Recommended Approach to a Pleural Effusion on Chest X-ray

Pulmonary Critical CareDIAGNOSIS OFPLEURAL EFFUSION:A SYSTEMATIC APPROACHBy Emmet E. McGrath, MB, PhD, MRCPI, and Paul B. Anderson, MA, FRCPC E 1.0 HourNotice <strong>to</strong> CE enrollees:A closed-book, multiple-choice examinationfollowing this article tests your under standing ofthe following objectives:1. Identify 3 characteristics of abnormal pleuralfluid.2. Describe a systematic algorithm for the investigationof pleural effusions.3. Differentiate between extudates and transdates.To read this article and take the CE test online,visit www.ajcconline.org and click “CE Articlesin This Issue.” No CE test fee for AACN members.©2011 American Association of Critical-Care Nursesdoi: 10.4037/ajcc2011685In most diseases related <strong>to</strong> pleural effusion, the fluid analysisyields important diagnostic information, and in certain cases,fluid analysis alone is enough for diagnosis.The many importantcharacteristics of pleural fluid are described, as are othercomplementary investigations that can assist with the diagnosisof common and rare pleural effusions. For a systematicreview of pleural effusion, a literature search for articles onthe practical investigation and diagnosis of pleural effusionwas done. Articles included guidelines, expert opinion, experimentaland nonexperimental studies, literature reviews, andsystematic reviews published from May 2003 through June2009.The search yielded 1 guideline, 2 meta-analyses, 9 literaturereviews, 1 randomized control trial, and 9 clinical studies.On the basis of class IIa or class I evidence from these articles,a step by step approach is recommended for investigating apleural effusion, beginning with assessment of the medicalhis<strong>to</strong>ry, clinical examination, radiology, pleural fluid evaluation,and finally, if no diagnosis is forthcoming, a pleuralbiopsy under image guidance or thoracoscopy. (AmericanJournal of Critical Care. 2011;20:119-128)www.ajcconline.org AJCC AMERICAN JOURNAL OF CRITICAL CARE, March 2011, Volume 20, No. 2 119

In a normal pleural space, fluid enters and exits at a constant, equal rate because of theongoing filtration of a small amount of low-protein liquid in normal microvessels. Nearthe end of the 19th century, Starling and Tubby 1 hypothesized that the exchange of microvascularfluid and solutes was governed by the balance between hydrostatic pressure,osmotic pressure, and membrane permeability, and they devised the Starling equation:QF = LP × A[(PCAP - PPL) - ζD(piCAP - piPL)]where QF is fluid movement, LP is the filtration coefficient, A is the surface area of the pleura,ζD is the reflection coefficient for protein movement across the pleura (PL), P is the hydrostaticpressure of the pulmonary capillary bed (CAP), and pi is the oncotic pressure of pleural space. 1,2This equation formed the basis for understanding fluid accumulation in the pleural space,where the hydrostatic forces that filter water out of the vessel are balanced by osmotic forcesthat reabsorb water back in<strong>to</strong> the vessel. In the pleura, reabsorption is facilitated by the extensivelymphatic system on the diaphragm and mediastinal surfaces of the parietal pleura. 2The most commonconditions causingeffusions arecardiac failure,pneumonia,and malignantneoplasm.Fluid accumulation in the pleural space indicatesdisease. The accumulation is associated with manymedical conditions that predispose <strong>to</strong> fluid accumulationvia many different mechanisms, includingincreased pulmonary capillary pressure, decreasedoncotic pressure, increased pleuralmembrane permeability, and obstructionof lymphatic flow. 3The most common conditions thatresult in effusions are cardiac failure,pneumonia, and malignant neoplasm.Diagnosis of a pleural effusion beginswith obtaining the patient’s clinical his<strong>to</strong>ryand doing a physical examinationand is followed by chest radiographyand analysis of pleural fluid in appropriateinstances. If necessary, theprocess continues with further investigativestudies, such as computed <strong>to</strong>mography (CT)of the thorax, pleural biopsy, thoracoscopy, and,occasionally, bronchoscopy.MethodsArticles used for this review were selected byusing MEDLINE <strong>to</strong> search for English-language articleson pleural effusion published from May 2003About the AuthorsEmmet E. McGrath was a clinical lecturer in respira<strong>to</strong>rymedicine and Paul B. Anderson was a consultant respira<strong>to</strong>ryphysician in the Department of Respira<strong>to</strong>ry Medicine,Northern General Hospital, Sheffield, England at the timethis article was written.Corresponding author: Dr Emmet McGrath, Senior Lecturerin Respira<strong>to</strong>ry Medicine, Department of Respira<strong>to</strong>ryMedicine, Birmingham Heartlands Hospital, Birmingham,United Kingdom (e-mail: e.e.mcgrath@bham.ac.uk).through June 2009. In addition, hand searching ofthe bibliographies of retrieved articles was done byrecognized world authorities on pleural effusions.Keywords used for the search were pleural effusion,investigation, and diagnosis. All types of evidence(guidelines, expert opinion, experimental and nonexperimentalstudies, literature