Diagnostic Approach to Pleural Effusion in Adults JOSÉ M. PORCEL, M.D., Arnau de Vilanova University Hospital, Lleida, Spain RICHARD W. LIGHT, M.D., Saint Thomas Hospital, Nashville, Tennessee P leural effusion develops when more fluid enters the pleural space than is removed. Potential mechanisms of pleural fluid accumulation include: increased interstitial fluid in the lungs second- ary to increased pulmonary capillary pressure (i.e., heart failure) or permeability (i.e., pneu- monia); decreased intrapleural pressure (i.e., atelectasis); decreased plasma oncotic pres- sure (i.e., hypoalbuminemia); increased pleu- ral membrane permeability and obstructed lymphatic flow (e.g., pleural malignancy or infection); diaphragmatic defects (i.e., hepatic hydrothorax); and thoracic duct rupture (i.e., chylothorax). Although many different dis- eases may cause pleural effusion, the most common causes in adults are heart failure, malignancy, pneumonia, tuberculosis, and pulmonary embolism, whereas pneumonia is the leading etiology in children.1,2 Initial Evaluation of Pleural Effusion The history and physical examination are critical in guiding the evaluation of pleural effusion (Table 1). Signs and symptoms of an effusion vary depending on the underlying disease, but dyspnea, cough, and pleuritic chest pain are common. Chest examination of a patient with pleural effusion is nota- ble for dullness to percussion, decreased or absent tactile fremitus, decreased breath sounds, and no voice transmission. Posteroanterior and lateral chest radio- graphs usually confirm the presence of a pleural effusion, but if doubt exists, ultra- sound or computed tomography (CT) scans are definitive for detecting small effusions and for differentiating pleural fluid from pleural thickening.3 Small amounts of pleural fluid not readily seen on the standard frontal view may be recognized in a lateral decubitus view (Figures 1a and 1b). On a posteroante- rior radiograph, free pleural fluid may blunt the costophrenic angle; form a meniscus laterally; or hide in a subpulmonic location, simulating an elevated hemidiaphragm. Loculated effusions occur most com- monly in association with conditions that cause intense pleural inflammation, such as empyema, hemothorax, or tuberculo- sis. Occasionally, a focal intrafissural fluid collection may look like a lung mass. This situation most commonly is seen in patients with heart failure. The disappearance of the apparent mass when the heart failure is treated definitively establishes the diagnosis of pseudotumor (i.e., vanishing tumor). Heart failure is by far the most common cause of bilateral pleural effusion, but if car- diomegaly is not present, other causes such as malignancy should be investigated. Large effusions may opacify the entire hemithorax and displace mediastinal struc- tures toward the opposite side. More than The first step in the evaluation of patients with pleural effusion is to determine whether the effu- sion is a transudate or an exudate. An exudative effusion is diagnosed if the patient meets Light’s criteria. The serum to pleural fluid protein or albumin gradients may help better categorize the occasional transudate misidentified as an exudate by these criteria. If the patient has a transuda- tive effusion, therapy should be directed toward the underlying heart failure or cirrhosis. If the patient has an exudative effusion, attempts should be made to define the etiology. Pneumonia, cancer, tuberculosis, and pulmonary embolism account for most exudative effusions. Many pleural fluid tests are useful in the differential diagnosis of exudative effusions. Other tests help- ful for diagnosis include helical computed tomography and thoracoscopy. (Am Fam Physician 2006;73:1211-20. Copyright © 2006 American Academy of Family Physicians.) Downloaded from the American Family Physician Web site at www.aafp.org/afp. Copyright© 2006 American Academy of Family Physicians. For the private, noncommercial use of one individual user of the Web site. All other rights reserved. Contact copyrights@aafp.org for copyright questions and/or permission requests. 