Epidemiology, clinical presentation, and diagnostic evaluation of parapneumonic effusion and empyema in adults

INTRODUCTION — A parapneumonic effusion is a pleural effusion that forms in the pleural space adjacent to a pneumonia. When microorganisms infect the pleural space, a complicated parapneumonic effusion or empyema may result. An empyema can also develop in the absence of an adjacent pneumonia.

The epidemiology, microbiology, clinical presentation, and diagnostic evaluation of parapneumonic effusions and empyema are reviewed here. The management of parapneumonic effusions and empyema in adults and children is discussed separately. (See "Management and prognosis of parapneumonic pleural effusion and empyema in adults" and "Epidemiology, clinical presentation, and evaluation of parapneumonic effusion and empyema in children" and "Management and prognosis of parapneumonic effusion and empyema in children".)

DEFINITIONS — A parapneumonic effusion refers to the accumulation of fluid in the pleural space in the setting of an adjacent pneumonia (table 1).

●An uncomplicated or simple parapneumonic effusion refers to a free-flowing effusion that is sterile.

●A complicated parapneumonic effusion refers to an effusion that has been infected with bacteria or other microorganisms (eg, positive Gram stain or biochemical evidence of marked inflammation).

●An empyema refers to a collection of pus within the pleural space, which can develop when pyogenic bacteria, fungi, parasites, or mycobacteria invade the pleural space, either from an adjacent pneumonia or from direct inoculation (eg, from penetrating trauma) or other source. Empyema that develops from an adjacent pneumonia is a subclass of a complicated parapneumonic effusion.

●A complex effusion refers to an effusion with internal loculations.

●A uniloculated effusion is one that is without internal septae (free-flowing or fixed).

EPIDEMIOLOGY

Incidence — Parapneumonic effusions and empyema are relatively common complications of pneumonia. Since the advent of antibiotics, their overall incidence has declined dramatically to approximately two to three percent of all pneumonias [1]. However, epidemiologic studies suggest that rates are again slowly rising [2-5]. As examples, one study of 4424 patients with empyema reported an increase in the incidence by 2.8 percent per year between the years of 1987 and 2004 [6]. Similarly, another study of over 11,000 patients admitted to hospital with empyema found a 2.2-fold rise in the incidence of pleural infection between 1995 and 2003 in patients <19 years and a 1.2-fold rise in patients >19 years of age [2]. Continued increases have been noted from 2008 to 2018 in the United Kingdom, particularly in those aged >60 years, with a seasonal variability associated with influenza infections [7].

Rates are highest in hospitalized patients with one report suggesting that 20 to 40 percent of patients hospitalized with pneumonia have a parapneumonic effusion and 5 to 10 percent of those progress to empyema (ie, about 32,000 patients per year) [3].

Differences in reporting rates between studies may reflect differences in how studies define parapneumonic effusion; studies evaluating the incidence of computed tomography (CT)-defined fluid in pneumonia report higher rates than those that are defined by fluid chemistry and culture.

Empyema may be more common in men than women, although the reasons for this are unknown [8-10].

Risk factors — Other than risk factors for the development of pneumonia, commonly cited risk factors for the development of a parapneumonic effusion include aspiration, poor dental hygiene, malnutrition, and alcohol or intravenous drug abuse [11-15]. Others include immunosuppression, age (<18 years, >65 years), partially treated pneumonia, influenza [7], and gastroesophageal reflux. (See "Overview of community-acquired pneumonia in adults", section on 'Risk factors'.)

Prior use of inhaled glucocorticoids for chronic obstructive pulmonary disease (COPD) or asthma has been associated with reduced incidence of parapneumonic effusions [16], a surprising outcome given the increased incidence of community acquired pneumonia (CAP) in these patients. The reason for this inverse association is unknown but may be due to an altered inflammatory response from inhaled glucocorticoids or due to a lower threshold to seek medical attention in patients with asthma or COPD.

The presence of preexisting pleural fluid (eg, secondary to heart failure, liver disease) also favors growth of microorganism in the pleural space and is likely a contributing factor in the risk for developing pleural space infections.

PATHOGENESIS — Most parapneumonic effusions and empyemas are due to underlying pneumonia, and are believed to develop in three stages (figure 1). Patients can present at any stage of development (table 2).

Stage 1 (simple or uncomplicated parapneumonic effusion) — Stage 1 is an early stage of development when interstitial fluid increases during pneumonia and moves across the adjacent visceral pleural membrane. In this stage, fluid is free-flowing and has exudative characteristics with a protein content greater than 0.5 of the serum value and/or a lactate dehydrogenase (LDH) level more than 0.6 than in the serum (but usually <1000 international units [IU]/L). The white cell count is variable but typically has neutrophilic predominance. Fluid will have a normal pH and glucose level and no evidence to support infection with microorganisms. (See 'Thoracentesis and pleural fluid analysis' below.)