reviews, and systematicreviews) in English were examined, but onlymaterial that involved the practical and clinical investigationand diagnosis of pleural effusion in adultswas included in the final review. Case reports wereexcluded. Important articles by world leading expertscited in articles from the search were also reviewed.ResultsAlthough many articles have been written onpleural effusion, the number included in our reviewwas markedly reduced when only English-languagepublications about the clinical and practical investigationand diagnosis of this abnormality weresought by using the key words investigation anddiagnosis <strong>to</strong>gether. The number was also reducedfurther when case reports were excluded. From May2003 through June 2009, a <strong>to</strong>tal of 1 guideline, 3 2meta-analyses, 4,5 9 literature reviews, 6-14 1 randomizedcontrolled trial, 15 and 9 clinical studies 16-24 on thepractical investigation and diagnosis of pleural effusionwere published in peer-reviewed journals.Evidence was assessed as class I, class IIa, classIIb, indeterminate, and class III. 25 Class I evidence isdefinite recommendation of a practice that is basedon the results of at least 1 randomized controlledtrial and is considered a definitive standard of careor best practice. Good evidence and expert opinionare considered class II. Fair <strong>to</strong> good evidence is classifiedas class IIb. A finding of no benefit or harm120 AJCC AMERICAN JOURNAL OF CRITICAL CARE, March 2011, Volume 20, No. 2 www.ajcconline.org

is considered indeterminate; usually evidence isclassified as indeterminate because it is insufficient(eg, evidence provided by preliminary data). ClassIII evidence is evidence that is not useful and maybe harmful.Clinical Features of <strong>Pleural</strong> <strong>Effusion</strong>sThe clinical his<strong>to</strong>ry and physical examinationcan be quite helpful in indicating appropriate investigation.Patients with pleural effusions usually havedyspnea, cough, and occasional sharp nonradiatingchest pain that is often pleuritic. A his<strong>to</strong>ry of cardiac,renal, or liver impairment can suggest a transudativeeffusion. A his<strong>to</strong>ry of cancer can besuggestive of a malignant pleural effusion. Recentleg swelling or deep-vein thrombosis may result inan effusion related <strong>to</strong> pulmonary embolism. 6 A his<strong>to</strong>ryof recent or current pneumonia suggests a parapneumoniceffusion, either complicated (empyema) oruncomplicated. Previous trauma may result in hemothoraxor chylothorax. Previous exposure <strong>to</strong> asbes<strong>to</strong>s iscommon in patients who have a benign effusion related<strong>to</strong> the exposure or have mesothelioma. 6 Recentesophageal dilatation or endoscopy can result inesophageal rupture. Certain medications, includingamiodarone, methotrexate, pheny<strong>to</strong>in, and nitrofuran<strong>to</strong>in,can cause pleural effusions. 3 Rheuma<strong>to</strong>idarthritis and other au<strong>to</strong>immune conditions canalso result in effusions.A sign such as hemoptysis may be associatedwith a malignant neoplasm, pulmonary embolism,or severe tuberculosis. Fever occurs in tuberculosis,empyema, and pneumonia. Weight loss can be associatedwith a malignant neoplasm and tuberculosis.Physical findings such as ascites may indicatecirrhosis, ovarian cancer, or Meig syndrome. 6 Unilateralleg swelling can strongly indicate pulmonaryembolism, and bilateral leg swelling is associatedwith transudates such as those caused by heart or liverfailure. A pericardial friction rub occurs in pericarditis.In general, the clinician moves from suspecting effusionon the basis of clinical his<strong>to</strong>ry and examination<strong>to</strong> proving an effusion exists by means of chest radiographybefore fluid sampling is considered.For a unilateral pleural effusion evident on chestradiographs, the differential diagnosis is extensive(Table 1). The differential diagnosis for bilateralpleural effusions is narrower and includes causes oftransudative effusions, such as cardiac, hepatic, andrenal failure and hypoalbuminemia, and in rare cases,malignant neoplasm, pulmonary embolism, andrheuma<strong>to</strong>id arthritis. Congestive heart failure is themost common cause of bilateral pleural effusion,and in patients with clinical or radiological evidenceTable 1Causes of transudative andexudative unilateral effusion aTypeCommonLess commonRareExudateParapneumonic effusionsMalignant neoplasmPulmonary embolismRheuma<strong>to</strong>id arthritisBenign effusion associatedwith exposure <strong>to</strong> asbes<strong>to</strong>sPancreatitisAfter myocardial infarctionsyndromeAu<strong>to</strong>immune diseasesAfter coronary artery bypasssurgerySubphrenic, hepatic, orsplenic abscessUremiaChylothoraxOther infectionsDrug inducedRadiotherapyEsophageal rupturea Based on data from Maskell et al. 3of congestive heart failure, investigation of the effusionneed not go any further. 3 Diagnostic thoracentesisis required only if a patient has bilateral effusionsthat are unequal in size, has an effusion that doesnot respond <strong>to</strong> therapy, has pleuriticchest pain, or is febrile. 