Congestive heart failure and cirrhosis are the most common causes of transudative pleural effusions, while pneumonia and malignancy are responsible for the majority of exudative effusions. Plain chest films are valuable in confirming the presence of an effusion, providing clues to the cause, and directing the method of pleural fluid sampling. Thoracentesis is safe and simple, and it is diagnostic in about 75% of cases. Pleural biopsy is indicated for unexplained exudative effusions, most of which are found to result from malignancy or tuberculosis. Medical thoracoscopy, if available, is the procedure of choice for patients with these effusions. Thoracoscopy permits visually directed pleural biopsies and allows for therapeutic intervention at the time of diagnosis, obviating the need for subsequent invasive procedures.
More than 1 million cases of pleural effusion occur in the United States each year. Pleural effusions are a common presentation of many pulmonary and systemic diseases, including congestive heart failure (CHF), malignancy, sarcoidosis, and infection. About 25% to 30% of patients seen by a pulmonary consultant have evidence of pleural disease.
In this article, we will describe the steps to follow when evaluating a patient who has suspected pleural effusion. Specifically, we will discuss the clinical presentation, chest film findings, and the role of thoracentesis and medical thoracoscopy.
CAUSES OF PLEURAL EFFUSION
The pleural space is 10 to 20 µm in width and normally contains about 0.1 mL/kg of fluid. A volume greater than 7 to 14 mL is abnormal. Many mechanisms can result in abnormal amounts of pleural fluid, including:
Generally, transudative effusions are formed in response to increased hydrostatic pressure, while exudative effusions form when pleural inflammation or disrupted lymphatic drainage results in increased protein leak or decreased protein removal from the pleural space. In CHF, pleural effusions are secondary to pulmonary venous hypertension. Neoplasms can cause pleural effusions by direct involvement of the pleura, by lymphatic obstruction, or in association with a post-obstructive pneumonia. Pleural effusions associated with pulmonary embolism are secondary to increased capillary permeability, pleuropulmonary hemorrhage, and increased hydrostatic pressure.
Many patients with pleural effusions are asymptomatic. However, when symptoms arise, they do so because of pleural inflammation or the effusion's effects on mechanics. The most common symptoms of pleural effusion are dyspnea, nonproductive cough, and pleuritic chest pain.
The mechanisms by which dyspnea occurs are not well understood, but they do not appear to correlate with blood oxygen levels or the size of the pleural effusion. Dyspnea is probably related to increased thoracic cage size, which affects respiratory muscle function. Nonproductive cough may occur secondary to lung compression and resultant bronchial irritation.
Pleuritic chest pain is associated with inflammation of the parietal pleura. Pain is occasionally referred to the abdomen. If the central portion of diaphragmatic pleura is involved, patients experience pain in the lower chest and ipsilateral shoulder simultaneously. Historical features, including underlying disease processes, drug use, and radiation therapy, can alert you to the possibility of pleural effusion in a patient with less common symptoms.
Several physical findings suggest the presence of pleural effusion. Tactile fremitus is lost over the area of effusion because voice-induced vibrations are attenuated by the fluid adjacent to aerated lung. This finding is more sensitive than the use of percussion for detecting pleural fluid collections.
Absent or diminished breath sounds over the area of effusion are characteristic. Pleural friction rubs are occasionally noted in the initial stages or as the effusion resolves and are caused by roughened pleural surfaces moving across one another.
Radiography. Once you suspect a pleural effusion, obtain chest films to confirm its presence and to look for other abnormalities that can help identify the cause (Table 1). Small pleural effusions appear on plain films as blunting of the normally sharp costophrenic angles. These effusions usually represent more than 100 mL of fluid. If the effusion volume is greater than about 500 mL, chest films are 100% sensitive (Figure 2).1
Lateral decubitus films are essential to confirm that the effusion is not loculated. Loculated effusions may be misinterpreted as parenchymal infiltrates.
