Pediatric Empyema 

Updated: Sep 30, 2016
Author: Peter H Michelson, MD; Chief Editor: Girish D Sharma, MD, FCCP, FAAP 



Bacterial pneumonia with associated pleural empyema is the most common cause of pleural effusion found in the pediatric population. Parapneumonic effusions are predominately exudative and occur in as many as 50-70% of patients admitted with a complicated pneumonia.

See the image below.

Most parapneumonic effusions treated with the appr Most parapneumonic effusions treated with the appropriate antimicrobials of sufficient duration resolve without the development of complications or sequelae. The series of radiographs represent a patient treated with thoracentesis alone. Figure A illustrates the patient at presentation. Note the amount of fluid present. In Figure B, the radiograph demonstrates progression of the pleural fluid accumulation with further airspace disease and scoliosis noted. Despite the radiographic evidence of disease progression, the patient was clinically improving. Figure C illustrates the radiograph at follow-up, 6 months following completion of therapy. Resolution of the parapneumonic effusion with no evidence of pleural thickening or fibrosis occurred.

The pulmonary infections of these patients extend into the pleural space and require more extensive therapy, with associated increased morbidity and extended hospital stay. Involvement of the pleural space with pulmonary infections has been recognized since ancient times. Aristotle identified the increased morbidity and mortality associated with empyema and described drainage of empyema fluid with incision. The practice of surgical drainage as part of therapy for empyema has continued into the era of modern medicine. In his 1901 text, The Principles and Practice of Medicine, Sir William Osler, MD, stated that empyema should be treated as an ordinary abscess, "with incision and drainage."[1] Of note, Osler underwent a rib resection for his own postpneumonic empyema, from which he ultimately expired.

Complicated parapneumonic effusions are appearing more frequently by most accounts, with reported increases in incidence rates in both in Europe and the United States. In England, the rate of admission with a diagnosis of empyema increased over the last decade, most notably in children aged 1-4 years. In addition, the identification of Streptococcus pneumoniae as the primary pathogen has also been reported, both in both the United States and abroad.[2] Whereas the overall rate of parapneumonic effusions may have stabilized over the last decade, the rate of bacterial resistance, specifically methicillin-resistant Staphylococcus aureus, has predominated.[3]


The definition of a parapneumonic effusion is a pleural effusion associated with either bacterial pneumonia or lung abscess or, rarely, external introduction of organisms associated with chest wall trauma. The development of parapneumonic effusions is gradual, with the pleural fluid collection most commonly divided into 3 stages. The progression of pleural fluid collection evolves from stage 1-3.

In stage 1, the exudative stage, the pleural inflammation from a contiguous infection results in increased permeability and a small fluid collection. At this stage, the effusion is thin and amenable to thoracentesis alone, contains neutrophils, has normal pH and glucose levels, and is often sterile. Stage 2, the fibrinopurulent stage, is characterized by invasion of the organism into the pleural space, progressive inflammation, and significant polymorphonuclear (PMN) leukocyte invasion. The increase in fibrin deposition also results in partitions or loculations within the pleural space. Inflammation is characterized by progressive decreases in the pleural fluid glucose and pH levels and increased protein and lactate dehydrogenase (LDH) levels. The last stage, stage 3, is the organizing stage, in which a pleural peel is created by the resorption of fluid and is associated with fibroblast proliferation that can result in parenchymal entrapment.[4]

Empyema is defined by the presence of intrapleural pus and, for definition purposes, represents an advanced parapneumonic effusion. Complicated parapneumonic effusions (CPE) refer to those fluid collections that require thoracentesis, tube thoracostomy, or surgery for their resolution.

The accumulation of pleural effusions can rapidly occur in the presence of infection. The pleural surface is a mesothelial membrane that covers the lungs and chest wall. The resultant pleural space is a potential space, containing only small volumes of transudative fluid, with a protein content of less than 1.5 g/dL. This fluid is normally composed of lymphocytes, macrophages, and mesothelial cells, with an absence of neutrophils. Interaction of the mesothelium with the invading bacteria, PMNs, and resultant inflammatory mediators can increase pleural permeability. Further PMN recruitment ensues, which results in the increased production of neutrophil chemotactic mediators, ultimately leading to significant pleocytosis.