Stage 2 (complicated parapneumonic effusion and empyema) — Stage 2 is a fibrinopurulent stage, whereby bacterial invasion across the damaged pleural mesothelium stimulates an inflammatory response resulting in fibrin deposition and loculations within the pleural space. Characteristic findings include an exudative effusion with a high white cell count, pH <7.20, glucose <2.2 mmol/L (<40 mg/dL) and LDH >1000 IU/L. Without pus, this is termed a complicated parapneumonic effusion, but if frank pus is found, then this is called an empyema. Notably, it is possible that fluid characteristics may vary among pockets of fluid/loculations. (See 'Thoracentesis and pleural fluid analysis' below.)

Empyema can also present independently of pneumonia, when bacteria or other pathogens directly inoculate or invade the pleural space (figure 1). In these cases, empyema can be due to esophageal rupture, blunt or penetrating chest trauma, hematogenous spread, mediastinitis with pleural extension, bronchogenic carcinoma that has breached the pleura allowing bacterial translocation, infected congenital cysts of the airway or esophagus, or extension from sources outside the thorax (eg, liver abscess or cervical or thoracic spine infections). Postsurgical etiologies (eg, bronchopleural fistula from lobectomy) can also be independent of pneumonia.

Stage 3 (chronic organization) — In stage 3, the pleural fluid begins to organize [17]. In the later stages, a fibrinous pleural covering ("fibrous peel") develops and may encase the lung, hindering full re-expansion (trapped lung), impairing lung function, and creating the potential for additional infection. In many cases, pleural fluid has been organized and cannot be withdrawn for analysis. This thickened pleura usually resolves over three to six months but in some cases forms a true scar. (See "Diagnosis and management of pleural causes of nonexpandable lung".)

MICROBIOLOGY — Pyogenic bacteria such as Streptococcus pneumoniae, oral streptococci and anaerobes, and Staphylococcus aureus are the most common causes of parapneumonic effusions and parapneumonic empyema (table 3) [15,18]. Parapneumonic pleural space infections that result from community acquired pneumonia (CAP) tend to be caused by the pyogenic bacteria that cause CAP (eg, S. pneumoniae, S. aureus). Parapneumonic pleural space infections that result from aspiration tend to be polymicrobial and caused by oral streptococci and anaerobes. However, the spectrum of potential pathogens is wide and varies based on the route of acquisition (eg, parapneumonic versus nonparapneumonic), site of acquisition (eg, community or hospital-acquired) and geography (figure 1) [15,18,19].

For nonparapneumonic complicated pleural effusions and empyema, the list of causative organisms is more extensive and varies considerably with the source (figure 2). For example, empyema that results from diaphragmatic translocation from an intraabdominal infection is likely caused by gastrointestinal flora. Thus, the infection source, local epidemiology, and patient-specific risk factors are important when evaluating a patient with nonparapneumonic empyema.

It is prudent that the clinician be aware of their local microbiology and antibiotic resistance patterns and be cognizant of the likelihood of infections that are polymicrobial (eg, coexisting anaerobes).

Bacteria — S. pneumoniae (pneumococcus) is the most common bacterial cause of CAP, and among the most common causes of community-acquired parapneumonic effusions and empyema. The reported prevalence varies among studies, ranging from approximately 10 to 20 percent of culture-positive cases [15,20-22]. (See "Pneumococcal pneumonia in patients requiring hospitalization", section on 'Pleural effusion and empyema'.)

Microaerophilic streptococci (eg, S. anginosus, S. intermedius, S. constellatus) and anaerobes, which colonize the oropharynx, are leading causes of empyema, and presumably reach the pleural space by spread from aspiration pneumonia [4,15,21,23]. Empyema related to aspiration is often polymicrobial. Among anaerobes the most frequently isolated organisms include Fusobacterium species (eg, F. nucleatum), Prevotella species, Peptostreptococcus species, and Bacteroides species (eg, B. melaninogenicus; B. fragilis) [9,11,24-26]. Anaerobic bacteria have been cultured in 36 to 76 percent of human empyemas [24,27]. (See "Aspiration pneumonia in adults", section on 'Microbiology' and "Anaerobic bacterial infections" and "Anaerobic bacterial infections", section on 'Pleuropulmonary infections'.)

Staphylococcus aureus accounts for approximately 10 to 15 percent of parapneumonic effusions and empyema overall [15,18,21]. The prevalence of S. aureus is highest among hospital-acquired empyemas (accounting for approximately one-third of cases), but S. aureus empyema can also occur secondary to CAP. Gram-negative rods, including Escherichia coli, other Enterobacteriaceae, Pseudomonas aeruginosa, and Klebsiella species, collectively account for approximately 8 to 10 percent of cases [15,21].

