INTRODUCTION — Pleural effusions can develop as a result of over 50 different pleuropulmonary or systemic disorders. Following diagnostic thoracentesis, the cause of a pleural effusion is not evident in up to 25 percent of patients [1,2]. However, no universally accepted definition exists for an "undiagnosed effusion."
This topic will review the approach to pleural effusions for which the diagnosis is unclear after initial clinical assessment and investigation. The initial assessment of a patient with a pleural effusion is presented separately. (See "Imaging of pleural effusions in adults" and "Diagnostic evaluation of a pleural effusion in adults: Initial testing" and "Ultrasound-guided thoracentesis".)
HISTORY — The first step for the clinician is to revisit the patient's history, paying particular attention to drugs, occupational exposures, risk factors for pulmonary embolism or tuberculosis, and comorbid conditions.
A careful drug history may reveal that the patient is taking nitrofurantoin, amiodarone, ovarian stimulation therapy, or a drug that can produce a lupus-like syndrome [3,4] (see "Drug-induced lupus"). Immunotherapy-related pleural effusion, although uncommon, has also been described [5].
Occupational asbestos exposure, which might suggest a benign asbestos pleural effusion, may have occurred many years earlier [6]. Benign asbestos effusions are usually unilateral, exudative, and about a third have an elevated pleural eosinophil count [7].
Many systemic disorders (eg, lupus, hypothyroidism, amyloid, yellow nail syndrome) can cause an effusion and must not be overlooked (see "Pulmonary manifestations of systemic lupus erythematosus in adults" and "Overview of amyloidosis"). Patients with yellow nail syndrome typically have friable, yellow nails that grow slowly and may also have lymphedema [8,9]. Abnormal nail fold capillaroscopy has also been reported [8]. A high serum immunoglobulin G4 (IgG4) level should raise the possibility of IgG4 pleuritis [10].
Pleural fluid may originate from extrapleural sources, most commonly from transdiaphragmatic movement. A detailed history and examination for ascites, urinary tract obstruction, and hepatic and pancreatic diseases are important. (See "Hepatic hydrothorax".)
TIME COURSE — Some effusions resolve spontaneously, but the time required for resolution varies depending upon the underlying etiology. Uncomplicated parapneumonic effusions and effusions from pulmonary embolism, tuberculous pleurisy, and postcardiac injury syndrome may persist for several weeks. (See "Clinical presentation, evaluation, and diagnosis of the nonpregnant adult with suspected acute pulmonary embolism" and "Epidemiology, clinical presentation, and diagnostic evaluation of parapneumonic effusion and empyema in adults" and "Tuberculous pleural effusion" and "Evaluation and management of pleural effusions following cardiac surgery".)
Benign asbestos pleural effusion, rheumatoid pleurisy, and radiation pleuritis often persist for months to years. Other effusions that may persist for years include those caused by lymphatic abnormalities (eg, yellow-nail syndrome and pulmonary lymphangiectasia) and trapped lung [9,11]. Malignant pleural effusions, on the other hand, seldom resolve spontaneously. (See "Management of malignant pleural effusions".)
REANALYSIS OF PLEURAL FLUID — The pleural fluid should be analyzed carefully: vital diagnostic clues may lie in simple inspection (especially for chylous effusions) or the smell (in urinothorax), but are often overlooked (see "Pleural effusion of extra-vascular origin (PEEVO)"). Most undiagnosed pleural effusions are exudative, because the differential diagnosis of a transudative effusion is limited and the cause can usually be recognized by the clinical presentation (table 1) (table 2) (see "Management of nonmalignant pleural effusions in adults"). A few points are worth noting when categorizing an undiagnosed effusion as a transudate or exudate:
●One common problem is that transudative pleural fluid from patients receiving diuretic therapy may have an elevated protein level and be mistakenly classified as an exudate. In such cases, the N-terminal pro-brain natriuretic peptide (NTproBNP) level in the pleural fluid, especially if higher than 1500 pg/mL, can be used to determine if the pleural fluid is a transudate due to heart failure [12,13]. NTproBNP has been shown to correctly diagnose heart failure as a cause of most effusions that have been misclassified as exudates by Light's criteria and can help avoid repeated invasive investigations in patients with a strong clinical suspicion of cardiac failure [14] (see "Diagnostic evaluation of a pleural effusion in adults: Initial testing"). If NTproBNP is unavailable, the fluid can be considered a transudate if the total protein gradient (serum minus pleural fluid) is greater than 3.1 g/dL in a patient with convincing clinical picture of heart failure who is taking diuretics [15].
●Constrictive pericarditis, which can be a sequel of prior cardiac surgery, infection, or remote radiation therapy, can also be difficult to diagnose. The resulting pleural effusion is typically transudative, but can be exudative with effusive-constrictive pericarditis. (See "Constrictive pericarditis: Diagnostic evaluation and management".)