1212 American Family Physician www.aafp.org/afp Volume 73, Number 7 U April 1, 2006 SORT: KEY RECOMMENDATIONS FOR PRACTICE Clinical recommendations Evidence rating References Thoracentesis should be performed in all patients with more than a minimal pleural effusion unless clinically evident heart failure is present. C 5, 33 An effusion is exudative if it meets any of the following three criteria: (1) the ratio of pleural fluid protein to serum protein is greater than 0.5, (2) the pleural fluid lactate dehydrogenase (LDH) to serum LDH ratio is greater than 0.6, (3) pleural fluid LDH is greater than two thirds of the upper limit of normal for serum LDH. C 1, 5, 8 The serum-effusion protein or albumin gradients can be used to diagnose the presence of a transudate after diuresis. C 1, 9, 10 In a lymphocyte-predominant exudate, a pleural fluid adenosine deaminase greater than 40 U per L (667 nkat per L) indicates that the most likely diagnosis is tuberculosis. C 26-28 If malignancy is a concern and cytologic examination is nondiagnostic, thoracoscopy should be considered. C 5, 39, 40 A = consistent, good-quality patient-oriented evidence; B = inconsistent or limited-quality patient-oriented evidence; C = consensus, disease- oriented evidence, usual practice, expert opinion, or case series. For information about the SORT evidence rating system, see page 1135 or http://www.aafp.org/afpsort.xml. TABLE 1 Causes of Pleural Effusions: History, Signs, and Symptoms Condition Potential causes of the pleural effusion History Abdominal surgical procedures Postoperative pleural effusion, subphrenic abscess, pulmonary embolism Alcohol abuse or pancreatic disease Pancreatic effusion Artificial pneumothorax therapy Tuberculous empyema, pyothorax-associated lymphoma, trapped lung Asbestos exposure Mesothelioma, benign asbestos pleural effusion Cancer Malignancy Cardiac surgery or myocardial injury Pleural effusion secondary to coronary artery bypass graft surgery or Dressler’s syndrome Chronic hemodialysis Heart failure, uremic pleuritis Cirrhosis Hepatic hydrothorax, spontaneous bacterial empyema Childbirth Postpartum pleural effusion Esophageal dilatation or endoscopy Pleural effusion secondary to esophageal perforation Human immunodeficiency virus infection Pneumonia, tuberculosis, primary effusion lymphoma, Kaposi sarcoma Medication use Medication-induced pleural disease Remote inflammatory pleural process Trapped lung Rheumatoid arthritis Rheumatoid pleuritis, pseudochylothorax Superovulation with gonadotrophins Pleural effusion secondary to ovarian hyperstimulation syndrome Systemic lupus erythematosus Lupus pleuritis, pneumonia, pulmonary embolism Trauma Hemothorax, chylothorax, duropleural fistula Signs Ascites Hepatic hydrothorax, ovarian cancer, Meigs’ syndrome Dyspnea on exertion, orthopnea, peripheral edema, elevated jugular venous pressure Heart failure, constrictive pericarditis Pericardial friction rub Pericarditis Unilateral lower extremity swelling Pulmonary embolism Yellowish nails, lymphedema Pleural effusion secondary to yellow nail syndrome* Symptoms Fever Pneumonia, empyema, tuberculosis Hemoptysis Lung cancer, pulmonary embolism, tuberculosis Weight loss Malignancy, tuberculosis, anaerobic bacterial pneumonia *—Yellow nail syndrome results from an abnormality of lymphatics and consists of the triad of yellow nails, lymphedema, and pleural effusion. April 1, 2006 U Volume 73, Number 7 www.aafp.org/afp American Family Physician 1213 one half of these massive pleural effusions are caused by malignancy; other causes are complicated parapneumonic effusion, empy- ema, and tuberculosis.4 If the mediastinum is shifted toward the side of the effusion or is midline in a patient with a massive pleural effusion, either an endobronchial obstruction (e.g., lung cancer) or a mediastinum encase- ment by tumor (e.g., mesothelioma) should be considered. Thoracentesis Except for patients with obvious heart fail- ure, thoracentesis should be performed in all patients with more than a minimal pleural effusion (i.e., larger than 1 cm height on lateral decubitus radiograph, ultrasound, or CT) of unknown origin.5 In the context of heart failure, diagnostic thoracentesis is only indicated if any of the following atypical circumstances is present:1,5 (1) the patient is febrile or has pleuritic chest pain; (2) the patient has a unilateral effusion or effusions of markedly disparate size; (3) the effusion is not associated with cardiomegaly, or (4) the effusion fails to respond to management of the heart failure. Thoracentesis is urgent when it is sus- pected that blood (i.e., hemothorax) or pus (i.e., empyema) is in the pleural space, because immediate tube thoracostomy is indicated in these situations. If difficulty in obtaining pleural fluid is encountered because the effusion is small or loculated, ultrasound-guided thoracentesis minimizes the risk for iatrogenic pneumothorax.6 In most instances, analysis of the pleural fluid yields valuable diagnostic information or definitively establishes the cause of the pleu- ral effusion. This is the case when malignant cells, microorganisms, or chyle are found, or when a transudative effusion is found in the setting of heart failure or cirrhosis. Observing the gross appearance of the pleu- ral fluid may suggest a particular cause. For example, turbidity of the pleural fluid can be caused either by cells and debris (i.e., empy- ema) or