Atypical bacteria, such as Legionella, Mycoplasma, and Chlamydia species, are rare but reported causes of parapneumonic effusions and empyema. (See "Clinical manifestations and diagnosis of Legionella infection" and "Mycoplasma pneumoniae infection in adults" and "Mycoplasma pneumoniae infection in adults", section on 'Pneumonia'.)

The relative prevalence of pathogens also varies with geography. As an example, Burkholderia pseudomallei is endemic in Southeast Asia, South Asia, and northern Australia. In areas of high prevalence, such as Thailand, B. pseudomallei is a common cause of pneumonia, and has been reported to cause up to 20 percent of pleural infections [28]. (See "Melioidosis: Epidemiology, clinical manifestations, and diagnosis".)

Selected patients may also have increased risk of specific organisms. Patients with diabetes mellitus are at increased risk of empyema secondary to Klebsiella pneumoniae [29]. In patients with influenza, the major causes of bacterial superinfection and empyema have been Staphylococcus aureus, Streptococcus pneumoniae, and Streptococcus pyogenes [30]. Anaerobic empyema with aspiration pneumonia often is seen in the aspiration-prone patient who presents relatively late in the infection with pneumonitis involving the superior segment of a lower lobe or posterior segment or an upper lobe.

Mycobacteria — Mycobacterial effusion and empyema are considerably less frequent than bacterial empyema, and usually result from reactivation of latent tuberculosis (TB) in the subjacent lung or pleural space. Tuberculous empyema should be considered in patients who live in an endemic area and/or who have risk factors for TB. [31]. Tuberculous empyema, in which the TB organism can be found by stain or by culture in the pleural effusion, should be differentiated from tuberculous pleurisy in which a lymphocytic effusion occurs from the immunologic response to tuberculous proteins and the TB organisms are few and only reliably found in the pleural tissue. Further details regarding the manifestations of tuberculous empyema are discussed separately. (See "Tuberculous pleural effusion".)

Less commonly, atypical mycobacteria are also associated with parapneumonic effusion and empyema, (eg, Mycobacterium abscessus, Mycobacterium avium, Mycobacterium kansasii) [32]. (See "Overview of nontuberculous mycobacterial infections".)

Other pathogens — Fungal pleural infection is rare (<1 percent of cases) with Candida species being responsible for the majority of cases [33]. Candida pneumonia and empyema typically occur in the setting of disseminated infections in highly immunocompromised patients or, in the case of empyema, as a complication of thoracic surgery [33]. Aspiration of Candida from the oropharynx is unlikely to cause either pneumonia or empyema. (See "Candida infections of the abdomen and thorax".)

Less commonly reported fungal pleural infections are caused by cryptococcus and Aspergillus species, both of which most often occur in immunocompromised hosts [33,34].

Parasites are rare causes of pleural infections, but Entamoeba histolytica, Echinococcus granulosus and Paragonimus westermani can cause pleural effusions [35,36].

Viruses do not typically cause empyema. For example, no influenza viruses were detectable by polymerase chain reaction in any pleural fluid samples in a prospective study evaluating pleural space infections following influenza [7]. However, secondary bacterial infections can complicate viral pneumonias. As an example, in patients with influenza, the major causes of bacterial superinfection and empyema have been Staphylococcus aureus, Streptococcus pneumoniae, and Streptococcus pyogenes [30].

CLINICAL FEATURES — The clinical findings of a parapneumonic effusion and empyema are nonspecific and, with the exception of decreased fremitus, overlap with those of pneumonia. Among those with pneumonia, risk factors for empyema (eg, aspiration), and persistent or new fever, and/or lack of clinical response despite appropriate antibiotics should heighten the suspicion for the development of a parapneumonic effusion or an empyema.

History and examination — Common clinical features on history include cough, fever, pleuritic chest pain, dyspnea, and sputum production. Compared with those with pneumonia alone or pneumonia with simple parapneumonic effusion, patients with empyema may report a longer course with several days of fever and malaise, with one trial reporting duration of symptoms as long as two weeks [8]. Presentation may also be more insidious and delayed in patients with anaerobic infections (eg, those with aspiration, poor dental hygiene) with some patients presenting with loss of appetite and weight loss over weeks to months [11-13].

Physical examination may identify the presence of pleural fluid with dullness on percussion, decreased breath sounds, and decreased fremitus. Occasionally, egophony (e-to-a sound) is present at the upper edge of the effusion. Although decreased vocal fremitus classically differentiates a pleural effusion from lung consolidation (associated with increased vocal fremitus), these findings are often absent and therefore, not useful. Thus, radiographic imaging is crucial to the complete evaluation. (See 'Diagnostic imaging' below.)