Activation of the coagulation cascade is common with pleural empyema. The pleural inflammatory response favors increased procoagulant activity, as well as depressed fibrinolytic activity, which favors fibrin deposition. Loculations result with these fibrin strands covered rapidly by a meshwork of fibroblasts that both proliferate and deposit basement membrane proteins onto the surface of the pleura. These proteins obscure the separation of the visceral and parietal pleura and lead to the formation of the pleural peel.

Following initiation of appropriate therapy, the inflammatory cellular and cytokine production recedes, and the PMN predominance of the parapneumonic effusion decreases. An influx of macrophages assists in the clearance of PMNs, with resolution of the inflammatory process. Migration of pleural mesothelial cells into areas of denuded mesothelium results in the reepithelialization of the pleura and recovery of normal function; however, following exuberant pleural inflammation, the potential for pleural fibrosis and restrictive lung disease is enhanced. The mechanisms that lead to either the development of pleural fibrosis or pleural repair with normal recovery are not well understood and need further investigation and characterization.



United States

Parapneumonic effusions are predominately exudative and occur in as many as 50-70% of patients admitted with a complicated pneumonia. These patients have extension of their pulmonary infection into the pleural space and require more extensive therapy, with associated increased morbidity and extended hospital stay. Although incidence rates appear to be increasing, they remain at approximately 10 cases per 10,000.[3]


With an increased incidence of pneumonia and tuberculosis worldwide, the frequency of CPE is likely to be even higher. Literature emerging from Asia suggests empyema is a significant concern, although these retrospective reviews cannot accurately describe the incidence, and variable rates of identification suggest more standardized practices are needed.[5]


Early recognition of pneumonia with parapneumonic effusion, effective intervention to identify the infecting organism, and initiation of definitive therapy reduce the morbidity and complications associated with this process.


The bacteriology of the infection varies with patient age. In the pediatric population, the most commonly implicated organisms are S pneumoniae, S aureus, and group A streptococci. The latter may be observed as a complication of an infectious skin disorder, such as varicella, impetigo, or infectious eczema. Haemophilus influenzae is rarely encountered since the advent of the H influenzae B vaccine. Methicillin-resistant S Aureus is a concern in the older age group.




Clinical manifestations of empyema include the following:

  • Most patients with empyema present with clinical manifestations of bacterial pneumonia. Their symptoms are characterized by an acute febrile response, pleuritic chest pain, cough, dyspnea, and, possibly, cyanosis.

  • The inflammation of the pleural space may cause abdominal pain and vomiting.

  • Patients frequently exhibit characteristic splinting of the affected side.

  • Symptoms may be blunted, and fever may not be present in patients who are immunocompromised.


Physical examination findings can vary as well.

  • Auscultation reveals crackles, decreased breath sounds, and, possibly, a pleural rub, if the process is recognized before a large amount of fluid accumulates.

  • Dullness to percussion and decreased breath sounds are likely but difficult to elicit in the younger child, who, because of discomfort, may be less cooperative with the examination.

  • Physical findings and presentation may vary depending on the organism and the duration of the illness.


See the list below:

  • Increased pleural permeability associated with pneumonia or lung abscesses, as well as contiguous infections of the esophagus, mediastinum, or subdiaphragmatic region, may extend to involve the pleura.

  • Similarly, retropharyngeal, retroperitoneal, or paravertebral processes may extend to adjacent structures and involve the pleura, as well.

  • Host factors that contribute to alterations in pleural permeability, such as noninfectious inflammatory diseases, infection, trauma, or malignancy, may allow accumulation of fluid in the pleural space, which becomes secondarily infected.

  • The bacteriology of the pleural space varies with patient age. In the pediatric population, the most common implicated organisms are S pneumoniae, S aureus, and group A streptococci. H influenzae is rarely observed since the advent of the H influenzae B vaccine.

  • Because of the use of oral antibiotics before the recognition of the parapneumonic effusion, most specimens cultured are sterile; thus, the relative incidences of the aforementioned organisms are not known.