by a high lipid level (i.e., chylothorax). A uniformly blood-stained fluid (i.e., hema- tocrit greater than 1 percent) narrows the differential diagnosis of the pleural effusion to malignancy, trauma (including recent cardiac surgery), pulmonary embolism, and pneu- monia.3,7 If the hematocrit of the pleural fluid exceeds half the simultaneous peripheral blood hematocrit, the patient has hemothorax. Although common, chest radiography is not necessary after thoracentesis unless air is obtained during the procedure; the patient develops symptoms such as dyspnea, cough, or chest pain; or tactile fremitus is lost over the upper part of the aspirated hemithorax.5 Pleural Effusion Figure 1A. Posteroanterior radiograph dem- onstrating blunting of the left costophrenic angle. Figure 1B. Left lateral decubitus of the same patient demonstrating a large amount of free pleural fluid. 1214 American Family Physician www.aafp.org/afp Volume 73, Number 7 U April 1, 2006 Pleural Effusion Analysis of Pleural Fluid Pleural effusions are either transudates or exudates based on the biochemical charac- teristics of the fluid, which usually reflect the physiologic mechanism of its formation. TRANSUDATIVE EFFUSIONS Transudates result from imbalances in hydrostatic and oncotic forces and are caused by a limited number of recognized clinical conditions such as heart failure and cirrho- sis. Less common causes include nephrotic syndrome, atelectasis, peritoneal dialysis, constrictive pericarditis, superior vena caval obstruction, and urinothorax. Transudative effusions usually respond to treatment of the underlying condition (e.g., diuretic therapy). EXUDATIVE EFFUSIONS In contrast, exudates occur when the local fac- tors influencing the accumulation of pleural fluid are altered. Exudates present more of a diagnostic dilemma. Pneumonia, malignancy, and thromboembolism account for most exu- dative effusions in the United States (Table 2).1 In clinical practice, exudative effusions can be separated effectively from transudative effusions using Light’s criteria. These criteria classify an effusion as exudate if one or more of the following are present: (1) the ratio of pleural fluid protein to serum protein is greater than 0.5, (2) the ratio of pleural fluid lactate dehydrogenase (LDH) to serum LDH is greater than 0.6, or (3) the pleural fluid LDH level is greater than two thirds of the upper limit of normal for serum LDH. Light’s criteria are nearly 100 percent sen- sitive at identifying exudates, but approxi- mately 20 percent of patients with pleural effusion caused by heart failure may fulfill the criteria for an exudative effusion after receiving diuretics.8 In these circumstances, if the difference between protein levels in the serum and the pleural fluid is greater than 3.1 g per dL, the patient should be classified as having a transudative effusion.9 A serum- effusion albumin gradient greater than 1.2 g per dL also can indicate that the pleu- ral effusion is most likely a true transuda- tive effusion.10 However, neither protein nor albumin gradients alone should be the pri- mary test used to distinguish transudative effusions from exudative effusions because they result in the incorrect classification of a significant number of exudates. This lower sensitivity may be caused by the fact that a single test is employed as opposed to TABLE 2 Leading Causes of Pleural Effusion in the United States* Cause Annual incidence Transudate Exudate Congestive heart failure 500,000 Yes No Pneumonia 300,000 No Yes Cancer 200,000 No Yes Pulmonary embolism 150,000 Sometimes Sometimes Viral disease 100,000 No Yes Coronary-artery bypass surgery 60,000 No Yes Cirrhosis 50,000 Yes No *−Based on analysis of patients subjected to thoracentesis. Reprinted with permission from Light RW. Clinical practice. Pleural effusion. N Engl J Med 2002;346:1971. TABLE 3 Likelihood of Exudates Using the Pleural Fluid to Serum Protein Ratio Pleural fluid to serum protein ratio Likelihood ratio* Probability of exudate with overall risk of 10% (%) Probability of exudate with overall risk of 30% (%) r0.71 93.03 91.0 98.0 0.66 to 0.70 31.81 78.0 93.0 0.61 to 0.65 4.24 32.0 64.0 0.56 to 0.60 3.58 29.0 61.0 0.51 to 0.55 1.50 14.0 39.0 0.46 to 0.50 0.48 5.3 18.0 0.41 to 0.45 0.27 3.2 11.0 0.36 to 0.40 0.15 2.2 7.9 0.31 to 0.35 0.07 1.1 4.1 b0.30 0.04 0.0 0.0 * A likelihood ratio of 1 does not change the likelihood of disease. Likelihood ratios above 1 increase the risk of disease and likelihood ratios below 1 reduce the risk of disease (see http://www.aafp.org/afppoems.xml). Adapted with permission from Heffner JE, Highland K, Brown LK. A meta-analysis derivation of continuous likelihood ratios for diagnosing pleural fluid exudates. Am J Respir Crit Care Med 2003;167:1595. April 