Laboratory findings — There are no specific laboratory blood tests that are diagnostic of a parapneumonic effusion. Laboratory findings usually reflect that of infection such as leukocytosis, left shift, elevated C-reactive protein. In some cases, bacteremia can co-occur and the infecting organism can be identified from blood cultures (up to 12 percent of cases) [8]. (See "Clinical evaluation and diagnostic testing for community-acquired pneumonia in adults".)

DIAGNOSTIC EVALUATION — Our approach is generally similar to that outlined by the American Association for Thoracic Surgery (AATS), the European Association for Cardio-Thoracic Surgery (EACTS), the American College of Chest Physician (ACCP), the British Thoracic Society (BTS) pleural disease guideline development group, and the European Respiratory Society/European Society of Thoracic Surgeons group (algorithm 1) [37-41].

Diagnostic imaging — Chest radiography, ultrasonography, and CT all play a key role in the evaluation and management of parapneumonic effusions and empyema [42]. In all patients with pneumonia in whom recent imaging has not been obtained, the chest radiograph should be the initial imaging modality obtained for evidence of pleural fluid. Once a pleural effusion is detected on chest radiography, ultrasonography is typically performed at the bedside to evaluate the nature of the effusion and feasibility of sampling or drainage. CT is generally performed when complications (eg, loculations) are suspected, complex interventions are planned, and/or more detail of the underlying anatomy is expected to help with management.

All three imaging modalities have their advantages and disadvantages and have been inadequately compared. However, one retrospective analysis of 66 patients suggested that ultrasonography was more sensitive than chest radiography (69 versus 61 percent) but less sensitive than CT (69 versus 76 percent) for the diagnosis of a complicated parapneumonic effusion [43]. (See "Imaging of pleural effusions in adults".)

Magnetic resonance imaging (MRI) and positron emission tomographic scanning are not useful, although MRI may be valuable when chest wall or spinal involvement is suspected [44]. Their role in evaluating effusions due to malignancy is discussed separately. (See "Imaging of pleural plaques, thickening, and tumors" and "Overview of the initial evaluation, diagnosis, and staging of patients with suspected lung cancer", section on 'Clinical-directed imaging'.)

Rarely, additional imaging may be required for those with empyema not associated with pneumonia, including contrast imaging for the esophagus when a ruptured esophagus is suspected, or CT of the abdomen when empyema due to an intraabdominal process (eg, liver abscess) is suspected (figure 2).

Chest radiography — Most parapneumonic effusions or empyemas are initially suspected during the routine evaluation of patients with suspected or known pneumonia.

Free-flowing pleural effusions accumulate in the most dependent part of the thoracic cavity [45]. In an upright patient, some free-flowing effusions are subtle lying in a subpulmonic location (<75 mL). Other pleural effusions can be appreciated on lateral chest radiography as blunting of the posterior costophrenic angle (>75 mL) or on anteroposterior chest radiography as blunting of the lateral costophrenic angle (>175 mL), while large effusions may obscure the diaphragm (>500 mL) and demonstrate a meniscus sign. Occasionally, the entire hemithorax may be occupied by an effusion with associated underlying lung collapse.

However, chest radiography has limitations. The meniscus sign and dependent layering of fluid is often absent in complex/loculated parapneumonic effusions where the radiograph may show a lenticular pleural-based opacity (image 1). In addition, large effusions may hide underlying pneumonia and large consolidations may hide small effusions; in both situations the threshold to obtain an ultrasound and/or chest CT should be low. Demonstrating such limitations, in a study of 61 patients with chest CT-proven parapneumonic effusions, anteroposterior and lateral chest radiography missed approximately 10 percent of pleural effusions, in most cases due to the coexistence of lower lobe consolidation [46]. Additional details regarding radiographic characteristics of pleural effusions are discussed separately. (See "Imaging of pleural effusions in adults", section on 'Conventional radiography'.)

In the past, lateral decubitus radiographs were often performed to determine the extent to which an effusion is freely flowing within the pleural space and evaluate the safety of thoracentesis. As the availability of ultrasonography and CT advances, this test is rarely performed. (See 'Thoracentesis and pleural fluid analysis' below.)

Ultrasonography — Free-flowing or loculated pleural effusions as well as underlying consolidation or solid masses can be appreciated on chest ultrasonography. Ultrasonography is now typically performed at the bedside for most pleural effusions to evaluate the size and characteristics of the effusions. Although ultrasonography is best utilized for selecting and guiding needle or catheter placement for thoracentesis, it also provides important prognostic and therapeutic information [47]. Data that support the use of pleural ultrasonography are discussed separately. (See "Bedside pleural ultrasonography: Equipment, technique, and the identification of pleural effusion and pneumothorax", section on 'Identification of pleural effusion using ultrasonography' and "Ultrasound-guided thoracentesis".)