6 The effusionsusually improve quite quickly oncediuretic therapy is started.Diagnostic ThoracentesisThoracentesis with analysis ofthe fluid can quickly narrow the differentialdiagnosis of an effusion.Most aspirates consist of a straw-yellowfluid; this finding is nonspecificbecause it occurs in many types of pleural effusion.However, other appearances can be helpful (Table 2).The fluid may be bloodstained in conditions suchas pneumonia, malignant neoplasm, pulmonaryTransudateLeft ventricular failureCirrhotic liver diseaseHypoalbuminemiaPeri<strong>to</strong>neal dialysisNephrotic syndromePulmonary embolismHypothyroidismMitral stenosisConstrictive pericarditisMeig syndromeUrinothoraxSuperior vena cavaobstructionOvarian hyperstimulationSymp<strong>to</strong>ms includedyspnea, cough,and occasionallysharp, nonradiating,pleuriticchest pain.www.ajcconline.org AJCC AMERICAN JOURNAL OF CRITICAL CARE, March 2011, Volume 20, No. 2 121

Table 2Relationship between pleural fluidappearance and causes aCausePseudochylothorax and chylothoraxUrinothoraxAnaerobic empyemaChylothoraxAspergillus infectionEmpyemaAmebic liver abscessEsophageal ruptureTrauma, pulmonary embolism, benignasbes<strong>to</strong>s-related effusion, pneumonia,malignant neoplasm, aftermyocardial infarction syndromea Based on data from Maskell et al. 3Figure 1 Milky pleural aspirate.Table 3Light’s criteria for distinguishing betweenpleural exudates and transudatesFluid appearance/odorMilky whiteUrinePutridBile stainedBlackTurbid“Anchovy” brownFood particlesBloodstainedFluid is an exudate if 1 or more of the following criteria are met1. Ratio of pleural fluid level of lactate dehydrogenase (LDH) <strong>to</strong> serumlevel of LDH is greater than 0.62. <strong>Pleural</strong> fluid level of LDH is more than two-thirds the upper limit ofthe reference range for the serum level of LDH3. Ratio of pleural fluid level of protein <strong>to</strong> serum level of protein isgreater than 0.5embolism with infarction, benign effusion related<strong>to</strong> asbes<strong>to</strong>s exposure, and trauma. 19 In franklybloody effusions, a fluid hema<strong>to</strong>crit level greaterthan half the serum hema<strong>to</strong>crit level is indicativeof hemothorax. 3 If the fluid is particularly turbid ormilky (Figure 1), centrifugation is useful in differentiatingempyema from chylothorax or pseudochylothorax.In empyema, the supernatant fluid is clearwhere the cell debris forms a pellet. In chylothoraxor pseudochylothorax, the fluid remains uniformdue <strong>to</strong> the high lipid content. 3 Anchovy-brown fluidand black fluid are indicative of amebic liverabscess and Aspergillus infection, respectively.Once aspirated, the fluid is sent for biochemical,microbiological, and cy<strong>to</strong>logical analyses. Biochemicalanalyses include determination of protein, pH,lactic dehydrogenase, glucose, and albumin levels.Because it enters the pleural space with water, proteinhas become an important marker in the differentiationof transudates from exudates.A pleural effusion with a protein level less than30 g/L indicates a transudate, whereas one with alevel greater than 30 g/L indicates an exudate, providedthe serum protein level is within the normalreference range. When a protein level greater than30 g/L is used as the only basis for determining thetype of effusion, 10% of exudates and 15% of transudatesare misclassified. 26 Consequently, the use ofLight’s criteria (Table 3) is recommended when theprotein level is between 25 and 35 g/L. 27 Use of thesecriteria requires the additional measurement of apatient’s serum levels of lactic dehydrogenase andprotein. Although Light’s criteria are almost 100%sensitive for exudates, in a prospective comparativestudy of pleural effusions in 172 patients, Porcel et al 28found that approximately 20% of the patients withheart failure who were taking diuretics also metLight’s criteria for an exudate. In such instances, ifthe difference between serum and pleural levels ofprotein is greater than 31 g/L, the effusion shouldbe classified as a transudate. 29 Albumin levels canalso be used in this manner: a difference of morethan 12 g/L between serum and fluid levels indicatesa transudate. 30 Of note, a large percentage ofexudates will be misclassified if these gradients areused as the only method of differentiating betweentransudates and exudates. 6A pH less than 7.2 in infected effusions indicatesa complicated parapneumonic effusion (empyema)until proved otherwise, and insertion of chest drainand fluid removal are priorities. 3,6 A low pH can alsooccur in esophageal rupture, rheuma<strong>to</strong>id arthritis,and malignant neoplasm associated with poor outcome.6 Elevated levels of lactate dehydrogenase occurin lymphoma and tuberculosis; levels greater than1000 U/L (<strong>to</strong> convert units per liter <strong>to</strong> microkatalsper liter, multiply by 0.0167) are associated withempyema. 6 In an exudative effusion, pleural glucose122 AJCC AMERICAN JOURNAL OF CRITICAL CARE, March 2011, Volume 20, No. 2 www.ajcconline.org