1, 2006 U Volume 73, Number 7 www.aafp.org/afp American Family Physician 1215 Pleural Effusion the three-test combination of the standard criteria described above. Another approach to the classification of pleural effusions is to apply continuous or multilevel likelihood ratios (Table 311). FURTHER TESTING FOR EXUDATES In patients with exudative effusion, the fol- lowing pleural fluid tests should be per- formed on fluid obtained during the initial thoracentesis: cell counts and differential, glucose, adenosine deaminase (ADA), and cytologic analysis. Bacterial cultures and pH should be tested if infection is a concern12 (Tables 43,5,13 and 55,13-24). Pleural fluid for total white blood cell (WBC) count and differential cell count should be sent in an anticoagulated tube. TABLE 4 Routine Pleural Fluid Tests for Pleural Effusion Test Test value Suggested diagnosis Comments Adenosine deaminase (ADA) >40 U per L (667 nkat per L) Tuberculosis (>90 percent), empyema (60 percent), complicated parapneumonic effusion (30 percent), malignancy (5 percent), rheumatoid arthritis5 In the United States, ADA is not routinely requested because of the low prevalence of tuberculous pleurisy. Cytology Present Malignancy Actively dividing mesothelial cells can mimic an adenocarcinoma. Glucose <60 mg per dL (3.3 mmol per L) Complicated parapneumonic effusion or empyema, tuberculosis (20 percent), malignancy (<10 percent), rheumatoid arthritis5 In general, pleural fluids with a low glucose level also have low pH and high LDH levels. Lactate dehydrogenase (LDH) >Two thirds of upper limits of normal for serum LDH Any condition causing an exudate Very high levels of pleural fluid LDH (>1,000 U per L) typically are found in patients with complicated parapneumonic pleural effusion and in about 40 percent of those with tuberculous pleurisy.5 LDH fluid to serum ratio >0.6 Any condition causing an exudate Most patients who meet the criteria for an exudative effusion with LDH but not with protein levels have either parapneumonic effusions or malignancy.3 Protein fluid to serum ratio >0.5 Any condition causing an exudate A pleural fluid protein level >3 mg per dL suggests an exudate, but when taken alone this parameter misclassifies more than 10 percent of exudates and 15 percent of transudates.13 Red blood cell count >100,000 per mm3 (100  106 per L) Malignancy, trauma, parapneumonic effusion, pulmonary embolism A fluid hematocrit <1 percent is nonsignificant.13 White blood cell count and differential >10,000 per mm3 (10  109 per L) Empyema, other exudates (uncommon) In purulent fluids, leukocyte count is commonly much lower than expected because dead cells or other debris account for much of the turbidity. Eosinophils >10 percent Not diagnostic The presence of air or blood in the pleural space is a common cause. No diagnosis is ever obtained in as many as one third of patients with eosinophilic pleural effusion.3 Lymphocytes >50 percent Malignancy, tuberculosis, pulmonary embolism, coronary artery bypass surgery Pleural fluid lymphocytosis >90 percent suggests tuberculosis or lymphoma. Neutrophils >50 percent Parapneumonic effusion, pulmonary embolism, abdominal diseases In about 7 percent of acute tuberculous pleurisy and 20 percent of malignant pleural effusions, a neutrophilic fluid predominance can be seen.5 Information from references 3, 5, and 13. 1216 American Family Physician www.aafp.org/afp Volume 73, Number 7 U April 1, 2006 Pleural Effusion If the fluid is sent in a plastic or glass tube without anticoagulation, the fluid may clot, resulting in an inaccurate count.25 The pre- dominant WBC population is determined by the mechanism of pleural injury and the timing of the thoracentesis in relation to the onset of the injury. Thus, the finding of neutrophil-rich fluid heightens suspicion for parapneumonic pleural effusion (an acute process), whereas a lymphocyte-predomi- nant fluid profile suggests cancer or tuber- culosis (a chronic process). Pleural fluid for pH testing should be col- lected anaerobically in a heparinized syringe andmeasuredinablood-gasmachine.3 Frank pus should not be sent for pH determination because thick, purulent fluid may clog the blood-gas machine. A low pleural fluid pH TABLE 5 Optional Pleural Fluid Tests for Pleural Effusion Test Test value Suggested diagnosis Comments Amylase >Upper limit of normal Malignancy (<20 percent), pancreatic disease, esophageal rupture5,16 Obtain when esophageal rupture or pancreatic disease is suspected. The amylase in malignancy and esophageal rupture is of the salivary type. Cholesterol >45 to 60 mg per dL (1.16 to 1.55 mmol per L) Any condition causing an exudate Measure if chylothorax or pseudochylothorax is suspected. This parameter taken alone misclassifies 