  • Anaerobic infections secondary to aspiration and fungal or mycobacterial infections in immunosuppressed patients are also reported.

  • Mycoplasma pneumoniae, viruses, and atypical pneumonias can also present with exudative pleural effusions, although mononuclear cells primarily characterize them.

  • A study investigated the risk factors of empyema after acute viral infection and the association between empyema and the use of nonsteroidal anti-inflammatory drugs (NSAIDs). The study found that NSAIDs use during acute viral infection is associated with an increased risk of empyema in children.[6]





Laboratory Studies

The following studies are indicated in empyema:

  • CBC count

  • Blood culture: Blood culture is obtained to assist in the identification of the offending organism. In pediatric patients, in whom sputum production is uncommon, identifying the cause of the pulmonary symptoms early in the course of a pulmonary infection is difficult. However, with parapneumonic effusions, the patient may become bacteremic as the organism invades into the pleural space, and a blood culture may reveal the organism.

  • Serum lactate dehydrogenase (LDH) level

  • Total protein level

  • Glucose concentration

  • Bacterial, mycobacterial, and fungal cultures

  • Serologic studies of the aspirated pleural fluid

  • pH level

  • Amylase concentration

  • Lipid stain or triglyceride level

  • Cell count and differential: Although the pleural fluid obtained at thoracentesis is typically purulent, with an elevated WBC count and a predominance of polymorphonuclear cells (PMNs), an effusion evaluated early in the infectious process may well be more transudative, with a less cellular WBC count and a differential that is less PMN predominant. Regardless of the cell count and differential, the treatment should be based on clinical course, pending the culture results. Cytokine analyses of pleural fluid have been performed in experimental settings and may prove to add prognostic value on the degree of inflammation present and may be beneficial in determining treatment course in the near future.

Imaging Studies

See the list below:

  • Radiologic evaluation is the primary tool for the diagnosis of a parapneumonic effusion. Standard chest radiography is the first step to assess for pleural fluid. See the image below.

    Radiographic imaging of a parapneumonic effusion m Radiographic imaging of a parapneumonic effusion may be useful in assessing the stage of the effusion and the type of drainage needed. In Figure A, the left heart border is obscured, and more than 50% of the left hemithorax is filled with an effusion, as evidenced by a fluid meniscus. In Figure B, the effusion is demonstrated to be fluid because it layers out on a decubitus film. In Figure C, the lateral radiograph again demonstrates the fluid meniscus and filling of the posterior sulcus. These findings suggest tube thoracostomy placement may be sufficient to drain this pleural process.
  • Examination should include upright views of the chest to examine the diaphragmatic margins, which are obliterated with pleural fluid collections. Because as much as 400 mL may be required before these costophrenic angles are obscured in older children and adolescents, further diagnostic imaging may be needed.

  • In cases in which the effusion is moderate, radiography may reveal displacement of the mediastinum to the contralateral hemithorax, as well as scoliosis.

  • Indistinct diaphragmatic contours merit lateral decubitus views of the chest.

  • Free-flowing pleural effusions suggest less complicated parapneumonic processes, which may not require extensive diagnostic and therapeutic interventions. Ultrasonography that reveals the absence of loculations suggests that effective treatment can be achieved without surgical intervention. The absence of free layering fluid on the decubitus films does not exclude the possibility of a loculated pleural effusion.[7]

  • Consider ultrasonography or CT imaging to identify the presence of consolidated lung or fibrinous septations. In patients with complex fluid collections, chest CT imaging has emerged as the study of choice. Chest CT imaging can be used to detect and define pleural fluid and image the airways, guide interventional procedures, and discriminate between pleural fluid and chest consolidation.

Other Tests

See the list below:

  • Pleural fluid polymerase chain reaction (PCR), latex agglutination (or counter immunoelectrophoresis [CIE] for specific bacteria) may be helpful if the cause of the infection cannot be ascertained from stain or culture results.


See the list below:

  • Thoracentesis can provide both significant diagnostic information and therapeutic relief for parapneumonic effusions.