Chest computed tomography — Chest CT (typically with intravenous contrast to help identify the pleural membranes) is the most sensitive method for detecting small amounts of pleural fluid (>2 mL). CT can also easily appreciate loculations (image 2) and underlying airway, pleural, and parenchymal abnormalities (eg, pneumonia, abscess, fistulae, diaphragmatic defects, esophageal rupture, or masses that suggest underlying cancer) as well as chest tube positioning [48-50].

During the fibrinopurulent and organizing stages of complicated parapneumonic effusions and empyema (see 'Pathogenesis' above), radiographic contrast enhancement of the pleural surfaces assists in delineating pleural fluid loculations and in characterizing empyema. As examples:

●Loculations, often lenticular in shape, may be appreciated by tapered borders and obtuse angles between the fluid and the chest wall with an absence of the meniscus sign.

●Parietal pleural thickening is seen in 86 percent and pleural enhancement in 96 percent of patients with empyema [49]. Thickening of the visceral and parietal pleura is suggestive of empyema when associated with significant (usually >30 mm) separation of the pleural surfaces (split pleura sign), which can be appreciated in up to 68 percent of patients [49,51].

●Pleural infection is also associated with increased attenuation of extracostal fat [49].

Air within the pleural fluid (eg, pockets or bubbles of air, gas-liquid level) may suggest associated pneumothorax, bronchopleural fistula, air introduced during thoracentesis, a nonexpandable lung after pleural drainage or rarely gas-producing anaerobic organisms.

Older empyemas that have spontaneously organized and remained undetected over years may exhibit minor or major calcification (eg, tuberculous empyema).

Distinguishing pleural fluid from pleural masses and distinguishing empyema from a lung abscess are discussed below. (See 'Differential diagnosis' below.)

Thoracentesis and pleural fluid analysis — Most cases of suspected parapneumonic effusion or empyema should at minimum be sampled unless the effusion is too small or sampling is deemed unsafe [52]. Results guide further management of the effusion. Diagnostic thoracentesis is typically performed under ultrasound guidance. In some cases, therapeutic thoracentesis is simultaneously performed when symptomatic relief is needed or when drainage is indicated (eg, frank pus is observed). Indications and contraindications and technique for thoracentesis are discussed separately. (See "Ultrasound-guided thoracentesis" and "Bedside pleural ultrasonography: Equipment, technique, and the identification of pleural effusion and pneumothorax".)

In the past, it was considered safe to sample pleural fluid when a free-flowing effusion was demonstrated on chest radiography with at least 1 cm depth to the chest wall on a lateral decubitus film [53]. However, most pleural effusions are now sampled under ultrasound guidance; since there is no definition of what is considered a "safe" amount for sampling by ultrasonography, much of this decision is at the discretion of the ultrasonographer, although most experts would consider a pleural space of >1 cm (between parietal and visceral pleural) safe. Occasionally, CT guidance may be needed for sampling fluid, particularly fluid that is located in small loculated pockets within the pleural space; in such cases, the largest and most accessible loculation is generally chosen. A pleural fluid thickness cutoff of 2 to 2.5 cm has been suggested to guide thoracentesis by chest CT, because smaller effusions on CT are likely to resolve with antibiotics alone [54,55].

Fluid obtained by thoracentesis should be sent for the following:

●Cell count with differential and chemistries – A total protein, lactate dehydrogenase (LDH), and glucose should be obtained together with serum values for protein and LDH. A neutrophil predominance is more common in patients with bacterial pneumonia while a lymphocyte predominance may indicate tuberculous or fungal etiologies. Distinguishing exudates from transudates is discussed separately. (See "Pleural fluid analysis in adults with a pleural effusion", section on 'Routine pleural fluid biomarkers'.)

●Microbiologic analysis – Microbiologic analysis of the pleural fluid with appropriate stains and cultures (eg, aerobic, anaerobic, mycobacterial, fungal) is critical. Although sampling should ideally occur before the administration of antibiotics, thoracentesis should not delay prompt antimicrobial therapy. Samples should be drawn directly from the pleural space because cultures from previously placed catheters or tubing can be colonized or contaminated with bacteria or fungi [56].

Pleural fluid should be inoculated directly into blood culture bottles (aerobic and anaerobic) in addition to the usual sterile containers used for standard Gram stain and culture, in order to maximize diagnostic yield [57,58]. In a series of 53 patients with suspected pleural infection and culture data from both standard and blood culture bottles obtained at the same time, the number of patients with identifiable pathogens increased by 21 percent (95% CI 8.9 to 20.8 percent) [55]. A sterile container without culture media is acceptable if only a small amount of fluid is available.