10 percent of exudates and 20 percent of transudates.13 Culture Positive Infection Obtain in all parapneumonic pleural effusions because a positive Gram stain or culture should lead to prompt chest tube drainage.14,15 Hematocrit fluid to blood ratio r0.5 Hemothorax Obtain when pleural fluid is bloody. Hemothorax most often originates from blunt or penetrating chest trauma. Interferon* Different cutoff points Tuberculosis17 Consider when ADA is unavailable or nondiagnostic and tuberculosis is suspected. NT-proBNP >1,500 pg per mL Heart failure18 If available, consider testing when heart failure is suspected and exudate criteria are met.19 pH <7.20 Complicated parapneumonic effusion or empyema, malignancy (<10 percent), tuberculosis (<10 percent), esophageal rupture5 Obtain in all nonpurulent effusions if infection is suspected. A low pleural fluid pH indicates the need for tube drainage only for parapneumonic pleural effusions. Polymerase chain reaction† Positive Infection20,21 Consider when infection is suspected. Sensitivity of polymerase chain reaction to detect Mycobacterium tuberculosis in pleural fluid varies from 40 to 80 percent and is lower in patients with negative mycobacterial cultures. Triglycerides >110 mg per dL (1.24 mmol per L) Chylothorax Obtain when pleural fluid is cloudy or milky. Chylothorax is caused by lymphoma or trauma. Not all chylous pleural effusions appear milky white or whitish. Tumor markers‡ Different cutoff points Malignancy Consider when malignancy is suspected and thoracoscopy is being considered. Except for telomerase activity,22 individual tests tend to have low sensitivity (< 30 percent) when looking for the utmost specificity.23,24 ADA = adenosine deaminase; NT-proBNP = N-terminal pro-b-type natriuretic peptide. *—ADA measurement is cheaper, easier, and quicker to perform than interferon for diagnosing tuberculosis. †—For example, Streptococcus pneumoniae and Mycobacterium tuberculosis. ‡—For example, carcinoembryonic antigen (CEA), CA 15.3 and CA 549 (markers for breast carcinoma), CYFRA 21-1 (marker for lung carcinoma), CA 125 (marker for ovarian and endometrial carcinoma), human epidermal growth factor receptor (HER-2/neu) gene amplification, telomerase. Information from references 5 and 13 through 24. April 1, 2006 U Volume 73, Number 7 www.aafp.org/afp American Family Physician 1217 Pleural Effusion value has prognostic and therapeutic impli- cations for patients with parapneumonic and malignant pleural effusions. A pH value less than 7.20 in a patient with a parapneumonic effusion indicates the need to drain the fluid.14,15 In a patient with malignant pleural effusion, a pleural fluid pH value less than 7.30 is associated with a shorter survival and poorer response to chemical pleurodesis.1 When a pleural fluid pH value is not avail- able, a pleural fluid glucose concentration less than 60mg per dL can be used to identify complicated parapneumonic effusions.14 ADA is an enzyme that plays an impor- tant role in lymphoid cell differentiation. A pleural fluid ADA level greater than 40 U per L (667 nkat per L) has a sensitivity of 90 to 100 percent and a specificity of 85 to 95 percent for the diagnosis of tuberculous pleurisy.3,5,26-28 The specificity rises above 95 percent if only lymphocytic exudates are considered.29,30 In areas where the preva- lence of tuberculosis is low, the positive pre- dictive value of pleural ADA declines but the negative predictive value remains high. Cultures for both aerobic and anaero- bic bacteria will identify the responsible microorganism in about 40 percent of para- pneumonic effusions (70 percent if fluid is grossly purulent).5 The yield with culture is increased if blood-culture bottles are inocu- lated at the bedside with the pleural fluid. In addition, both pleural fluid and sputum should be cultured for mycobacteria when tuberculous pleuritis is suspected. The yield of sputum cultures in tuberculous pleu- ral effusion varies from 10 to 60 percent, largely dependent on the extent of associated pulmonary involvement.31 Because delayed hypersensitivity plays a major role in the pathogenesis of tuberculous pleuritis, it is not possible to isolate Mycobacterium tuber- culosis from pleural fluid samples in more than 60 to 70 percent of patients.5,26 The use of broth medium (e.g., BACTEC radio- metric system) with bedside inoculation provides higher yields and faster results (one to two weeks) than conventional methods.32 Smears of the pleural fluid for mycobacteria are rarely positive (5 percent)5 unless the patient has a tuberculous empyema. About one third of patients with tuberculous pleu- ritis have a negative tuberculin skin test.26 Cytology is positive in approximately 60 percent of malignant pleural effusions.33 Negative test