  • The presence of pus establishes the presence of an empyema, and the determination of the Gram stain, cell differential, and chemistries helps to guide therapy.

  • Performing thoracentesis before the initiation of antibiotics increases the diagnostic yield of the fluid cultures and allows for more specific antimicrobial therapy.



Medical Care

Treatment of parapneumonic effusions should address control of the infection and often involves drainage of the pleural fluid and reexpansion of the affected lung tissue.

  • Appropriate antibiotic selection should be based on the Gram stain and culture of the pleural fluid; however, because a large number of patients may have already received antibiotics at the time of thoracentesis, an empiric selection of the most appropriate antibiotics is necessary.

    • Base the choice on the most common pathogens that cause pneumonia within the patient's age range and geographic location.

    • When the organism is identified, change the antibiotics to most specifically cover for the pathogen.

    • Patients with empyema should receive a longer course of therapy analogous to necrotizing pneumonia, but the response to therapy determines the duration of treatment. The patient receives 10-14 days of intravenous antibiotics and receives treatment until he or she is afebrile, off supplemental oxygen, and appropriately responds to therapy.

    • Continuation of oral antibiotics may be recommended for 1-3 weeks after discharge but is not required for less complicated infections.

  • The most controversial area in the management of parapneumonic effusions is the identification of patients who would benefit from pleural drainage and the selection of the appropriate drainage intervention.

    • No clinical studies have effectively contrasted antibiotic treatment without drainage to currently available drainage techniques. However, long-term follow-up studies show no differences in pulmonary function or exercise capacity between the groups. The therapeutic discussion rests on available clinical, radiologic, and laboratory evidence; host factors; and individualization to make the appropriate decision.

    • The pulmonologist, intensivist, interventional radiologist, or surgeon can perform simple tube thoracostomy with an underwater seal.

    • Diagnostic thoracentesis and chest tube drainage are effective therapies in more than 50% of patients. Prompt drainage of a free-flowing effusion prevents the development of loculations and a fibrous peel.

    • Remove the tube when the lung re-expands and drainage ceases. If the fluid is not free flowing, undertake further radiologic imaging to better define the pleural space disorder.

  • Clinical resolution is not hastened by chest physical therapy used as an adjunct to standard treatment in children hospitalized with acute pneumonia.[8]

  • In addition to the benefit of CT and ultrasonographic imaging to characterize loculated pleural effusions, the radiologist has become significantly involved in the treatment of complicated parapneumonic effusions (CPE).

    • The ability of the interventional radiologist to assist in the placement of small-bore catheters, specifically localized to loculated pleural fluid collections, has helped to facilitate drainage. Furthermore, with smaller diameter tubes, patients have tolerated tube placement better, with less associated morbidity.

    • In addition, radiologists can lyse adhesions directly using imaging during the tube placement.

    • Finally, interventional radiologists, using fibrinolytics, have further improved the care of the complicated empyema by improved management of loculations and amelioration of fibrous peel formation and fibrin deposition.

  • Numerous studies have documented the effectiveness of intrapleural fibrinolytics to treat obstructed thoracostomy tubes, increase drainage in multiloculated effusions, and to lyse adhesions; however, initial studies report on the use of urokinase, the fibrinolytic most commonly described prior to 1998, evolving to the use of tissue plasminogen activator (tPA), which has become the most frequently used treatment. Increased use of tPA has shown it to be well tolerated, effective, and less costly than surgery.[9, 10] Randomized trials of chest tube drainage with fibrinolytics versus simple drainage and surgical therapy need to be undertaken to fully assess the relative clinical value of fibrinolytic therapy in the more complicated patient.

Surgical Care

Controversy continues regarding the surgical treatment of CPE.

  • For the uncomplicated free-flowing parapneumonic effusion, surgical intervention is rarely needed; however, the multiloculated persistently symptomatic effusion, for which initial therapy may have been delayed, is likely to require more than conservative management.

