Other pathogens — Parapneumonic effusions have also been reported in up to 10 percent of viral and 20 percent of Mycoplasma pneumonia. However, these effusions are typically small and rarely require intervention in the absence of other underlying diseases [2,51,52], such as sickle cell disease. Parapneumonic effusions also are fairly common complications of pulmonary tuberculosis [53] and some fungal infections (eg, coccidioidomycosis, also known as Valley fever) [54].

PATHOPHYSIOLOGY — Parapneumonic effusion occurs in three stages; the rate of progression is variable and can happen over a matter of hours to days:

●Uncomplicated effusion – In the first stage, the exudate is a simple effusion that is characterized by normal glucose concentration, normal pH, and a low cellular count. The pleural fluid usually layers out on lateral decubitus chest radiographs (image 1). At this stage, the effusion is due to increased pulmonary interstitial fluid passing the visceral pleura into the pleural space and to increased permeability of the pleural space capillaries. This stage tends to last approximately 24 to 72 hours [55].

●Fibrinopurulent – The second stage is known as the fibrinopurulent stage, which is triggered by bacterial invasion of the pleural space, causing empyema. In this stage, large numbers of polymorphonuclear cells accumulate in the pleural fluid; fibrin deposition on the pleural surfaces leads to thickening of the exudate and the formation of loculations, making drainage difficult [56,57]. During this stage, the pleural fluid pH and glucose concentration decrease because of anaerobic utilization of glucose by the neutrophils and bacteria [56,57]. Lysis of neutrophils increases the lactate dehydrogenase concentration to values often in excess of 1000 international units/L. Pleural effusions in the fibrinopurulent stage are sometimes called complicated pleural effusions. Loculated pleural effusions usually do not layer out on decubitus radiographs. This stage may last up to 7 to 10 days [55].

●Organizational – The third stage is known as the organizational stage. In this stage, fibroblasts grow on both parietal and visceral pleural surfaces, forming an inelastic membrane "pleural peel" that restricts lung reexpansion, impairs lung function, and creates a persistent pleural space with ongoing potential for infection [29]. At this stage, thoracentesis may yield a "dry tap" [4]. This stage typically occurs by two to four weeks after initial development of the empyema [55].

The above classifications of parapneumonic effusion/empyema may be helpful in understanding the pathophysiology of disease, but there are no data that correlate these stages with specific management strategies [58].

CLINICAL PRESENTATION — The clinical presentation of the child with parapneumonic effusion/empyema depends to some extent upon when the child presents for medical attention. Some children present with symptoms directly related to the parapneumonic effusion/empyema, whereas others are seen earlier in the course and are treated for pneumonia but fail to respond. Thus, any child who remains febrile or unwell 48 hours after initiation of antibiotic therapy for pneumonia should be reevaluated for potential complications, with repeat examination and chest radiograph [29,59]. (See "Community-acquired pneumonia in children: Outpatient treatment", section on 'Monitoring response' and "Pneumonia in children: Inpatient treatment", section on 'Treatment failure'.)

Symptoms and signs — The most common presenting signs and symptoms in children with parapneumonic effusion/empyema are persistent fever, malaise, decreased appetite, cough, chest pain, and dyspnea [1,20,29,55]. Children with parapneumonic effusion/empyema may lie on the affected side to splint the involved hemithorax and provide temporary pain relief [29].

On physical examination, children with parapneumonic effusion/empyema may present with a range of pulmonary signs and symptoms, from mild tachypnea to overt respiratory distress. Breaths are often shallow due to pleuritic pain and resultant splinting. Fever and cough are present in approximately 90 percent of patients [20,60]. Occasionally, they present in septic shock with hypotension.

Chest examination may reveal a small degree of "new" scoliosis, related to the child's splinting toward the affected side [5,61]. In addition, there may be dullness to percussion, decreased air exchange, and possibly a pleural rub, on the side of the fluid collection [29,55].

Mediastinal shift and tension hydrothorax have been reported in patients with large volumes of pleural fluid (>1000 mL) that cause compression of the lung [62-64]. Hypoalbuminemia is common, particularly in children with large effusions [65]; this is likely caused by shifting of albumin from the serum to the pleural space. Other causes of hypoalbuminemia, including urinary or fecal protein loss and malnutrition/catabolic state, should also be considered.