A putrid odor of fluid is considered diagnostic of anaerobic infection; the Gram stain will also help identify anaerobes because of the unique morphology of some anaerobic Gram-negative rods. (See "Anaerobic bacterial infections" and "Anaerobic bacterial infections", section on 'Pleuropulmonary infections'.)

Blood and pleural culture results yield a diagnosis in approximately 60 percent of cases [21]. There are several reasons why bacteria may not be identified in culture for the remaining 40 percent:

•Anaerobic organisms may be difficult to culture

•Anaerobic cultures are not specifically requested

•Sampling is often performed after a patient has received antibiotics

•Sterile inflammatory fluid may be aspirated adjacent to an infected loculus of infection

•Current culture methods are insufficiently sensitive

•Bacteria may be located in the pleural membranes rather than in the fluid [59]

Some centers have begun routine molecular analysis of parapneumonic effusions to detect S. pneumoniae infection by rapid antigen detection assays or broad-range 16S ribosomal DNA polymerase chain reaction. These centers report a much higher detection rate for S. pneumoniae than historical case series [60].

Special stains and cultures should be requested when unusual organisms or organisms that require special culture conditions are suspected.

●pH – The pH should be drawn directly into an arterial blood gas syringe and determined with a blood gas analyzer within one hour of sampling. Residual lidocaine and heparin falsely decrease the pH and air in the syringe falsely increases the pH; therefore, the same needle that is used to anesthetize the pleural space should not collect the pH sample. It is not necessary to run the sample through an analyzer if frank pus is collected since that feature is diagnostic of an empyema. (See "Pleural fluid analysis in adults with a pleural effusion".)

The pH is the most useful test when determining the therapeutic course, the details of which are discussed separately. (See 'Complicated parapneumonic effusion and empyema' below and "Management and prognosis of parapneumonic pleural effusion and empyema in adults".)

The differential diagnosis of pleural fluid acidosis, a feature typically associated with a complicated parapneumonic effusion or empyema is discussed below. (See 'Differential diagnosis' below.)

●Cytology – Cytologic examination for appropriate stains (eg, mycobacteria, actinomyces, nocardia) can be sent when organisms requiring special stains are needed. In addition, malignancy can cause pleural fluid acidosis; thus, sending fluid for cytology for malignant cells is prudent.

Some studies have demonstrated that pleural fluid analysis may substantially differ from one locule to another, limiting the value of pleural fluid analysis in this instance [61]. Thus, in cases where the results are inconsistent with clinical findings, repeat sampling should be considered.

Novel biomarkers of infection (eg, C-reactive protein, procalcitonin, STREM-1) in pleural fluid have been evaluated for possible utility in distinguishing empyemas from uncomplicated pleural effusions, but were found to be no more useful than the more traditional pleural chemistries [62-66]. Another retrospective study showed that higher levels of pleural vascular endothelial growth factor (VEGF) and interleukin-8 (IL-8) were associated with complicated parapneumonic effusions [65]. Further prospective studies of serum or pleural biomarkers that define a population requiring pleural fluid drainage are needed.

Other tests — Although blood cultures are frequently negative in patients with parapneumonic effusion and empyema, they should be obtained. Growth from cultures can help make the microbiologic diagnosis as well as identify concurrent bacteremia. The need for additional microbiologic testing should be determined on a case-by-case basis (eg, sputum cultures, urine S. pneumoniae antigen testing, interferon-gamma release assays for tuberculosis, galactomannan, cryptococcal antigen).

Procalcitonin's usefulness for distinguishing bacterial from nonbacterial parapneumonic effusions is not well studied but is not likely to be high. In one study, serum procalcitonin levels >0.18 ng/mL were associated with a sensitivity of 83 percent and specificity of 81 percent for the pleural effusion having a bacterial infectious etiology [64]. However, this marker rises in the setting of invasive and systemic infections; thus the sensitivity for determining whether bacteria are present in a contained space infection is questionable. Neither high nor low procalcitonin levels are likely to change the need for drainage. (See "Procalcitonin use in lower respiratory tract infections", section on 'Procalcitonin biology'.)

DIFFERENTIAL DIAGNOSIS — Several conditions need to be entertained when evaluating a suspected parapneumonic effusion or empyema.

●Pleural fluid exudates – The differential diagnosis of an exudative pleural effusion is listed in the table (table 4). Further studies may be needed on pleural fluid or pleural biopsy may be required to distinguish these from one another, the details of which are discussed separately. (See "Pleural fluid analysis in adults with a pleural effusion".)