  • The surgical literature supports the use of thoracotomy to remove the pleural peel and lyse the adhesions, if the patient does not respond promptly to treatment. Length of stay and long-term morbidity are reduced by this more aggressive approach, but this must be contrasted with the increased cost and short-term morbidity associated with thoracotomy and decortication. This treatment regimen is very effective, with a reported 95% success rate for patients with fibrinopurulent effusions. Because no prospective comparative studies have contrasted the current techniques, decortication is likely to remain a treatment of choice for advanced empyema.

  • Video-assisted thoracoscopic surgery (VATS) has proven to be an effective and less-invasive replacement for the limited decortication procedure.[11, 12] Thoracoscopic debridement closely imitates open thoracotomy and drainage. Mechanical removal of purulent material and the breakdown of adhesions can be easily accomplished via this route. VATS results in more rapid relief of symptoms, earlier hospital discharge, and significantly less discomfort and morbidity.

  • Despite the benefits, a small percentage of patients still progress to require thoracotomy. As with fibrinolytic therapy, those patients in which this therapy has been most effective are those slightly less affected in whom earlier and potentially more aggressive treatment has been initiated.

  • The definitive approach is thoracotomy drainage with mechanical release of the pleural peel and lysis of adhesions. Studies of decortication and debridement report 95% effectiveness for empyemas in the fibrinopurulent stage. These outcomes are determined by selected clinical outcomes, such as resolution of symptoms; however, these studies are subject to selection bias and do not account for the morbidity associated with the procedure, as well as increased costs associated with an operative procedure and the associated anesthesia risk. Furthermore, most children heal well in the long run, even without immediate surgical intervention.

  • A meta-analysis reviewed the differences between primary operative treatment with nonoperative management and revealed striking reductions in length of stay, duration of tube thoracostomy, and duration of antibiotics.[13] However, these data are susceptible to critics who point out the same concerns regarding selection biases that have been listed above. An additional review of surgical options demonstrated that early VATS resulted in shortening hospitalization times, but other outcomes such as duration of symptoms or antibiotic use were less dramatic between surgical treatments. In conclusion, early intervention of any sort is likely to improve outcomes, but early VATS is the surgical approach now most preferred in managing children with empyema.

  • Alternative procedures, such as rib resection and open drainage or pleural obliteration, are rarely needed in the pediatric population.

  • To most effectively determine the optimal therapeutic intervention, a carefully designed, multicentered, randomized, clinical trial would help to develop evidence-based standards for the treatment of complicated parapneumonic effusion in children.

  • Marhuenda et al conducted a prospective, randomized, multicenter clinical trial comparing the efficacy of drainage plus urokinase therapy with that of video-assisted thoracoscopic surgery in the treatment of pediatric parapneumonic empyema.[14] The study included a total of 103 patients (age, younger than 15 years); 53 patients were randomly assigned to receive treatment with thoracoscopy, and 50 were assigned to treatment with urokinase. There were no differences in demographic characteristics or in the main baseline characteristics between the 2 groups. No statistically significant differences were found between the thoracoscopy group and the urokinase group in the median postoperative stay (10 vs 9 days), the median hospital stay (14 vs 13 days), or the number of days febrile after treatment (4 vs 6 days). A second intervention was required in 15% of the children in the thoracoscopy group and in 10% in the urokinase group. The investigators concluded that drainage plus therapy with urokinase is as effective as video-assisted thoracoscopic surgery as first-line treatment of septated parapneumonic empyema in children.[14]

  • Proesmans et al reported on the use of a standardized medical treatment of parapneumonic empyema as a first-step nonsurgical approach in a tertiary care center.[15] The purpose of the study was to evaluate the need for surgery and to collect data on disease course, outcome, and microbiology. The study cohort consisted of 132 children treated for parapneumonic empyema between 2006 and 2013. Of the 132 children, 20% required surgical intervention. Median duration of in-hospital fever was 5 days. The duration of fever correlated with pleural LDH levels and pleural glucose levels and was inversely correlated with pleural pH. On the basis of pleural polymerase chain reaction results, 85% of the cases were caused by Streptococcus pneumoniae (40% were of serotype 1). After introduction of a standardized primary medical approach (chest drainage with or without fibrinolysis), the need for surgical rescue interventions remained at 20%overall.[15]


See the list below:

  • Consultants may include pediatric surgeons (thoracic or general), interventional radiologists, intensivists, and pulmonologists.