●Effusions that are classically exudative can be transudative in some cases (eg, malignancy), usually a result of concomitant causes of transudative fluid formation (table 2) [11,16,17]. One study of 638 malignant pleural effusions from 103 patients reported a significant reduction in pleural fluid protein (8 g/L/100 days) and pH (0.04/100 days) accompanied by a small rise in lactate dehydrogenase over time [17]. (See "Diagnostic evaluation of a pleural effusion in adults: Initial testing".)
●In patients with a lymphocytic effusion, further investigation should be considered for tuberculosis (TB), sarcoidosis, lymphoma (including primary pleural lymphoma), chylothorax, and pseudochylothorax [18]. Flow cytometry may be helpful in the evaluation of pleural lymphoma [19]. Pleural fluid adenosine deaminase (ADA) may be helpful when TB is suspected [20]. A meta-analysis of 63 studies showed that pleural fluid ADA level provides 92 percent sensitivity and 90 percent specificity in diagnosing tuberculous pleuritis [21]. (See "Management of malignant pleural effusions".)
●A viral etiology has been postulated for effusions in which no causes were found despite investigation, especially if the effusion resolved without specific treatment. A wide range of viruses have been associated with a pleural effusion [22]. However, there are no pathognomonic features that allow easy diagnosis of a viral pleural effusion. No studies to date have studied how common viral infections are in otherwise undiagnosed effusions.
●Polyserositis involving the pleura and pericardium (and at times the peritoneum) have been associated with autoimmune diseases, especially lupus, though in many cases the cause was not identified. Genetic testing is helpful if familial Mediterranean fever is suspected.
TRAPPED LUNG — The diagnosis of trapped lung should be suspected in any patient with a stable chronic pleural effusion, particularly if there is a past history of pneumonia, pneumothorax, thoracic surgery, or hemothorax [11]. A trapped lung occurs when a remote inflammatory process causes a fibrin peel to form on the visceral pleural surface, thereby preventing the lung from expanding to the chest wall. This generates a negative pleural pressure favoring movement of fluid from the parietal pleural capillaries into the pleural space, until a new steady state is reached. Other causes of trapped lung include rheumatoid pleurisy, uremic pleurisy, tuberculosis, and malignancy. Diagnosis and treatment of nonexpandable lung are discussed separately. (See "Mechanisms of pleural liquid turnover in the normal state" and "Mechanisms of pleural liquid accumulation in disease" and "Diagnosis and management of pleural causes of nonexpandable lung" and "Measurement and interpretation of pleural pressure (manometry): Indications and technique".)
DIAGNOSTIC EVALUATION — If clinical examination and pleural fluid analysis fail to result in a diagnosis, additional investigations with imaging and pleural biopsy will be needed. Bronchoscopy is rarely helpful in identifying the cause of an undiagnosed pleural effusion, unless the chest radiograph or computed tomography (CT) show parenchymal abnormalities or the patient presents with hemoptysis [23-25].
Pleural pressure — The gold standard for determining if a lung is "trapped" or nonexpandable is by radiology after removal of pleural effusion or air. Although pleural manometry has attracted strong interests [11], studies cast significant doubts on its clinical value and validity in a routine setting. In a study of 70 patients undergoing thoracentesis for malignant pleural effusion, 54 percent had incomplete lung expansion on post-drainage radiographs [26]. Although elevated pleural elastance was associated with a higher likelihood of nonexpandable lung, discordance between postprocedural radiograph findings and pleural manometry was found in one-third of patients, including those with normal elastance but nonexpandable lung on radiographs and vice versa. A two-center, randomized trial of 124 patients undergoing thoracentesis, found no differences in the rates of procedure-related chest discomfort when patients undergoing manometry were compared with those who did not undergo manometry [27]. (See "Diagnosis and management of pleural causes of nonexpandable lung" and "Measurement and interpretation of pleural pressure (manometry): Indications and technique".)
Imaging — CT imaging of the thorax with pleural phase contrast enhancement should be performed in virtually all patients with an undiagnosed pleural effusion [28]. Irregular or thickened pleura with contrast enhancement suggests pleural inflammation or malignancy and identifies optimal sites for CT-guided needle aspiration or cutting needle biopsy. CT may also reveal signs of invasion of underlying or adjacent structures that would suggest malignant pleural disease. (See "Imaging of pleural plaques, thickening, and tumors".)
CT pulmonary angiography may be considered if pulmonary embolism (PE) is a consideration. Two studies of 60 and 230 patients with pulmonary emboli showed that 47 to 48 percent had an effusion, the majority of which were small and did not require drainage [18,29]. Large effusions should alert clinicians to alternative or additional causes of fluid formation other than pulmonary embolism.