●Pleural fluid acidosis – Pleural fluid acidosis and/or low glucose, although highly suspicious for a complicated parapneumonic pleural effusion or empyema, can be associated with other diseases including malignancy, tuberculosis, rheumatoid pleuritis, lupus pleuritis, and urinothorax [67]. It may also be seen in patients who have a central venous catheter that is misplaced in the pleural space and is infusing isotonic fluid such as saline. These conditions can be excluded when clinically indicated with appropriate serology or further analysis of the pleural fluid. Pleural space infections caused by urease-splitting organisms such as Proteus species may result in a spuriously elevated pleural pH [68]. (See "Pleural effusion of extra-vascular origin (PEEVO)".)

●Pleural effusion or masses on imaging – Distinguishing loculated pleural fluid from a pleural mass on chest radiography or CT can be challenging but differences in the CT attenuation can help distinguish fluid from solid masses. However, if organization is advanced, occasionally, a pleural biopsy will be needed to assure that pleural based malignancy is excluded.

Chest CT can also help distinguish a lung abscess from empyema. Empyemas are more likely to compress adjacent lung rather than erode it, whereas lung abscesses are more likely to erode adjacent structures. Empyemas typically have thinner, smoother walls than lung abscesses, which tend to have thicker walls and irregular luminal and exterior surfaces. Empyemas tend to form an obtuse angle of interface with the chest wall, compared with lung abscesses, which commonly have an acute angle, although this feature is nonspecific. (See "Imaging of pleural effusions in adults", section on 'Computed tomography'.)

DIAGNOSIS — In most patients, the diagnosis of a parapneumonic effusion and empyema is made by a constellation of clinical signs and symptoms with confirmatory chemical and microbiologic data from pleural fluid (table 2 and algorithm 1). Rarely, a complicated parapneumonic effusion or empyema requires definitive diagnosis on pleural biopsy, most often by thoracoscopy, when microorganisms are demonstrated in or cultured from affected pleural tissue. Occasionally, a retrospective diagnosis may be made when patients present in the late stages of organization (ie, stage 3 (see 'Pathogenesis' above)) and fibrous peel is identified during thoracoscopy that requires decortication.

Uncomplicated parapneumonic effusion — A parapneumonic effusion is considered uncomplicated or simple when:

●The pleural effusion is free-flowing and too small to sample.

●A free-flowing small effusion has a neutrophilic exudate (an elevated protein level >0.5 percent of serum and/or a lactate dehydrogenase (LDH) level >0.6 that in the serum), a normal pH, a normal glucose level, and does not contain microorganisms.

In most cases, uncomplicated parapneumonic effusions resolve with appropriate antibiotic therapy and drainage is not generally necessary. However, the threshold to repeat imaging and obtain pleural fluid sampling should be low if patients do not respond adequately to antimicrobial therapy. (See "Management and prognosis of parapneumonic pleural effusion and empyema in adults", section on 'Uncomplicated parapneumonic effusion (antibiotics alone)'.)

Complicated parapneumonic effusion and empyema — A parapneumonic effusion is considered complicated when an exudative effusion has any one or more of the following characteristics: has a pH <7.2 and/or contains evidence of microorganism invasion by culture or Gram stain. In the absence of pH data, a low glucose level <40 mg/dL may be used instead. Although a serum LDH >1000 international units (IU)/L may support the diagnosis, it is non-specific and may simply indicate infection or significant inflammation elsewhere. It is typically large, loculated, or has associated pleural thickening. When it is bacterial, it generally contains a large number of neutrophils. For several reasons, cultures of fluid from complicated parapneumonic effusions are sometimes negative. Although a positive culture is confirmatory, this is not necessary for the diagnosis [69]. An empyema (due to pneumonia) is a complicated parapneumonic effusion in which frank pus is seen on pleural fluid sampling. (See 'Thoracentesis and pleural fluid analysis' above.)

In most cases of complicated parapneumonic effusion, drainage is indicated. Pleural pH <7.2 is the most useful predictor of a complicated clinical course [70]. If pleural pH is not measured, a pleural fluid glucose value <40 mg/dL and/or pleural fluid LDH value >1000 IU/L, or significant loculations are also predictive of the need for tube thoracostomy [70]. Empyema is an absolute indication for chest tube drainage. (See "Management and prognosis of parapneumonic pleural effusion and empyema in adults", section on 'Complicated pleural effusion and empyema (antibiotics plus drainage)'.)

The use of the RAPID score helps risk-stratify patients with pleural infection by five characteristics (renal failure, age, purulence, infectious source, and dietary factors) and may identify those at low, medium, and high risk of mortality from a pleural infection [71]. A calculator is available that estimates 3- and 12-month mortality (calculator 1).

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: Pleural effusion".)

SUMMARY AND RECOMMENDATIONS

●Definition and epidemiology – A parapneumonic effusion is a pleural effusion that forms in the pleural space adjacent to a pneumonia (table 1). When bacteria or other pathogens infect the pleural space, a complicated parapneumonic effusion or empyema may result. Parapneumonic effusions and empyema are common complications of pneumonia, but an empyema can also develop without the presence of an adjacent pneumonia. The incidence is approximately two to three percent of all pneumonias but appears to be rising. (See 'Introduction' above and 'Epidemiology' above.)