levels less than 28.8 mg/dL (<strong>to</strong> convert milligramsper deciliter <strong>to</strong> millimoles per liter, multiply by0.0555) occur in tuberculosis, malignant neoplasm,empyema, rheuma<strong>to</strong>id arthritis, systemic lupus erythema<strong>to</strong>sus,and esophageal rupture. Rheuma<strong>to</strong>idarthritis probably is not responsible for a pleuraleffusion if the fluid glucose level is greater than1.6 mmol/L. 3 Cell differential analysis can also helpnarrow the differential diagnosis. Fluid lymphocy<strong>to</strong>sisoccurs in conditions such as tuberculosis, sarcoidosis,chylothorax, rheuma<strong>to</strong>id arthritis, andmalignant neoplasm, including lymphoma. 30,31<strong>Pleural</strong> fluid predominated by neutrophils isassociated with pulmonary embolism, parapneumoniceffusion, acute tuberculosis, and benign effusionrelated <strong>to</strong> exposure <strong>to</strong> asbes<strong>to</strong>s. 31A predominance of eosinophils in pleural fluidhas no diagnostic value, and up <strong>to</strong> one-third of thistype of effusion is never diagnosed. 32 However, effusionswith mostly eosinophils have been associatedwith air or blood in the pleural space. 6Routine microbiological analysis includes sendinga fluid sample for Gram and Ziehl-Neelson stainsand cultures <strong>to</strong> detect mycobacteria and other bacteria.Cy<strong>to</strong>logical analysis is extremely important ifa malignant neoplasm is suspected, and the analysesare positive for a malignant tumor in 60% ofpatients who have such a neoplasm. 3 If the firstsample is negative for tumor, sending a secondsample is worthwhile because analysis of 2 samplesmarkedly increases the chance of diagnosis ofmalignant neoplasm. 33Levels of adenosine deaminase (ADA) levels areparticularly useful in areas where the prevalence oftuberculosis is high. An ADA level greater than 40 U/L(<strong>to</strong> convert units per liter <strong>to</strong> nanokatals per liter,multiply by 16.667) has a sensitivity of more than90% and a specificity of about 85% for the presenceof tuberculosis. 4,34 In lymphocyte-predominant effusions,the specificity of ADA for tuberculosis increases<strong>to</strong> more than 95%. Elevated ADA also occurs withmalignant neoplasm, empyema, and rheuma<strong>to</strong>idarthritis. 35 Of note, ADA levels may be normal inthe pleural fluid of patients positive for humanimmunodeficiency virus who have tuberculosis. 36Several more unusual tests of pleural fluid canbe performed if clinical findings indicate 3,6 (Table 4).The results may be helpful in making the relevantdiagnosis. In particular, polymerase chain reaction,tumor markers, and complement are currently ofmuch interest. Polymerase chain reaction is usefulin diagnosing Strep<strong>to</strong>coccus pneumoniae infectionand tuberculosis. For tuberculosis, the sensitivity is40% <strong>to</strong> 60% and the specificity is greater than 90%,Table 4Useful further tests of pleural fluid forassessing the causes of pleural effusion aTestCholesterol levelTriglyceride levelHema<strong>to</strong>crit levelAmylase levelNT-proBNP levelCreatinine levelPolymerase chain reactionTumor markers CEA, CA15.3, CA 549 levelsCYFRA 21-1 levelCA 125, HER-2/neu levelComplement C4 levelCentrifugationTable 5<strong>Pleural</strong> fluid characteristics of chylothoraxand pseudochylothoraxResults of fluid analysisTriglyceride level, mmol/L (mg/dL)Cholesterol level, mmol/L (mg/dL)Cholesterol crystalsChylomicronsSuggested diagnosisChylothorax or pseudochylothorax (seeTable 5)Hemothorax if >50% serumhema<strong>to</strong>crit levelPancreatitis or esophageal rupture,depending on isotypeHeart failure, if elevatedUrinothorax if pleural fluid creatinine>serum creatinineTuberculosis or Strep<strong>to</strong>coccus pneumoniaeinfectionBreast carcinomaLung carcinomaOvarian, endometrial, and breast cancersRheuma<strong>to</strong>id arthritis, if reducedUsed <strong>to</strong> differentiate empyema fromchylothorax or pseudochylothoraxAbbreviations: CA, cancer antigen; CEA, carcinoembryonic antigen; CYFRA, cy<strong>to</strong>keratin-19fragments; HER, human epidermal growth recep<strong>to</strong>r; NT-proBNP, N-terminalpro B-type natriuretic peptide.a Based on data from Porcel and Light. 6Pseudochylothorax5.18 (200)Yes, commonalthough the last percentage is reduced in patientswho have cultures negative for mycobacteria. 6,37Tumor markers in pleural fluid may be usefulin certain clinical situations, although in generalthe sensitivity is quite low (1.24 (110)

Figure 2 Chest radiograph and computed <strong>to</strong>mography scan ofthorax show a right-sided effusion (arrow).In frankly bloodyeffusions, a fluidhema<strong>to</strong>crit of morethan half the serumhema<strong>to</strong>crit indicateshemothorax.Removal of all thefluid before thecomputed <strong>to</strong>mographyis done willresult in less thanoptimal imaging.greater for cy<strong>to</strong>keratin-19 fragments 21-1 was predictiveof a lower survival (4 vs 11.7 months foradenocarcinoma, 0.3 vs 8.4 months for squamouscell carcinoma). This tumor marker combinationremained an independent predic<strong>to</strong>r ofpoor outcome after adjustments weremade for age and tumor type. 17Complement levels in pleuralfluid have traditionally been analyzedwhen au<strong>to</strong>immune disease such asrheuma<strong>to</strong>id arthritis or systemic