results are related to factors such as the type of tumor (e.g., commonly negative with mesothelioma, sarcoma, and lymphoma); the tumor burden in the pleu- ral space; and the expertise of the cytolo- gist. The diagnostic yield may be somewhat improved by additional pleural taps. Sub- mission of 10 mL of pleural fluid appears adequate for cytologic processing.34 A second thoracentesis should be consid- ered in the following situations: (1) suspected malignant effusion and the initial pleural fluid cytologic examination is negative; (2) a parapneumonic effusion with borderline biochemical characteristics of the pleural fluid for indicating chest tube drainage; and (3) suspected acute tuberculous pleurisy with initial nondiagnostic pleural ADA levels. Other diagnostic procedures IMAGING TECHNIQUES Helical CT has become the first-line modal- ity for imaging of pulmonary circulation in a patient suspected of having pulmonary embolism, supplanting ventilation-perfusion scintigraphy. Helical CT also can identify alternative explanations for the pleural effu- sion, can diagnose deep venous thrombosis when combined with CT venography of the pelvis and lower extremities, and can distin- guishmalignant from benign pleural disease. CT findings suggestive of malignant disease are the presence of pleural nodules or nodu- lar pleural thickening (Figure 2), circumfer- ential or mediastinal pleural thickening, or infiltration of the chest wall or diaphragm. However, a recent study35 found that this was true in less than 20 percent of patients with malignant pleural effusion. Positron emis- sion tomography seems promising for dif- ferentiating between benign and malignant pleural diseases (sensitivity 97 percent and specificity 88.5 percent in one study).36 BRONCHOSCOPY Bronchoscopy is useful whenever an endo- bronchial malignancy is likely, as suggested 1218 American Family Physician www.aafp.org/afp Volume 73, Number 7 U April 1, 2006 Pleural Effusion by one or more of the following characteris- tics: a pulmonary infiltrate or a mass on the chest radiograph or CT scan, hemoptysis, a massive pleural effusion, or shift of the medi- astinum toward the side of the effusion. PERCUTANEOUS PLEURAL BIOPSY Closed-needle biopsy of the pleura for his- tologic examination classically has been recommended for undiagnosed exudative effusions when tuberculosis or malignancy is suspected. The combination of histology (80 percent sensitivity) and culture (56 per- cent sensitivity) of pleural biopsy tissue estab- lishes the diagnosis of tuberculosis in up to 90 percent of patients.3,5 However, this diag- nosis is strongly suggested by a high ADA level in the pleural fluid, as detailed above, thus avoiding the need for a confirmatory biopsy in most patients. Cytology is superior to blind pleural biopsy for the diagnosis of pleural malignancy. In one case series,37 needle biopsy of the pleura was positive in only 17 percent (20 of 119) of patients with malignancy involving the pleura but a negative pleural fluid cytology. The diagnostic yield from pleural biopsy is higher when it is used with some form of image guidance to identify areas of particular thickening or nodularity.38 THORACOSCOPY Because thoracoscopy is diagnostic in more than 90 percent of patients with pleural malignancy and negative cytology, it is the preferred diagnostic procedure in patients with cytology-negative pleural effusion who are suspected of having pleural malignancy.39 Moreover, thoracoscopy offers the pos- sibility of effective pleurodesis during the procedure. Criteria for Referral No diagnosis is ever established for approxi- mately 15 percent of patients.1 Observation is probably the best option if the patient is improving and there are no parenchymal infiltrates or pleural nodules, because most pleural effusions that are undiagnosed after a thorough initial evaluation are benign.40,41 Figure 3 is a suggested algorithm for the investigation of pleural effusions. Pulmonary consultation should be obtained when thoracentesis is technically difficult; the etiology is uncertain after ini- tial thoracentesis; or drainage of the pleural space is advised (e.g., symptomatic large or massive pleural effusion, hemothorax, empyema, or complicated parapneumonic effusion). The Authors JOSÉ M. PORCEL, M.D., F.A.C.P., F.C.C.P., is professor of Medicine and vice dean at the Faculty of Medicine, University of Lleida, Spain. He also is chief of the Internal Medicine Department at the Arnau de Vilanova University Hospital, Lleida. Dr. Porcel received his medical degree from the Autonomous University of Barcelona, Spain and completed a residency in internal medicine at Vall d’Hebron University Hospital, Barcelona. He was a research fellow in immunology at King’s College