Laboratory findings — Most children with bacterial pneumonia, with or without parapneumonic effusion/empyema, have blood leukocytosis. If C-reactive protein or erythrocyte sedimentation rate is measured, these usually are elevated. These findings are somewhat useful to distinguish between bacterial and other causes of pneumonia, but findings overlap and cannot reliably differentiate between the etiologies. In addition, some patients with bacterial pneumonia who are severely ill or immunocompromised may have normal or low blood leukocyte counts. Hypoalbuminemia is common. (See "Community-acquired pneumonia in children: Clinical features and diagnosis", section on 'Laboratory evaluation'.)

Secondary thrombocytosis (platelet count of >500,000/microL) is common; in one series, it was noted in 93 percent of children [66]. Thrombocytosis peaks at two to three weeks into the illness and usually returns to normal after three weeks. Antiplatelet therapy is not necessary [29]. Hemolytic anemia has been reported, primarily associated with mycoplasma pneumonia. In addition, pneumococcal infection may be complicated by hemolytic uremic syndrome [67].

EVALUATION

History and examination — In addition to the findings that support a diagnosis of parapneumonic effusion, as described above, it is important to assess the severity of illness and look for underlying conditions that predispose to the development of parapneumonic effusion/empyema, including immunodeficiencies. Pulse oximetry is useful in determining the presence and severity of hypoxia. Oxygen saturations <92 percent suggest severe disease [29,59]. The child's state of hydration should be assessed so that fluid therapy can be initiated, if necessary.

Radiologic evaluation — The evaluation of a child with suspected parapneumonic effusion/empyema begins with chest radiography. If a moderate or large effusion is seen on chest radiograph, and especially if the effusion appears non-free-flowing, then ultrasonography is indicated to more accurately determine the size of the effusion, characterize the nature of the fluid, and evaluate for septations and loculation. Computed tomography (CT) is not routinely necessary in the evaluation of children with parapneumonic effusion/empyema, and its use is generally limited to cases in which there is clinical suspicion for lung necrosis or abscess formation or to rule out other causes for effusion, if clinically indicated.

Plain radiographs — Posteroanterior or anteroposterior and lateral decubitus radiographs can help in making the diagnosis of pleural effusion and in determining the need for thoracentesis and/or chest tube placement. A decubitus or cross-table view permits free fluid to layer out on the dependent chest wall and can help to distinguish fluid from pleural thickening. In older children and adults, a decubitus layer of >1 cm is considered sufficient volume to enable extraction of fluid by thoracentesis [2,68]. (See "Epidemiology, clinical presentation, and diagnostic evaluation of parapneumonic effusion and empyema in adults".)

Plain radiographs do not reliably differentiate simple parapneumonic effusion from a complicated effusion/empyema [29,69]. However, they sometimes help to distinguish between free fluid and loculated effusion. A free-flowing effusion should layer with gravity, as noted on the right lateral decubitus view (image 1). By contrast, a loculated effusion will not be affected by gravity and will not be seen to layer on a decubitus chest radiograph. Of note, the loculations/septations themselves cannot be seen on plain radiograph. In general, ultrasonography is the imaging modality of choice to make this distinction. (See 'Ultrasonography' below.)

Radiographic signs of pleural effusion include obliteration of the costophrenic angle, the "meniscus sign" (a rim of fluid ascending the lateral chest wall), and scoliosis [29], particularly in patients with illness of more than one week's duration [5]. Air-fluid levels in the pleural space suggest the presence of gas-forming organisms, pneumothorax, perforated viscus, or bronchopleural fistula. On the posteroanterior view, opacification of more than one-quarter but less than one-half of the thorax suggests a moderate-sized effusion and opacification of more than one-half of the thorax suggests a large effusion [2]. Patients with moderate or large effusions on chest radiograph should be further evaluated with chest ultrasonography to determine the size and nature of the effusion [2]. If the radiograph shows opacity of the entire hemithorax ("whiteout"), then it is difficult to know whether this is caused by fluid, consolidation within the lung, or extrapulmonary mass. Ultrasonography can help to make that differentiation. Patients with severe respiratory distress or deviation of the trachea away from the involved lung (mediastinal shift) are likely to have a large pleural effusion and should be evaluated and treated urgently.