lupuserythema<strong>to</strong>sus is suspected (reducedlevels of complement). However,recently an association between transudatesand low complement levels wasdescribed. In a prospective study of 135 patientswith pleural effusion of known causes (cardiac failure,malignant neoplasm, parapneumonia, afterlung transplant, and Dressler syndrome), Shitrit etal 18 found that effusions related <strong>to</strong> congestive heartfailure could be differentiated from parapneumonicand pos<strong>to</strong>perative pleural effusions by the low levelof complement in the first type. Normal levels ofcomplement almost unequivocally indicate thatcongestive heart failure is not an etiological fac<strong>to</strong>r.Shitrit et al 18 also noted that in lung transplantrecipients, normal or high levels ofcomplement in pleural fluid mayindicate that the pleural effusion isnot attributable <strong>to</strong> the surgery, but <strong>to</strong>another secondary cause (eg, parapneumoniceffusions).A conventional posteroanteriorradiograph can show the presence ofan effusion (Figure 2). Lateral chestradiographs can be used <strong>to</strong> detectsmall effusions. If any doubt exists,thoracic ultrasound is useful in detecting smalleffusions, differentiating pleural fluid from pleuralthickening, and aspirating fluid from especiallysmall or loculated effusions. Thoracic CT with contrastmaterial can show pathological pulmonarychanges such as pneumonia or tumor and pleuralthickening and nodularity that may be amenable <strong>to</strong>percutaneous biopsy. An important step is <strong>to</strong> avoiddraining all the fluid before CT is done becauseremoval of all fluid results in less than optimalpleural imaging. If the clinical his<strong>to</strong>ry and findingsare suggestive of pulmonary embolism, then helicalCT may be indicated. 3If the results of fluid analysis and radiology arenot sufficient for diagnosing a persistent unexplainedexudative effusion, then a pleural biopsy is indicated.The biopsy may be blind, image guided, or thoracoscopic.As shown in a retrospective analysis of thediagnosis of pleural effusions in 414 patients byPrakash and Reiman, 38 blind closed biopsy has traditionallybeen of little benefit in diagnosing malignanteffusions in patients whose cy<strong>to</strong>logical assaysshowed no malignant cells. Indeed, fluid cy<strong>to</strong>logyis superior <strong>to</strong> blind closed pleural biopsy in thediagnosis of malignant disease. 6 Poe et al 39 retrospectivelyanalyzed 211 patients who underwentpleural biopsy during a 6-year period and foundthat blind closed pleural biopsy was useful in diagnosingtuberculosis, especially in patients with fluidlymphocy<strong>to</strong>sis. In these patients, the diagnostic ratewas greater than 90% when tissue was analyzed bothhis<strong>to</strong>logically and microbiologically. In a small butimportant study of 50 consecutive patients withsuspected pleural malignant disease whose cy<strong>to</strong>logicalassays showed no malignant cells, Maskell et al 15found that image-guided pleural biopsy greatlyassisted in the diagnosis of malignant neoplasmand was better than blind closed biopsy. Consequently,image-guided biopsy has essentiallyreplaced the blind biopsy technique. With imageguidance, areas of pleural thickening or nodularitycan be accurately biopsied, thus increasing thediagnostic rate. 15Image-guided biopsy is also useful in patientswho are <strong>to</strong>o weak <strong>to</strong> undergo thoracoscopy. Inpatients with no evidence of malignant disease, pleuralthickening, or pleural nodularity, or if the resultsof image-guided biopsy are negative for malignantdisease, thoracoscopy is indicated. 3 This techniqueallows close visual examination of the pleura andaccurate biopsy of abnormal tissue and is diagnosticin approximately 90% of cases. Thoracoscopy canbe performed by either a pulmonologist (medicalthoracoscopy) or a surgeon (video-assisted thoracoscopicsurgery). Drainage of fluid and pleurodesiscan be performed at the same time as the thoracoscopyif indicated, and in many medical facilities,124 AJCC AMERICAN JOURNAL OF CRITICAL CARE, March 2011, Volume 20, No. 2 www.ajcconline.org

Bilateral<strong>Pleural</strong> effusionChest radiographUnilateralEvidence of cardiac, liver, or renal failureEvidence of cardiac, liver, or renal failureYesNo evidenceoruneven effusionsorunresponsive <strong>to</strong> treatmentNo evidenceorunresponsive <strong>to</strong> treatmentYesDiagnostic thoracentesisAspirate for pH, protein, lactate dehydrogenase, glucose, Gramstain, and acid-fast bacillus stains with culture and sensitivitiesTransudateExudateDoes the fluid anaylsis indicate a diagnosis?NoYesInvestigate andtreat cause(eg, heart failure,cirrhosis,renal failure)Repeat fluid anaylsis and consider additional fluid tests (Table 4)Does the fluid anaylsis indicate a diagnosis?YesNoContrast-enhanced computed <strong>to</strong>mography of thoraxYesDiagnosisNoHemoptysis, lung infiltrates, or atelectasisConsider bronchoscopyYesDiagnosisNo, but effusion resolvingNoObserveConsider blind or image-guided pleural biopsyorMedical/surgical thoracoscopyFigure 3 <strong>Recommended</strong> algorithm for investigation of a pleural effusion.