Cardiomegaly with bilateral pleural effusions is most consistent with a diagnosis of CHF. The differential diagnosis of a normal heart size with bilateral effusions includes malignancy (most common), rheumatoid pleurisy, systemic lupus erythematosus (SLE), esophageal rupture, nephrotic syndrome, and cirrhosis with ascites. Massive effusions are most commonly caused by malignancy.
Ultrasonography. An ultrasound scan can detect as little as 5 mL of pleural fluid. Ultrasonography is extremely sensitive in identifying septations within a pleural fluid collection. Small or loculated effusions are best tapped under ultrasonographic guidance.
CT. A CT scan of the chest is not appropriate for the initial confirmation of pleural effusion. It is most helpful after thoracentesis for detecting suspected parenchymal or pleural abnormalities. A CT scan is ideal if pleural abnormalities are seen on the chest radiograph, because it allows imaging of the entire pleural space and differentiates pleural fluid from pleural thickening or pleural-based masses.2
In empyema, a CT scan often shows loculation and pleural enhancement. Asbestos-related pleural effusion may exhibit pleural calcifications and evidence of interstitial lung disease. In malignancy, a CT scan may show pulmonary masses and/or mediastinal adenopathy.
In most cases, the discovery of a pleural effusion warrants prompt thoracentesis to determine the cause as well as to direct initial therapy. If a layered effusion is 1 cm or greater in width, it is easily tapped at the bedside. Laboratory study of fluid removed by thoracentesis is diagnostic in about 75% of cases; relevant information for management is obtained in an additional 15% to 20%.3 Definitive diagnoses can be made in patients with malignancy, empyema, tuberculosis, fungal infection, lupus pleuritis (if lupus erythematosus cells are seen), chylothorax, urinothorax, and esophageal rupture.
However, it is appropriate to forgo thoracentesis in patients who have bilateral pleural effusions and uncomplicated CHF without fever, pleuritic chest pain, or severe hypoxemia. These patients can be monitored during therapy for CHF. Following diuretic therapy, the transudative effusions associated with CHF sometimes appear exudative, with increased protein and lactate dehydrogenase (LDH) levels.4,5
While there are no absolute contraindications to thoracentesis, relative contraindications include bleeding diathesis, anticoagulation, and small pleural effusions. The complication rate is about 20%; complications associated with thoracentesis include pneumothorax (12%), cough (9%) and, rarely, bleeding, empyema, and spleen or liver puncture.3
Only half of all pneumothoraces are clinically important; about one third of them require chest tube placement. Anxiety and pain at the site of thoracentesis are reported by about 20% of patients.
Large-volume taps to relieve dyspnea can result in unilateral pulmonary edema or hypoxemia; the latter occurs in up to 50% of patients. Despite this resultant hypoxemia, the patient's dyspnea is often relieved by thoracentesis. If it is not relieved and the pleural effusion reaccumulates, repeated thoracenteses for dyspnea alone are not indicated.
Routine use of post-tap chest films to check for pneumothorax is not necessary. Aspiration of air during thoracentesis often correlates with occurrence of a pneumothorax.6 However, obtaining a chest film following thoracentesis may help identify an underlying parenchymal abnormality.
The gross appearance of the fluid obtained often helps identify the cause of the effusion (Table 2). If the tap reveals a milky supernatant, lipid studies are indicated to confirm the diagnosis of chylothorax. The initial analysis of pleural fluid should include measurements of protein, LDH, and glucose concentrations; a cellular differential; and pH. If you suspect infection, order Gram, acid-fast, and potassium hydroxide stains and cultures for bacteria, acid-fast bacilli, and fungi.
Distinguishing an effusion as exudative or transudative is an important initial step. Transudative effusions characteristically have low protein and LDH concentrations. Exudative effusions have elevated cell counts with higher protein and LDH concentrations.
The classic criteria defining an exudate were established by Light and colleagues.7 An effusion is exudative if it meets 1 of the following criteria:
Although subsequent investigators have studied the sensitivity and specificity of other discriminators, such as bilirubin, cholesterol, and albumin levels, these are not superior to the measurement of LDH and protein levels.