Hospital, London, England. RICHARD W. LIGHT, M.D., F.C.C.P., is a professor in the Department of Medicine, Vanderbilt University, Nashville, Tenn., and director of the Pulmonary Disease Program at Saint Thomas Hospital, Nashville. He received his medical degree from Johns Hopkins University School of Medicine, Baltimore, Md., and completed a medical residency and a pulmonary fellowship at Johns Hopkins Hospital. Address correspondence to José M. Porcel, M.D., Arnau de Vilanova University Hospital, Alcalde Rovira Roure 80, 25198 Lleida, Spain (e-mail: jporcelp@yahoo.es). Reprints are not available from the authors. Author disclosure: Nothing to disclose. Figure 2. Axial computed tomography scan of a patient with multiple nodules in diaphragmatic pleura from metastatic breast cancer. April 1, 2006 U Volume 73, Number 7 www.aafp.org/afp American Family Physician 1219 Pleural Effusion REFERENCES 1. Light RW. Clinical practice. Pleural effusion. N Engl J Med 2002;346:1971-7. 2. Efrati O, Barak A. Pleural effusions in the pediatric population. Pediatr Rev 2002;23:417-26. 3. Light RW. Pleural diseases. 4th ed. Philadelphia: Lippincott Williams & Wilkins, 2001. 4. Porcel JM, Vives M. Etiology and pleural fluid charac- teristics of large and massive effusions. Chest 2003; 124:978-83. 5. Porcel JM, Light RW. Thoracentesis. PIER, American Col- lege of Physicians, 2004. Accessed online October 28, 2004, at: http://pier.acponline.org. Subscription required. 6. Jones PW, Moyers JP, Rogers JT, Rodriguez RM, Lee YC, Light RW. Ultrasound-guided thoracentesis: is it a safer method? Chest 2003;123:418-23. 7. Villena V, López-Encuentra A, García-Luján R, Echave- Sustaeta J, Martínez CJ. Clinical implications of appear- Evaluation of Patients with Pleural Effusion Figure 3. Algorithm for the evaluation of patients with pleural effusion. (CT = computed tomog- raphy; LDH = lactate dehydrogenase; ADA = adenosine deaminase.) Pleural effusion (>1 cm height on decubitus radiograph, ultrasound or CT) without clinically evident heart failure Perform thoracentesis; measure pleural fluid protein and LDH. Potential diagnosis (e.g., empyema, hemothorax, chylothorax) Are any of the following met? • Pleural to serum protein ratio >0.5 • Pleural to serum LDH ratio >0.6 • Pleural LDH > two thirds upper limit of normal serum LDH Additional pleural fluid testing (cultures, hematocrit, triglycerides) No Yes Exudate. Further diagnostic procedures: Transudate. Treat heart failure, cirrhosis, or nephrosis. ADA > 40 U per L (667 nkat per L) and lymphocytic effusion Obtain pleural fluid glucose; ADA; total and differential cell counts; cytologic analysis; and, if suspected infection, pH and cultures. Pleural fluid amylase Suspected pancreatic pleural effusion or esophageal rupture Consider antituberculous treatment. No diagnosis Positive helical CT: pulmonary embolism confirmed Perform helical chest CT. No diagnosis Consider bronchoscopy if hemoptysis, atelectasis, or pulmonary infiltrates are present. Symptoms improving? Consider pleural biopsy (blind, image-guided, or by thoracoscopy). Observe No Yes 1220 American Family Physician www.aafp.org/afp Volume 73, Number 7 U April 1, 2006 Pleural Effusion ance of pleural fluid at thoracentesis. Chest 2004; 125:156-9. 8. Porcel JM, Vives M, Vicente de Vera MC, Cao G, Rubio M, Rivas MC. Useful tests on pleural fluid that distin- guish transudates from exudates. Ann Clin Biochem 2001;38:671-5. 9. Romero-Candeira S, Fernández C, Martín C, Sánchez- Paya J, Hernández L. Influence of diuretics on the con- centration of proteins and other components of pleural transudates in patients with heart failure. Am J Med 2001;110:681-6. 10. Romero-Candeira S, Hernández L. The separation of transudates and exudates with particular reference to the protein gradient. Curr Opin Pulm Med 2004;10:294-8. 11. Heffner JE, Highland K, Brown LK. A meta-analysis derivation of continuous likelihood ratios for diagnos- ing pleural fluid exudates. Am J Respir Crit Care Med 2003;167:1591-9. 12. Barnes TW, Olson EJ, Morgenthaler TI, Edson RS, Decker PA, Ryu JH. Low yield of microbiologic studies on pleural fluid specimens. Chest 2005;127:916-21. 13. Sahn SA, Heffner JE. Pleural fluid analysis. In: Light RW, Gary Lee YC. Textbook of pleural diseases. London: Arnold, 2003:191-209. 14. Colice GL, Curtis A, Deslauriers J, Heffner J, Light R, Littenberg B, et al. Medical and surgical treatment of parapneumonic effusions: an evidence-based guide- line. Chest 2000;118:1158-71. 15. Davies CW, Gleeson FV, Davies RJ, Pleural Diseases Group, Standards of Care Committee, British Thoracic Society. BTS guidelines for the management of pleural infection. Thorax 2003;58(suppl 2):ii18-28. 