●Pathogenesis – Parapneumonic effusion and empyema may be part of a developmental spectrum that includes three stages (table 2). Not every patient progresses through each stage.

•Stage 1 is typically an early stage where fluid is free-flowing and small, such that resolution occurs with antibiotics alone.

•Stage 2 is a fibrinopurulent stage where infection precipitates organization and loculations.

•Stage 3 is characterized by a thickened fibrinous coating that may limit lung expansion. (See 'Pathogenesis' above.)

●Microbiology – The microbiology of complicated pleural effusions and empyema vary with source of infection (figure 1), route of acquisition, local epidemiology, and patient-specific risk factors (table 3). (See 'Microbiology' above.)

•Parapneumonic – Parapneumonic pleural space infections typically result as complications of community-acquired pneumonia (CAP) or aspiration pneumonia. Those associated with CAP tend to be caused by the pyogenic bacteria that cause CAP (eg, Streptococcus pneumoniae and Staphylococcus aureus). In contrast, those associated with aspiration pneumonia tend to be polymicrobial and caused by oral streptococci and anaerobes.

•Non-parapneumonic – The list of pathogens associated with non-parapneumonic pleural space infections is more extensive and varies by source (figure 1).

●Suspecting parapneumonic effusion – Among patients with pneumonia, the presence of risk factors (eg, aspiration, poor dental hygiene, alcohol or intravenous drug abuse, immunosuppression, young or old age, partially treated pneumonia, and gastroesophageal reflux) and the development of a persistent or new fever despite appropriate antibiotics should heighten the suspicion for the development of a parapneumonic effusion or an empyema. The clinical findings of a parapneumonic effusion and empyema are nonspecific and, with the exception of decreased fremitus, overlap with those of pneumonia. (See 'Risk factors' above and 'Clinical features' above.)

●Imaging – In all patients with pneumonia, the imaging modality that defined the pneumonia should be reexamined for evidence of pleural fluid. Once a pleural effusion is suspected on chest radiography, ultrasonography is typically performed at the bedside to evaluate the nature of the effusion and feasibility of sampling or drainage. CT is generally additionally performed when complications (eg, loculations) are suspected, complex interventions are planned, and/or more detail of the underlying anatomy is expected to help with management. (See 'Diagnostic imaging' above.)

●Thoracentesis – Most cases of suspected parapneumonic effusion or empyema should be sampled under ultrasound guidance unless the effusion is too small or sampling is deemed unsafe. (See 'Thoracentesis and pleural fluid analysis' above.)

•In some cases, therapeutic thoracentesis is simultaneously performed when symptomatic relief is needed or when drainage is indicated (eg, frank pus is observed).

•Fluid should be sent for cell count and differential, chemistries (protein and lactate dehydrogenase [LDH]), Gram stain and culture with additional inoculation of fluid into blood culture bottles (aerobic and anaerobic), pH (drawn directly into an arterial blood gas syringe and analyzed within one hour), and cytology. Special stains and cultures should be requested when unusual organisms or organisms that require special culture conditions are suspected.

●Differential diagnosis – Several conditions need to be entertained when evaluating a suspected parapneumonic effusion or empyema including other causes of an exudative effusion (table 4), pleural fluid acidosis (eg, malignancy, tuberculosis, rheumatoid pleurisy, lupus pleuritis, urinothorax, or infused saline), and pleural masses or lung abscesses. (See 'Differential diagnosis' above.)

●Diagnosis and classification – In most patients, the diagnosis of a parapneumonic effusion and empyema is made by a constellation of clinical signs and symptoms with confirmatory chemical and microbiologic data from pleural fluid (table 2). It is rare that pleural biopsy is required. In general, we and others categorize parapneumonic effusions as the following (see 'Diagnosis' above):

•Uncomplicated – A parapneumonic effusion is considered uncomplicated (simple) when the effusion is free-flowing and too small to sample or when a small free-flowing effusion has a neutrophilic exudate (an elevated protein level >0.5 that in serum and/or a LDH level >0.6 that in serum) with normal pH and glucose levels and does not contain microorganisms. These effusions typically resolve with antibiotics, although the threshold to reimage should be low to look for disease progression.

•Complicated – A parapneumonic effusion is considered complicated when an exudative effusion has any one or more of the following characteristics: has a pH <7.2 (or a low glucose level <40 mg/dL in the absence of pH data) and/or contains evidence of microorganism invasion by culture or Gram stain. It is typically large, loculated, and has associated pleural thickening.

•Empyema – An empyema is considered present when frank pus is seen on pleural fluid sampling.