thoracoscopy is the next step for patients whosecy<strong>to</strong>logical results are negative for malignant cells.Bronchoscopy is not routinely recommended inthe investigation of pleural effusion unless the patienthas hemoptysis or radiological features of malignantneoplasm such as a mass, massive pleural effusion, ora shift in the midline <strong>to</strong>ward the side of the effusion. 6In patients with massive effusion, drainage beforebronchoscopy is recommended <strong>to</strong> allow an adequateexamination without extrinsic compression. 3Recommendations Based on CurrentEvidenceInvestigation of a pleural effusion evident onchest radiographs should follow a stepwise approach(Figure 3) <strong>to</strong> diagnosis. Diagnosis begins with theclinical his<strong>to</strong>ry, physical examination, and chestradiography and is followed by thoracentesis whenappropriate. The majority of the recommendationsin Figure 3 represent class IIa evidence. An exceptionis the use of image-guided pleural biopsy (insteadof blind pleural biopsy) in the last step of the algorithm;this type of biopsy has been assessed byusing a randomized controlled trial and thereforerepresents class I evidence.Examination of the pleural fluid can narrowthe differential diagnosis considerably. The appearanceof the fluid and biochemical parameters canbe key <strong>to</strong> a direct diagnosis or can indicate the nextstep. In transudative effusions, the underlying causeshould be sought and treated. In exudative effusionsin which fluid analysis does not lead <strong>to</strong> immediatediagnosis, CT of the thorax should be performed.If the diagnosis is still not evident after CT, pleuralbiopsy (radiologically or medical thoracoscopy) isrecommended. In a few patients, the effusion maybegin <strong>to</strong> improve. In these instances, the patientshould be observed; the disease process may beresolving and further invasive investigation maynot be warranted.FINANCIAL DISCLOSURESNone reported.eLettersNow that you’ve read the article, create or contribute <strong>to</strong> anonline discussion on this <strong>to</strong>pic. Visit www.ajcconline.organd click “Respond <strong>to</strong> This Article” in either the full-text orPDF view of the article.REFERENCES1. Starling EH, Tubby A. On absorption from and secretionin<strong>to</strong> the serous cavities. J Physiol. 1894;16:140-155.2. Mitrouska I, Bouros D. The trans-exudative pleural effusion.Chest. 2002;122(5):1503-1505.3. Maskell NA, Butland RJ; <strong>Pleural</strong> Diseases Group, Standardsof Care Committee, British Thoracic Society. BTS guidelinesfor the investigation of a unilateral pleural effusion in adults.Thorax. 2003;58(suppl 2):ii8-ii17.4. Morisson P, Neves DD. Evaluation of adenosine deaminasein the diagnosis of pleural tuberculosis: a Brazilian metaanalysis.J Bras Pneumol. 2008;34(4):217-224.5. Go<strong>to</strong> M, Noguchi Y, Koyama H, Hira K, Shimbo T, Fukui T.Diagnostic value of adenosine deaminase in tuberculouspleural effusion: a meta-analysis. Ann Clin Biochem. 2003;40(pt 4):374-381.6. Porcel JM, Light RW. Diagnostic approach <strong>to</strong> pleural effusionin adults. Am Fam Physician. 2006;73(7):1211-1220.7. Chapman SJ, Cookson WO, Musk AW, Lee YC. Benign asbes<strong>to</strong>spleural diseases. Curr Opin Pulm Med. 2003;9(4):266-271.8. Romero-Candeira S, Hernández L. The separation of transudatesand exudates with particular reference <strong>to</strong> the proteingradient. Curr Opin Pulm Med. 2004;10(4):294-298.9. Kalomenidis I, Light RW. Pathogenesis of the eosinophilicpleural effusions. Curr Opin Pulm Med. 2004;10(4):289-293.10. Medford A, Maskell N. <strong>Pleural</strong> effusion. Postgrad Med J.2005;81(961):702-710.11. Froudarakis ME. Diagnostic work-up of pleural effusions.Respiration. 2008;75(1):4-13.12. Rahman NM, Munavvar M. Investigation of the patient withpleural effusion. Clin Med. 2009;9(2):174-178.13. Rahman NM, Chapman SJ, Davies RJ. <strong>Pleural</strong> effusion: astructured approach <strong>to</strong> care. Br Med Bull. 2005;72:31-47.14. An<strong>to</strong>ny VB, Loddenkemper R, As<strong>to</strong>ul P, et al. Management ofmalignant pleural effusions. Eur Respir J. 2001;18(2):402-419.15. Maskell NA, Gleeson FV, Davies RJ. Standard pleuralbiopsy versus CT-guided cutting-needle biopsy for diagnosisof malignant disease in pleural effusions: a randomisedcontrolled trial. Lancet. 2003;361(9366):1326-1330.16. Liang QL, Shi HZ, Qin XJ, Liang XD, Jiang J, Yang HB.Diagnostic accuracy of tumor markers for malignant pleuraleffusion: a meta-analysis. Thorax. 