16. Villena V, Pérez V, Pozo F, López-Encuentra A, Echave- Sustaeta J, Arenas J, et al. Amylase levels in pleural effu- sions: a consecutive unselected series of 841 patients. Chest 2002;121:470-4. 17. Villena V, López-Encuentra A, Pozo F, Echave-Sustaeta J, Ortuño-de-Solo B, Estenoz-Alfaro J, et al. Interferon gamma levels in pleural fluid for the diagnosis of tuber- culosis. Am J Med 2003;115:365-70. 18. Porcel JM, Vives M, Cao G, Esquerda A, Rubio M, Ribas MC. Measurement of pro-brain natriuretic peptide in pleural fluid for the diagnosis of pleural effusions due to heart failure. Am J Med 2004;116:417-20. 19. Porcel JM. The use of probrain natriuretic peptide in pleural fluid for the diagnosis of pleural effusions resulting from heart failure. Curr Opin Pulm Med 2005;11:329-33. 20. Falguera M, López A, Nogués A, Porcel JM, Rubio- Caballero M. Evaluation of the polymerase chain reac- tion method for detection of Streptococcus pneumoniae DNA in pleural fluid samples. Chest 2002;122:2212-6. 21. Pai M, Flores LL, Hubbard A, Riley LW, Colford JM Jr. Nucleic acid amplification tests in the diagnosis of tuberculous pleuritis: a systematic review and meta- analysis. BMC Infect Dis 2004;4:6. 22. Dikmen G, Dikmen E, Kara M, Sahin E, Dogan P, Özdemir N. Diagnostic implications of telomerase activ- ity in pleural effusions. Eur Respir J 2003;22:422-6. 23. Porcel JM, Vives M, Esquerda A, Salud A, Pérez B, Rodríguez-Panadero F. Use of a panel of tumor markers (carcinoembryonic antigen, cancer antigen 125, carbo- hydrate antigen 15-3, and cytokeratin 19 fragments) in pleural fluid for the differential diagnosis of benign and malignant effusions. Chest 2004;126:1757-63. 24. Porcel JM, Salud A, Vives M, Esquerda A, Rodriguez- Panadero F. Soluble oncoprotein 185(HER-2) in pleural fluid has limited usefulness for the diagnostic evaluation of malignant effusions. Clin Biochem 2005;38:1031-3. 25. Conner BD, Lee YC, Branca P, Rogers JT, Rodriguez RM, Light RW. Variations in pleural fluid WBC count and differential counts with different sample containers and different methods. Chest 2003;123:1181-7. 26. Porcel JM, Vives M. Differentiating tuberculous from malignant pleural effusions: a scoring model. Med Sci Monit 2003;9:CR175-80. 27. Goto M, Noguchi Y, Koyama H, Hira K, Shimbo T, Fukui T. Diagnostic value of adenosine deaminase in tuberculous pleural effusion: a meta-analysis. Ann Clin Biochem 2003;40:374-81. 28. Greco S, Girardi E, Masciangelo R, Capoccetta GB, Saltini C. Adenosine deaminase and interferon gamma measurements for the diagnosis of tuberculous pleu- risy: a meta-analysis. Int J Tuberc Lung Dis 2003;7: 777-86. 29. Porcel JM, Vives M. Adenosine deaminase levels in nontuberculous lymphocytic pleural effusions. Chest 2002;121:1379-80. 30. Jiménez Castro D, Díaz Nuevo G, Pérez-Rodríguez E, Light RW. Diagnostic value of adenosine deaminase in nontuberculous lymphocytic pleural effusions. Eur Respir J 2003;21:220-4. 31. Davies PD. Tuberculous pleuritis. In: Bouros D, ed. Pleu- ral disease. New York: Marcel Dekker, 2004:677-97. 32. Hasaneen NA, Zaki ME, Shalaby HM, El-Morsi AS. Polymerase chain reaction of pleural biopsy is a rapid and sensitive method for the diagnosis of tuberculous pleural effusion. Chest 2003;124:2105-11. 33. Maskell NA, Butland RJ, Pleural Diseases Group, Stand- ards of Care Committee, British Thoracic Society. BTS guidelines for the investigation of a unilateral pleural effusion in adults. Thorax 2003;58(suppl 2):ii8-17. 34. Sallach SM, Sallach JA, Vasquez E, Schultz L, Kvale P. Volume of pleural fluid required for diagnosis of pleural malignancy. Chest 2002;122:1913-7. 35. Arenas-Jiménez J, Alonso-Charterina S, Sánchez-Paya J, Fernández-Latorre F, Gil-Sanchez S, Lloret-Llorens M. Evaluation of CT findings for diagnosis of pleural effu- sions. Eur Radiol 2000;10:681-90. 36. Duysinx B, Nguyen D, Louis R, Cataldo D, Belhocine T, Bartsch P, et al. Evaluation of pleural disease with 18-fluorodeoxyglucose positron emission tomography imaging. Chest 2004;125:489-93. 37. Prakash UB, Reiman HM. Comparison of needle biopsy with cytologic analysis for the evaluation of pleu- ral effusion: analysis of 414 cases. Mayo Clin Proc 1985;60:158-64. 38. Maskell NA, Gleeson FV, Davies RJ. Standard pleural biopsy versus CT-guided cutting-needle biopsy for diagnosis of malignant disease in pleural effusions: a randomised controlled trial. Lancet 2003;361:1326-30. 39. Antony VB, Loddenkemper R, Astoul P, Boutin C, Gold- straw P, Hott J, et al. Management of malignant pleural effusions. Eur Respir J 2001;18:402-19. 40. Light RW. Update: management of the difficult to diag- nose pleural effusion. Clin Pulm Med 2003;10:39-46. 41. Venekamp LN, Velkeniers B, Noppen M. Does “idio- patic pleuritis” exist? Natural history of non-specific pleuritis diagnosed after thoracoscopy. Respiration 2005;72:74-8.