Patients should perform as much activity as can be tolerated.

  • Encourage the facilitation of deep breathing and airway clearance.

  • Use analgesics on an individual basis to facilitate airway clearance.



Medication Summary

Individualize intravenous antibiotics by both age and likelihood of the offending organism. In the case of a possible aspiration, such as in the patient who is debilitated or neurologically impaired, consider coverage for anaerobic infection.

Antibiotic, cephalosporin (second generation)

Class Summary

These agents are recommended for the most likely bacterial infections; the specific agent selected should be individualized based on the offending organism. They may include, but are not limited to, the following:

Cefuroxime (Zinacef)

A second-generation cephalosporin with good coverage for most staphylococcal and streptococcal organisms, which are the most common community-acquired causative agents; thus, this is the most often selected initial antibiotic.

Antibiotics, anaerobic infections

Class Summary

In situations in which an aspiration or likely anaerobic infection is the cause of the pneumonia, coverage for anaerobes is recommended.

Clindamycin (Cleocin)

Provides coverage for gram-positive organisms and anaerobes and is a possible agent for infections in patients at high risk for having aspirated PO contents as a cause of their infection.

Antibiotic, Miscellaneous

Class Summary

Vancomycin may be considered when methicillin-resistant S aureus is suspected or confirmed.

Vancomycin (Vancocin, Vancoled)

Classified as glycopeptide agent that has excellent gram-positive coverage, including methicillin-resistant S aureus. To avoid toxicity, current recommendation is to assay vancomycin trough levels after third dose drawn 0.5 h prior to next dosing. Use creatinine clearance to adjust dose in patients diagnosed with renal impairment.

Thrombolytic agents

Class Summary

These agents convert plasminogen to plasmin, leading to clot lysis. These agents are used to lyse adhesions in the pleural space.

Alteplase (Activase)

Tissue plasminogen activator exerts effect on fibrinolytic system to convert plasminogen to plasmin. Binds to fibrin in a thrombus and converts the entrapped plasminogen to plasmin, thereby initiating local fibrinolysis. Serum half-life is 4-6 min, but half-life is lengthened when bound to fibrin in clot. Used to restore function of central venous access devices that have become occluded due to thrombosis. Circulating plasma levels are not expected to reach pharmacologic concentrations.



Further Outpatient Care

See the list below:

  • Obtain follow-up radiographs and pulmonary function tests to determine prognosis of patients with empyema and to confirm resolution of pleural and parenchymal changes.

  • Consider a follow-up chest CT scan after the radiography findings clear.

Inpatient & Outpatient Medications

See the list below:

  • See Medication.


See the list below:

  • Children should receive their care in hospitals equipped to deal with ill children and staffed with the appropriate pediatric subspecialists.


See the list below:

  • Fibrothorax, a complication reported in the adult literature, is rarely observed in pediatric patients.


See the list below:

  • The prognosis for most patients with parapneumonic effusions is quite good.

  • Extended antibiotics may be needed in some patients with complicated parapneumonic effusions (CPE).

  • Despite the variability in presentation, most patients recover without sequelae.

  • Numerous studies have demonstrated resolution of the radiographic abnormalities by 3-6 months following therapy, with few to no symptoms reported at follow-up examination.

  • Pulmonary function testing performed following hospitalization has not shown marked abnormalities, regardless of clinical course. The only abnormality observed may be slight expiratory flow limitation. Mild obstructive abnormalities were the only findings observed in patients evaluated 12 years (±5) following recovery from empyema.

  • Some increased bronchial reactivity has been reported at later follow-up examinations; however, lung function and exercise response return to normal for most patients.

  • Early recognition of pneumonia with parapneumonic effusion, effective intervention to identify the causative organism, and initiation of definitive therapy reduce morbidity and complications associated with this process.

Patient Education

See the list below:

  • For excellent patient education resources, visit eMedicineHealth's Lung Disease and Respiratory Health Center. Also, see eMedicineHealth's patient education article Bacterial Pneumonia.