Chest radiographs also may reveal other findings that help to distinguish parapneumonic effusions from other causes of pleural effusion, such as heart failure (cardiomegaly) and malignancy (lymphadenopathy or bone lesions) [70].

Ultrasonography — Ultrasonography is the preferred imaging modality in the evaluation of patients with moderate to large parapneumonic pleural effusions [58]. It is useful in confirming the presence of fluid in the pleural space [70], determining the nature of the effusion [71], and quantifying the amount of effusion [72]. In particular, the presence of septations and cellular debris and the lack of free movement with gravity suggest a complicated (fibrinopurulent) effusion (image 2). In addition, ultrasonography can identify optimal sites for thoracentesis or chest tube insertion [73-75]. Ultrasonography is superior to CT because of better accuracy in detecting early loculations and septations [71]. Other advantages of ultrasonography compared with CT include its ready availability, lack of radiation exposure, and lack of need for sedation [29]. An ultrasonographic scoring system has been proposed to aid in determining the best approach to therapy [76]. However, the accuracy of ultrasonography depends on the skills and experience of the operator.

Chest computed tomography — Chest CT also can identify the presence of pleural fluid. However, studies have suggested that CT findings lack the accuracy to characterize the nature of fluid and the presence of empyema [77] and do not generally affect management decisions [78,79].

As a result of these limitations, CT should not be performed routinely in the evaluation of children with parapneumonic effusions [29]. However, CT may be useful in evaluating the type and extent of parenchymal disease (eg, necrosis and abscess formation) or to rule out other causes for effusion, if clinically indicated [58]. CT also may be useful in evaluating pulmonary disease in immunocompromised children [29]. In addition, CT may be required before surgical intervention (thoracotomy or thoracoscopy) in order to delineate anatomy and exclude intrapulmonary abscess. If CT scan is to be performed, we suggest using limited focused CT protocols to minimize radiation exposure. (See "Management and prognosis of parapneumonic effusion and empyema in children".)

Blood and sputum cultures — Blood cultures are recommended for children presenting with parapneumonic effusion [29] and are sometimes helpful in identifying the causative organism if pleural fluid culture is negative. Blood cultures are positive in 2 to 20 percent of children with complicated parapneumonic effusions; the yield may be influenced by a history of previous antibiotic treatment [1,80-83]. (See "Community-acquired pneumonia in children: Clinical features and diagnosis", section on 'Approach to microbiologic testing'.)

Bacterial culture also can be performed on sputum, if the patient is able to expectorate, or tracheal aspirates, if the patient is intubated. Cultures from the nasopharynx and throat are not reliable, because organisms cultured from these sites are not necessarily present in the lower airways [29]. The yield of all cultures decreases sharply after systemic or oral antibiotics are given. (See "Sputum cultures for the evaluation of bacterial pneumonia", section on 'Interpretation'.)

Pleural fluid analysis — Drainage of pleural fluid is indicated for some moderate and most large pleural effusions. For patients with no respiratory compromise and with small to moderate pleural effusions, treatment with antibiotics but without pleural drainage is a reasonable alternative [2]. (See "Management and prognosis of parapneumonic effusion and empyema in children", section on 'Moderate or large simple effusion (not loculated)'.)

If a drainage procedure is performed, the pleural fluid should be sent for microbial analysis and cell count, as outlined below. (See 'Microbial analysis' below and 'Cell count' below.)

If a noninfectious etiology is suspected (eg, malignancy or collagen-vascular disease), then additional evaluation of the pleural fluid may be helpful in establishing the cause [2,29]. (See 'Other studies' below and 'Differential diagnosis' below.)

The diagnostic evaluation of pleural fluid is discussed in detail elsewhere (see "Pleural fluid analysis in adults with a pleural effusion"). Information that can be obtained from pleural fluid analysis that may relate to the management of children with empyema is presented briefly below.