2008;63(1):35-41.17. Bielsa S, Esquerda A, Salud A, et al. High levels of tumormarkers in pleural fluid correlate with poor survival inpatients with adenocarcinoma<strong>to</strong>us or squamous malignanteffusions. Eur J Intern Med. 2009;20(4):383-386.18. Shitrit D, Ollech JE, Ollech A, et al. Diagnostic value ofcomplement components in pleural fluid: report of 135cases. Respir Med. 2008;102(11):1631-1635.19. Villena V, López-Encuentra A, García-Luján R, Echave-SustaetaJ, Martínez CJ. Clinical implications of appearance of pleuralfluid at thoracentesis. Chest. 2004;125(1):156-159.20. Gao ZC, Tian RX. Clinical investigation on diagnostic valueof interferon-gamma, interleukin-12 and adenosine deaminaseisoenzyme for tuberculous pleurisy. Chin Med J (Engl).2005;118(3):234-237.21. Chakrabarti B, Ryland I, Sheard J, Warbur<strong>to</strong>n CJ, Earis JE.The role of Abrams percutaneous pleural biopsy in theinvestigation of exudative pleural effusions. Chest. 2006;129(6):1549-1555.22. Jiménez D, Díaz G, García-Rull S, Vidal R, Sueiro A, Light RW.Routine use of pleural fluid cultures: are they indicated?Limited yield, minimal impact on treatment decisions.Respir Med. 2006;100(11):2048-205223. Qureshi NR, Rahman NM, Gleeson FV. Thoracic ultrasoundin the diagnosis of malignant pleural effusion. Thorax.2009;64(2):139-143.24. Yap E, Anderson G, Donald J, Wong CA, Lee YC, SivakumaranP. <strong>Pleural</strong> effusion in patients with pulmonary embolism.Respirology. 2008;13(6):832-836.25. ECC guidelines: part 1: introduction <strong>to</strong> the internationalguidelines 2000 for CPR and ECC. Circulation. 2000;102(suppl 1):I-1-I-11.26. Sahn SA, Heffner JE. <strong>Pleural</strong> fluid analysis. In: Light RW,Gary Lee YC. Textbook of <strong>Pleural</strong> Diseases. London, England:Arnold; 2003:191-209.27. Light RW. <strong>Pleural</strong> effusion. N Engl J Med. 2002;346(25):1971-1977.28. Porcel JM, Vives M, Vicente de Vera MC, Cao G, Rubio M,Rivas MC. Useful tests on pleural fluid that distinguish transudatesfrom exudates. Ann Clin Biochem. 2001;38(pt 6):671-675.29. Ansari T, Idell S. Management of undiagnosed persistentpleural effusions. Clin Chest Med. 1998;19(2):407-417.30. Romero-Candeira S, Fernández C, Martín C, Sánchez-Paya J,Hernández L. Influence of diuretics on the concentrationof proteins and other components of pleural transudatesin patients with heart failure. Am J Med. 2001;110(9):681-686.126 AJCC AMERICAN JOURNAL OF CRITICAL CARE, March 2011, Volume 20, No. 2 www.ajcconline.org

31. Light RW, Erozan YS, Ball WC Jr. Cells in pleural fluid: theirvalue in differential diagnosis. Arch Intern Med. 1973;132(6):854-860.32. Light RW. <strong>Pleural</strong> Diseases. 4th ed. Philadelphia, PA: LippincottWilliams & Wilkins; 2001.33. Garcia LW, Ducatman BS, Wang HH. The value of multiplefluid specimens in the cy<strong>to</strong>logical diagnosis of malignancy.Mod Pathol. 1994;7(6):665-668.34. Porcel JM, Vives M. Differentiating tuberculous from malignantpleural effusions: a scoring model. Med Sci Monit.2003;9:CR175-CR180.35. Burgess LJ, Maritz FJ, Le Roux I, Taljaard JJ. Use of adenosinedeaminase as a diagnostic <strong>to</strong>ol for tuberculouspleurisy. Thorax. 1995;50(6):672-674.36. Hsu WH, Chiang CD, Huang PL. Diagnostic value of pleuraladenosine deaminase in tuberculous effusions of immunocompromisedhosts. J Formos Med Assoc. 1993;92(7):668-670.37. Pai M, Flores LL, Hubbard A, Riley LW, Colford JM Jr.Nucleic acid amplification tests in the diagnosis of tuberculouspleuritis: a systematic review and meta-analysis. BMCInfect Dis. 2004;4:6.38. Prakash UB, Reiman HM. Comparison of needle biopsy withcy<strong>to</strong>logic analysis for the evaluation of pleural effusion:analysis of 414 cases. Mayo Clin Proc. 1985;60(3):158-164.39. Poe RH, Israel RH, Utell MJ, Hall WJ, Greenblatt DW, KallayMC. Sensitivity, specificity, and predictive values of closedpleural biopsy. Arch Intern Med. 1984;144(2):325-328.To purchase electronic or print reprints, contact TheInnoVision Group, 101 Columbia, Aliso Viejo, CA 92656.Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax,(949) 362-2049; e-mail, reprints@aacn.org.www.ajcconline.org AJCC AMERICAN JOURNAL OF CRITICAL CARE, March 2011, Volume 20, No. 2 127