The differential diagnosis of a transudative pleural effusion is limited (Table 3). Once confirmed, therapy consists of managing the underlying disease. The differential diagnosis of an exudative pleural effusion is extensive and requires additional analysis (Tables 4, 5, 6, and 7).8
Cytologic analysis is critical in any patient who has suspected malignancy or an exudate of unknown origin. In addition to identifying malignant cells, cytologic analysis with differential cell count can help narrow the differential diagnosis (Table 8). As little as 10 mL of fluid is adequate for diagnosing malignant pleural effusion.9 The diagnostic yield of thoracentesis for malignancy is approximately 60% and improves slightly with a repeated thoracentesis.
Obtain a fluid amylase level if you suspect esophageal rupture, pancreatitis, or pancreatic pseudocyst or if the patient has a left pleural effusion of unknown cause. The pleural fluid amylase level is always elevated in patients with chronic pancreatitis but can be normal in those with early, acute pancreatitis or in patients immediately after esophageal rupture.8
A glucose level of less than 60 mg/dL or a pleural fluid glucose/ serum glucose ratio of less than 0.5 is a classic finding in patients with rheumatoid lung disease. A rheumatoid factor that is greater than 1:320 and an antinuclear antibody titer that is greater than 1:160 suggest SLE or rheumatoid pleurisy; the presence of lupus erythematosus cells is diagnostic of SLE.8
Low glucose levels are common in the pleural fluid of patients with empyema, malignancy, or tuberculosis and are occasionally found in patients with SLE. In patients with rheumatoid lung disease or malignancy, low glucose levels are the result of abnormal glucose transport. Otherwise, the low levels are caused by increased use of glucose by inflammatory or malignant cells.
Normal pleural fluid pH is 7.6. Transudative effusions usually maintain a pH of 7.4 to 7.55; exudates have a lower pH--7.3 to 7.45. A low pleural fluid pH (less than 7.3) is caused by increased cellular acid production or decreased acid removal (resulting from pleural inflammation), or by the presence of tumor or fibrosis.
Low pleural fluid pH is common in patients who have esophageal rupture, empyema, rheumatoid lung disease, malignancy, or tuberculosis. Parapneumonic effusions that have a pH less than 7.1, glucose concentration less than 40 mg/dL, and LDH concentration greater than 1000 IU/L should be drained to promote resolution of infection.10
An elevated pleural fluid triglyceride level (greater than 110 mg/dL) confirms the diagnosis of chylothorax. However, if the triglyceride level is intermediate (50 to 110 mg/dL), lipoprotein electrophoresis is indicated to determine the presence of chylomicrons, which is diagnostic.11
Chylothorax occurs with disruption of the patient's thoracic duct, as in trauma; following thoracotomy; or with the presence of malignancy, tuberculosis, or pulmonary lymphangiomyomatosis.
Pleural fluid adenosine deaminase (ADA) can be a useful marker for diagnosis of tuberculosis, especially in the presence of a lymphocytic effusion. Its sensitivity varies from 78% to 99% and its specificity, from 85% to 97%.12 Pleural fluid ADA can be elevated in mesothelioma and in rheumatoid, malignant, and parapneumonic effusions. Although not widely available, pleural fluid interferon gamma is highly sensitive and specific for tuberculosis if appropriate cutoff values are used.13
PERCUTANEOUS PLEURAL BIOPSY
An exudative effusion that remains unexplained following pleural fluid analysis is often caused by tuberculosis or malignancy. Percutaneous pleural biopsy has a sensitivity of 75% for tuberculosis. The combination of pleural biopsy and fluid culture has a diagnostic sensitivity of 90% for tuberculosis.14
While pleural fluid cytology has a sensitivity of 60% to 90% for diagnosing malignant pleural effusion, closed pleural biopsy has a sensitivity of 40% to 75%. Biopsy occasionally leads to a diagnosis of fungal or parasitic infection, sarcoidosis, or rheumatoid pleurisy.8 Between one third and three fourths of all exudative effusions that are unexplained following thoracentesis and pleural biopsy eventually prove to be secondary to malignancy.14,15
Contraindications to pleural biopsy include an obliterated pleural space, an uncooperative patient, a bleeding diathesis, and anticoagulation. The most common complications include pneumothorax (3% to 15%), hemothorax (less than 2%), pain (1% to 15%), and vasovagal episodes (1% to 5%).7
About 20% to 25% of pleural effusions remain undiagnosed following thoracentesis and closed pleural biopsy; most of these effusions are exudative. Common causes include pleural tuberculosis, pulmonary embolism, and malignancy.