Positron emission tomography (PET)/CT has an emerging role: 18-fluorodeoxyglucose (FDG)-avidity confirms, but cannot differentiate, between inflammatory and malignant disease. Focal increased uptake of FDG in the pleura and the presence of solid pleural abnormalities on CT are suggestive of malignant pleural disease, and can guide biopsies. A PET/CT pattern composed of pleural uptake and increased effusion activity had an accuracy of 90 percent in predicting malignant pleural effusions in 31 patients with known extrapulmonary malignancy and a pleural effusion [30]. A negative PET/CT does not exclude malignancy [31]. PET/CT may also highlight extrapleural abnormalities that may be the cause of the effusion.
Pleural biopsy — Pleural biopsy typically follows CT scan in undiagnosed pleural effusions. A number of techniques for pleural biopsy are available.
●Percutaneous techniques include closed pleural biopsy and CT-guided cutting needle biopsy. The former is useful primarily when diseases such as tuberculosis (TB) are suspected (eg, because of a lymphocytic effusion or risk factors for TB). Ultrasound guidance can improve the yield of percutaneous biopsy [32]. CT guided biopsy is useful when a pleural-based mass is visible [33].
●Thoracoscopic pleural biopsy is increasingly used to diagnose malignancy when an obvious mass is not visible on CT, when percutaneous biopsy is negative, or when patchy disease is suspected [34].
●Open pleural biopsy by thoracotomy has been used in the past, but has been largely replaced by thoracoscopic pleural biopsy [35].
Image-guided cutting needle biopsy — CT- or ultrasound-guided cutting needle biopsy is typically used to diagnose malignancy when pleural-based soft tissue masses are identified. In a randomized trial, CT-guided biopsy had a sensitivity of 87 percent (versus 47 percent with Abrams needle biopsy) in patients with suspected malignant effusions with negative cytology [33]. In another study of 85 cases, the sensitivity and specificity were 76 and 100 percent, respectively, for image guided biopsy for pleural malignancies. A pneumothorax occurred in only 4 percent of cases [36]. Ultrasound-guided pleural biopsy had similar diagnostic sensitivity in suspected TB pleuritis as thoracoscopy in 196 patients [37]. In another small study of 63 patients with undiagnosed pleural effusion, ultrasound guided biopsy with an Abram needle demonstrated comparable diagnostic yield for pleural malignancy as CT-guided Tru-Cut biopsies [38].
Thoracoscopic pleural biopsy — Thoracoscopic pleural biopsy, in the form of video-assisted thoracoscopic surgery or pleuroscopy (also called medical thoracoscopy), has a diagnostic yield for malignant pleural effusions that is superior to percutaneous pleural biopsy [39-41]. It should be noted that a negative thoracoscopic biopsy does not fully exclude malignant pleural disease. False negative results (around 8 percent) can occur, especially with mesothelioma [2,42]. The use of new molecular diagnostic tools (especially immunohistochemistry for BRCA-associated protein-1 and fluorescent in situ hybridization for CDKN2A loss) has enhanced the diagnostic yield for mesothelioma [43]. In a large series of medical thoracoscopy [44], 80 of the 662 procedures performed for diagnostic purposes did not yield a diagnosis. It is also important that thorough examination of the pleural tissue be conducted for less common conditions such as IgG4 pleuritis and amyloidosis [45,46]. (See "Medical thoracoscopy (pleuroscopy): Equipment, procedure, and complications" and "Medical thoracoscopy (pleuroscopy): Diagnostic and therapeutic applications", section on 'Diagnostic evaluation of an exudative pleural effusion of unknown etiology'.)
Closed pleural biopsy — Closed pleural biopsy is useful in conditions where the pleura is diffusely involved (eg, tuberculous pleuritis or noncaseating granulomata in rheumatoid pleuritis) but less so if involvement is patchy (eg, pleural malignancies).
Closed pleural biopsy of the parietal pleura with a Cope or Abrams needle is operator dependent. In a retrospective analysis of 75 patients who underwent percutaneous pleural biopsy, sufficient pleural tissue was obtained from 72 percent of patients when three samples were obtained, which increased to 80 percent with four to six samples [47]. For all diagnoses, the sensitivity, specificity, positive predictive, and negative predictive values were 38, 100, 100, and 51 percent, respectively. Pneumothorax complicated 11 percent of procedures.
The initial pleural biopsy in tuberculous pleuritis is positive in approximately 70 percent of cases. A greater number of biopsies taken at a single session (six or more), or multiple, separate biopsy procedures can increase the sensitivity to 80 percent [48]. The combination of pleural fluid adenosine deaminase level and closed pleural biopsy may be adequate to establish a diagnosis of pleural TB [39]. (See "Tuberculous pleural effusion".)