CE Test Test ID A112002: Diagnosis of <strong>Pleural</strong> <strong>Effusion</strong>: A Systematic <strong>Approach</strong>.Learning objectives: 1. Identify 3 characteristics of abnormal pleural fluid. 2. Describe a systematic algorithm for the investigation of pleural effusions.3. Differentiate between extudates and transdates.1. If a recommended practice is supported by only 1 randomizedcontrol study and a few case studies, it is considered <strong>to</strong> be which ofthe following?a. Class I evidence c. Class IIb evidenceb. Class IIa evidence d. Class II evidence2. A patient who was a construction worker in the 1950s has dyspnea, acough, and sharp radiating chest pain. Which of the following is mostlikely <strong>to</strong> be the source of his pleural effusion?a. Pneumonia c. Pulmonary embolismb. Mesothelioma d. Rheuma<strong>to</strong>id arthritis3. The patient with a pleural effusion caused by a pulmonary embolismis likely <strong>to</strong> have which of the following?a. A pericardial friction rub c. Unilateral leg swellingb. Bilateral leg swelling d. Weight loss4. Which of the following is associated with blood stained pleural f luid?a. Aspergillus infection c. Pulmonary embolismb. Amebic liver abscess d. Pseudochylothorax5. Why is it important <strong>to</strong> compare serum protein levels <strong>to</strong> pleuralf luid protein levels when differentiating between an exudate anda transudate?a. Normal serum protein with normal pleural protein signifies a transudate.b. Normal serum protein with a pleural fluid less than a 30 g/L signifies anexudate.c. If the difference between serum and pleural protein levels is 12 g/L or less,the effusion is classified as a transudate.d. If the difference between serum and pleural protein levels is greater than31 g/L, the effusion is classified as a transudate.6. If a nurse is caring for a pneumonia patient undergoing a thoracentesisand the pleural f luid has a pH of 7.1, which of the followingactions should the nurse take?a. Schedule the patient for a computed <strong>to</strong>mography scanb. Assist with chest tube insertionc. Repeat the pleural fluid analysisd. Schedule the patient for a bronchoscopy7. <strong>Pleural</strong> f luid that has an elevated lactate dehydrogenase level isprobably due <strong>to</strong> which of the following?a. Rheuma<strong>to</strong>id arthritis c. Systemic lupus erythema<strong>to</strong>susb. Tuberculosis d. Sarcoidosis8. According <strong>to</strong> the algorithm, which of the following is the f irststep in the investigation of a pleural f luid following his<strong>to</strong>ry and chestradiograph?a. Computed <strong>to</strong>mography scan c. Thoracentesisb. Bronchoscopy d. <strong>Pleural</strong> biopsy9. A patient with congestive heart failure and a pleural effusion willmost likely have which of the following?a. Normal complement levels c. Elevated complement levelsb. Reduced complement levels d. No complement will be measured10. Which of the following diagnostic tests would be most benef icialin diagnosing malignant disease after inconclusive radiology andf luid analysis?a. Bronchoscopy c. Closed biopsyb. Image guided biopsy d. Thoroscopy11. Food particles in the pleural f luid can be a sign of which of thefollowing?a. Chylothorax c. Esophageal ruptureb. Trauma d. Anerobic empyema12. Which of the following can cause a pleural effusion with unilateralexudates?a. Malignant neoplasm c. Hypothyroidismb. Cirrhotic liver disease d. Mitral stenosis13. Pseudochylothorax can be differentiated from chylothorax if itcontains which of the following?a. Chylomicrons c. Triglycerides 1.32b. Cholesterol crystals d. Cholesterol 4.2Test ID: A112002 Contact hours: 1.0 Form expires: March 1, 2013. Test Answers: Mark only one box for your answer <strong>to</strong> each question. You may pho<strong>to</strong>copy this form.1. ❑a❑b❑c❑d2. ❑a❑b❑c❑d3. ❑a❑b❑c❑d4. ❑a❑b❑c❑d5. ❑a❑b❑c❑d6. ❑a❑b❑c❑d7. ❑a❑b❑c❑dFee: AACN members, $0; nonmembers, $10 Passing score: 10 Correct (77%) Synergy CERP Category: A Test writer: Marylee Bressie, RN, MSN, CCRN, CCNs, CENProgram evaluationName Member #Yes NoObjective 1 was met ❑ ❑ AddressObjective 2 was met ❑ ❑Objective 3 was met ❑ ❑City State ZIPContent was relevant <strong>to</strong> myCountry Phone E-mail addressFor faster processing, take nursing practice ❑ ❑this CE test online at My expectations were met ❑ ❑ RN License #1Statewww.ajcconline.org (“CE This method of CE is effectiveRN License #2StateArticles in This Issue”) or for this content ❑ ❑mail this entire page <strong>to</strong>:The level of difficulty of this test was:Payment by: ❑ Visa ❑ M/C ❑ AMEX ❑ Check❑ easy ❑ medium ❑ difficultAACN, 101 Columbia,To complete this program,Card #Expiration DateAliso Viejo, CA 92656. it <strong>to</strong>ok me hours/minutes.SignatureThe American Association of Critical-Care Nurses is accredited as a provider of continuing nursing education by the American Nurses Credentialing Center’s Commission on Accreditation.AACN has been approved as a provider of continuing education in nursing by the State Boards of Nursing of Alabama (#ABNP0062), California (#01036), and Louisiana (#ABN12). AACNprogramming meets the standards for most other states requiring manda<strong>to</strong>ry continuing education credit for relicensure.8. ❑a❑b❑c❑d9. ❑a❑b❑c❑d10. ❑a❑b❑c❑d11. ❑a❑b❑c❑d12. ❑a❑b❑c❑d13. ❑a❑b❑c❑d