Microbial analysis — Pleural fluid is typically sent for microbiologic analysis including Gram stain and bacterial culture [29]; these were reported to be positive in up to 49 percent of cases, with most studies reporting positive culture in less than 25 percent of cases [2]. However, pleural fluid cultures are often sterile because of prior administration of antibiotics [29].

Additional techniques may be used to increase the yield of microbiologic diagnosis in children who have received antibiotics. These include direct and enrichment culture for aerobic and anaerobic organisms, pneumococcal antigen detection (latex agglutination) [84,85], and specific or broad-range polymerase chain reaction (PCR) [34,85-88] and rapid tests for group A streptococcus [89].

●Pneumococcal antigen detection – Pneumococcal antigen detection in pleural fluid samples by latex agglutination is useful in the rapid diagnosis of pneumococcal empyema, particularly in patients who have received antibiotics before pleural fluid aspiration. The value of this assay was evaluated in a study of pleural fluid samples from 78 children with empyema [85]. Pleural fluid specimens were sent for standard culture, pneumococcal antigen detection, broad-range PCR (16S rDNA PCR), and specific PCR (pneumolysin PCR). A microbiologic diagnosis was confirmed by culture or broad-range PCR in 60 cases (77 percent). Among the 40 cases of pneumococcal empyema, 23 were diagnosed with PCR and culture and 17 with PCR alone (all 17 had received antibiotics before pleural fluid aspiration). Latex agglutination testing identified pneumococcal antigen in the pleural fluid of 90 percent of cases of pneumococcal empyema and was negative in 95 percent of nonpneumococcal empyema (sensitivity and specificity of 90 and 95 percent, respectively).

●PCR – Detection of specific pathogens in pleural fluid through PCR also can be used if the organism is not readily identified by culture or pneumococcal antigen detection [2]. If a broad-range PCR is not available, specific testing should be done for S. pneumoniae (pneumococcus) and S. pyogenes. Most reports describe substantially improved yield for PCR compared with conventional culture techniques [86-88,90,91]. Depending on the assay used, PCR results may not provide information about antibiotic resistance in the identified strain of S. pneumoniae or S. aureus. This is not a concern for group A streptococcus, because isolates are all susceptible to penicillin. (See "Pneumococcal pneumonia in children", section on 'Diagnosis'.)

●Rapid group A streptococcus test – Rapid tests for group A streptococcus may detect group A streptococcus antigen in culture-negative empyema fluids [89].

In a study of adults with complicated parapneumonic effusion, 16S rDNA sequencing in combination with conventional cultures significantly improved the yield of detecting bacteria and helped to guide antibiotic therapy [92].

New techniques such as cell free plasma next-generation sequencing have been reported to increase the yield in identifying pathogens, especially in cases of prior antibiotic therapy, but are not always available [93-95].

Cell count — Pleural fluid should be sent for differential cell count [29]. This is helpful in differentiating bacterial from mycobacterial, fungal, or malignant etiologies. Nucleated white blood cell counts above 50,000 cells/microL are typical of complicated parapneumonic effusions, including empyema, but cell counts may be between 10,000 and 50,000 early in the course of the illness. Chronic exudates (eg, from tuberculous pleurisy and malignancy) tend to have low nucleated cell counts. The differential cell count can be helpful in determining an etiology; mononuclear predominance suggests malignancy or tuberculosis [96-99]. (See "Tuberculous pleural effusion" and "Pleural fluid analysis in adults with a pleural effusion".)

Other studies

●Biochemical studies – Analysis of pleural fluid parameters, such as pH and levels of glucose, protein, and lactate dehydrogenase, rarely change the management of a patient with community-acquired pneumonia and are not recommended by the Pediatric Infectious Diseases Society/Infectious Diseases Society of America guideline [2]. Although some characteristics are associated with empyema (pH <7.0, glucose <40 mg/dL [2.2 mmol/L], lactate dehydrogenase >1000 international units), these measures do not help with decision-making in most cases [29,70,100,101]. Instead, the decision to drain a parapneumonic effusion can be made based on clinical criteria, including clinical status of the patient and radiographic findings suggestive of large or complicated effusions [70,102]. (See "Management and prognosis of parapneumonic effusion and empyema in children".)