TUBERCULOUS PLEURISY — Tuberculous pleurisy affects all age groups and should not be overlooked [49]. An initial negative skin test does not exclude the diagnosis of pleural tuberculosis (TB); the need for further diagnostic procedures in this setting is dictated largely by the likelihood of TB exposure.
Elevated levels of pleural fluid adenosine deaminase (ADA) and interferon-gamma increase the likelihood that a lymphocytic effusion is due to TB [50]. Empiric treatment for TB may be considered in selected PPD (purified protein derivative) positive patients with an undiagnosed, lymphocyte-predominant exudate with raised ADA levels and no alternative likely etiology. (See "Tuberculous pleural effusion", section on 'Diagnosis' and "Tuberculous pleural effusion".)
FOLLOW-UP — The cause of the pleural effusion may become obvious with long-term follow-up, though most cases remain of unknown etiology [2,51-53]. One study monitored 143 patients who had a non-diagnostic pleural fluid analysis and pleural biopsy. Follow-up ranging from one to six years revealed that 29 (20 percent) had malignancy, and one had tuberculosis (TB) [51]. However, another study found neither malignancy nor TB during a mean follow-up of 33 months in 53 patients with nonspecific pleuritis [52]. In another series of 53 patients labelled as having non-specific pleuritis after pleural fluid analyses and thoracoscopic biopsies and followed up for 24 months (median), two were diagnosed with malignancy [54]. On the other hand, a malignant diagnosis was revealed in 15 percent of 52 patients with non-specific pleurisy on thoracoscopic biopsies after a median follow-up of 35.5 months [55]. In a retrospective study of 83 patients with undiagnosed "idiopathic" effusions, 47 resolved, 20 improved, and 16 persisted during follow-up [53].
Biochemical pleural fluid analysis did not predict these outcomes [53]. During follow-up, particular attention is paid to identifying "treatable" causes of a pleural effusion (eg, TB, lymphoma) [56].
Overall, patients with idiopathic pleural effusions generally have favorable outcomes [2,53]. The optimal duration of follow-up has not been defined.
It should be emphasized that “chronic fibrinous pleuritis” is a descriptive pathological term and should not be considered a clinical diagnosis. Patients with histological results of chronic fibrinous pleuritis should be managed as having an undiagnosed pleural effusion.
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
●Although no universally accepted definition exists for an "undiagnosed effusion," the cause of a pleural effusion may not be evident following diagnostic thoracentesis in up to 25 percent of patients. (See 'Introduction' above.)
●The first step for the clinician is to revisit the patient's history, paying particular attention to drugs, occupational exposures, risk factors for pulmonary embolism and tuberculosis (TB), and comorbid conditions. (See 'History' above.)
●Uncomplicated parapneumonic effusions and effusions from pulmonary embolism, tuberculous pleurisy, and postcardiac injury syndrome may resolve spontaneously after several weeks, but malignant effusions seldom resolve spontaneously. Benign asbestos effusions, rheumatoid pleurisy, radiation pleuritis, yellow nail syndrome, and trapped lung may persist for months to years. (See 'Time course' above.)
●A trapped lung occurs when a remote inflammatory process causes a fibrin peel to form on the visceral pleural surface, thereby preventing the lung from expanding to meet the chest wall. The resultant negative intrapleural pressure contributes to pleural fluid accumulation. (See 'Trapped lung' above and "Measurement and interpretation of pleural pressure (manometry): Indications and technique".)
●Computed tomography (CT) scan of the thorax with contrast should be performed in virtually all patients with an undiagnosed pleural effusion. Additional imaging modalities that may be helpful include CT pulmonary angiography and positron emission tomography (PET)/CT scans. (See 'Imaging' above.)
●The decision whether to pursue more invasive procedures or simply observe the patient with an undiagnosed pleural effusion requires clinical judgment that is guided by the patient's presentation. As an example, the patient who is over 60 years of age, with an abnormal CT of the pleura, is more likely to have a malignancy as the cause of an undiagnosed exudate and needs further diagnostic evaluation. (See 'Diagnostic evaluation' above.)
●Closed pleural biopsy is most useful when the pleural abnormality is diffuse as in tuberculous pleurisy. (See 'Closed pleural biopsy' above.)
●Imaging-guided pleural biopsy is useful when a pleural based soft tissue abnormality is identified by CT scan or pleural ultrasonography. (See 'Image-guided cutting needle biopsy' above.)
●Thoracoscopic biopsy is helpful for malignant effusions, because of the patchy distribution on the pleural surface, and for tuberculous effusions in the setting of a negative closed biopsy. (See 'Thoracoscopic pleural biopsy' above.)
●Tuberculous pleurisy can affect any age group and should not be overlooked. Clues to the diagnosis include risk factors for TB exposure and a lymphocytic effusion, especially when pleural adenosine deaminase and interferon-gamma are elevated. (See 'Tuberculous pleurisy' above.)