Medical thoracoscopy has a diagnostic yield of more than 90% in the diagnosis of malignant pleural effusion. Compared with video-assisted thoracic surgery (VATS), it is less invasive and can be performed in the bronchoscopy suite under conscious sedation. The advantage of thoracoscopy over closed pleural biopsy is the ability to visualize the entire pleural surface and to perform biopsy of any abnormal pleura under direct vision.16 In addition, talc poudrage, which has an effectiveness of greater than 90% in achieving pleurodesis in malignant pleural effusion, can be performed in the same setting. Medical thoracoscopy is the procedure of choice in undiagnosed exudative pleural effusions. Fewer than 10% of pleural effusions remain undiagnosed after medical thoracoscopy.
An absolute contraindication to thoracoscopy is an obliterated pleural space resulting from adhesions; relative contraindications include coagulopathy, severe hypoxemia, and cardiac disease. The most common complications are post-procedure fever (15%) and persistent air leak (2%).
Fiberoptic bronchoscopy has no role in evaluating unexplained exudative effusions unless the patient has hemoptysis, radiographic evidence of lung mass, or mediastinal shift toward the side of the effusion, which would indicate possible endobronchial obstruction.
Bartter T, Akers SM, Pratter MR. The evaluation of pleural effusion.
McLoud TC, Flower CD. Imaging the pleura: sonography, CT, and MR imaging.
. 1991;156: 1145-1153.
Collins TR, Sahn SA. Thoracocentesis. Clinical value, complications, technical problems, and patient experience.
Chakko SC, Caldwell SH, Sforza PP. Treatment of congestive heart failure. Its effects on pleural fluid chemistry.
Shinto RA, Light RW. Effects of diuresis on the characteristics of pleural fluid in patients with congestive heart failure.
Am J Med
Petersen WG, Zimmerman R. Limited utility of
chest radiograph after thoracentesis.
Light RW, Macgregor MI, Luchsinger PC, Ball
WC. Pleural effusions: the diagnostic separation of transudates and exudates.
Ann Intern Med
. 1972;77: 507-513.
Sahn SA. State of the art. The pleura.
Am Rev Respir Dis
Sallach SM, Sallach JA, Vasquez E, et al. Volume of pleural fluid required for diagnosis of pleural malignancy.
Light RW, Girard WM, Jenkinson SG, George RB. Parapneumonic effusions.
Am J Med
. 1980; 69: 507-512.
Staats BA, Ellefson RD, Budahn LL, et al. The lipoprotein profile of chylous and nonchylous pleural effusions.
Mayo Clin Proc
. 1980; 55:700-704.
Bañales JL, Pineda PR, Fitzgerald JM, et al. Adenosine deaminase in the diagnosis of tuberculous pleural effusions. A report of 218 patients and review of the literature.
Valdes L, San Jose E, Alvarez D, et al. Diagnosis of tuberculous pleurisy using biological parameters adenosine deaminase, lysozyme and interferon gamma.
Enarson DA, Dorken E, Grzybowski S. Tuberculous pleurisy.
Can Med Assoc J
Menzies R, Charbonneau M. Thoracoscopy for the diagnosis of pleural disease.
Ann Intern Med.
Boutin C, Viallat JR, Cargnino P, et al. Thoracoscopy in malignant pleural effusions.
Am Rev Respir Dis.