In adults, biochemical studies are sometimes used to help distinguish between pleural exudates (due to infection, malignancy, and immunologic responses) and transudates (due to imbalances in hydrostatic and oncotic pressures, such as congestive heart failure or fluid from the cerebrospinal or peritoneal spaces). However, transudates are uncommon causes of pleural effusions in children, except in those with known predisposing disease. (See "Pleural fluid analysis in adults with a pleural effusion", section on 'Classification as exudative or transudative'.)

If biochemical studies are performed, it is important to keep pleural fluid specimens tightly capped and on ice to prevent spurious changes in pH or glucose levels. Testing should be performed immediately or fluid should be stored at 0ºC for no more than two hours; pH is lowered if fluid is stored at room temperature and raised upon exposure to air.

●Cytology – Pleural fluid should be sent for cytology if there is any suspicion that the effusion is not secondary to infection (eg, absence of pulmonary consolidation/pneumonia or fever, or presence of mediastinal mass or lymphadenopathy) [29]. Overtly bloody pleural fluid in the absence of a history of trauma also suggests malignancy.

Bronchoscopy — Flexible bronchoscopy is usually not indicated, and it is not routinely recommended for children with parapneumonic effusion, unless foreign body aspiration is suspected [29]. (See "Airway foreign bodies in children".)

Other tests — Additional tests may be indicated in the diagnosis or assessment of severity of parapneumonic effusions in children. These include:

●Tests commonly performed during monitoring and supportive care are:

•Complete blood count with differential helps to monitor the course of infection and identify comorbidities such as anemia; anemia is present at some point in approximately 20 percent of cases [4]. The white blood cell count may be helpful in monitoring progress, but it does not help in distinguishing complicated from uncomplicated effusions [56,103]. Secondary thrombocytosis (platelet count of >500,000/microL) is common. (See 'Laboratory findings' above.)

•C-reactive protein may be useful in monitoring progress, but it has not been established as a criterion for hospital discharge or when to convert to oral antibiotic treatment [2,29].

•Serum electrolytes are often obtained as part of routine care. If hyponatremia is discovered, the possibility of inappropriate secretion of antidiuretic hormone should be considered because it is occasionally associated with pneumonia and other pulmonary diseases [104,105]. (See "Pathophysiology and etiology of the syndrome of inappropriate antidiuretic hormone secretion (SIADH)", section on 'Pulmonary disease'.)

●Tests performed for selected patients include:

•Evaluation for tuberculosis should be performed if risk factors for tuberculosis are present or if there are signs and symptoms suggesting pulmonary tuberculosis (chronic cough, fevers, weight loss, or hilar adenopathy). If the clinical suspicion is high, an interferon-gamma release assay (IGRA) and sputum and/or gastric aspirates for acid-fast bacilli should be performed, in addition to tuberculin skin testing, since the tuberculin skin testing may be falsely negative. (See "Tuberculous pleural effusion" and "Tuberculosis disease in children: Epidemiology, clinical manifestations, and diagnosis".)

•Measurement of various serum titers (eg, mycoplasma, antistreptolysin O) may indicate the causative organism but are not useful for clinical management of pneumonia, because interpretation requires acute and convalescent samples. Instead, pathogens are typically identified through cultures of pleural fluid, with or without rapid tests that identify specific pathogens. (See 'Microbial analysis' above.)

•Evaluation for cystic fibrosis and immunodeficiency to be considered in children with parapneumonic effusion who have a history of recurrent bacterial infections, poor growth, or had Pseudomonas aeruginosa as the infecting organism.

•Blood gases (venous or capillary) are important to assess the degree of respiratory compromise in children in severe respiratory distress, requiring high concentrations of supplemental oxygen, or with suspected sepsis. These tests may show evidence of respiratory alkalosis in the tachypneic child, respiratory acidosis in a patient with acute respiratory failure, or metabolic acidosis in the septic child, but usually shows a mixed picture.

DIFFERENTIAL DIAGNOSIS — A majority of large pleural effusions in children are related to underlying bacterial pneumonia. Other considerations are:

●Tuberculosis – The possibility of tuberculosis in a patient with pleural effusion is suggested by the presence of relevant epidemiologic risk factors for tuberculosis or the presence of hilar adenopathy. Tuberculous pleural effusion is the second most common form of extrapulmonary tuberculosis (after lymphatic involvement) and is the most common cause of pleural effusion in areas where tuberculosis is endemic. Tuberculous effusion may progress to tuberculous empyema. (See 'Other tests' above and "Tuberculous pleural effusion".)

●Malignancy – Suggested by absence of acute fever or pneumonia and by evidence of an underlying mediastinal mass or lymphadenopathy, especially if associated with prodromal symptoms such as loss of appetite, weight loss, low-grade or intermittent fever, or night sweats (see 'History and examination' above and 'Radiologic evaluation' above). Evaluation for suspected malignancy includes cell counts and cytologic analysis of pleural fluid.

●Other – Other causes of pleural effusion in children include congestive heart failure and collagen-vascular diseases. Other types of fluid in the pleural space include chylothorax, hemothorax, and extravasated fluid from central venous catheters or ventriculoperitoneal shunts.

SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Pediatric pneumonia".)

SUMMARY AND RECOMMENDATIONS

●Epidemiology and etiology – Parapneumonic effusion/empyema in children occurs primarily in association with underlying bacterial pneumonia. Streptococcus pneumoniae (pneumococcus), community-acquired Staphylococcus aureus (including methicillin-resistant S. aureus [MRSA]) and group A streptococcus are the predominant causative organisms. (See 'Epidemiology' above and 'Etiology' above.)

●Clinical presentation – The most common presenting signs or symptoms in children with parapneumonic effusion/empyema are fever, cough, malaise, anorexia, chest pain, and dyspnea. Failure to improve after 48 hours of appropriate therapy for pneumonia is another important presenting scenario. (See 'Clinical presentation' above.)

●Initial evaluation – Children with the above clinical findings should be evaluated by chest radiography. If a moderate to large effusion is suspected, especially if it is non-free-flowing (loculated), then ultrasonography is the preferred imaging modality to confirm the presence of fluid in the pleural space and to evaluate for loculations and septations. Cultures of blood and sputum (if the patient is able to expectorate) should be obtained. (See 'Radiologic evaluation' above and 'Blood and sputum cultures' above.)

●Tests on pleural fluid – If thoracentesis or drainage of pleural fluid is performed, the pleural fluid should be sent for analysis (see "Management and prognosis of parapneumonic effusion and empyema in children", section on 'Thoracentesis'). The pleural fluid should be sent for the following studies:

•Gram stain and culture (aerobic and anaerobic). (See 'Microbial analysis' above.)

•Cell count and differential. (See 'Cell count' above.)

•Biochemical studies including pH, glucose, and lactate dehydrogenase should be considered. They are not routinely indicated in children with presumed parapneumonic effusion/empyema, because they do not affect clinical management decisions. However, they may be useful if a noninfectious cause of effusion is suspected, such as malignancy or collagen-vascular disease. (See 'Other studies' above.)

•Additional pleural fluid studies that may help to identify the cause, and which are particularly useful if the child received antibiotics before pleural fluid was obtained, include:

-Specific or broad-range polymerase chain reaction (PCR) studies.

-Pneumococcal antigen detection (latex agglutination studies). (See 'Microbial analysis' above.)

-Cytology, if malignancy is suspected (eg, because of lymphocytic predominance on cell count, mediastinal mass, or lymphadenopathy). (See 'Other studies' above.)

●Other tests – Other tests that may be helpful in identifying the causative organism, assessment of severity, presence of complications, and monitoring progress include (see 'Other tests' above):

•Tuberculin skin testing and sputum (or gastric aspirates) for acid-fast bacilli in patients with risk factors for tuberculosis.

•Complete blood count with differential.

•C-reactive protein may be useful in monitoring treatment response.

•Serum electrolytes (to detect inappropriate antidiuretic hormone syndrome).

•Children with parapneumonic effusion who have a history of recurrent bacterial infections, poor growth, or Pseudomonas aeruginosa as the infecting organism should be evaluated for cystic fibrosis and immunodeficiency. (See 'Other tests' above and "Cystic fibrosis: Clinical manifestations and diagnosis" and "Approach to the child with recurrent infections", section on 'Laboratory evaluation'.)