Pericardial Fluid Analysis 

Updated: Oct 30, 2015
  • Author: See Wun Flora Yau, DO; Chief Editor: Eric B Staros, MD  more...
  • Print

Reference Range

Pericardial fluid is collected via pericardiocentesis or open surgical drainage.

The specimen stability is as follows:

  • Room temperature - 7 days
  • Refrigerated - 7 days
  • Frozen - 28 days

Cells may degenerate during storage. Therefore, pericardial fluid sample for cytopathology study should be sent to the laboratory as soon as possible in a fresh state or refrigerated at 2-8º C.

The volume must be a minimum of 2 mL for each laboratory test.

Referring to each laboratory for more specific instructions about sample collection and transport is important.

Routine panels are as follows:

  • Cell count with differential
  • Glucose level
  • Pericardial fluid total protein
  • Pericardial fluid lactate dehydrogenase (LDH)
  • Serum Complement (anti-dsDNA, rheumatoid factor [RF], antinuclear antibody [ANA])
  • Gram stain and culture (at least 3 culture bottles from pericardial fluid)

Special panels are as follows:

  • Cytology (if malignancy is suspected)
  • Tumor markers (if malignancy is suspected)
  • Adenosine deaminase (if tuberculosis [TB] is suspected)
  • Polymerase chain reaction (PCR) for Mycobacterium tuberculi (if TB is suspected)
  • Pericardial interferon-gamma (interferon-gamma) if TB is suspected
  • Viral cultures
  • Molecular analysis (PCR) for bacteriological, viral, or fungal agents
  • B-type natriuretic peptide (BNP)


Bloody pericardial effusion  [1]

  • Iatrogenic: The most common cause in developed countries. This includes the effect of anticoagulant therapy, trauma, postinvasive cardiac procedures (ie, postpericardiotomy syndrome, transcatheter interventions).
  • Malignancy
  • Atherosclerotic heart disease (mainly complications of acute myocardial infarction)
  • Tuberculosis: This condition remains to be one of the most common causes of pericarditis/pericardial effusion in Africa and TB-dominant developing countries. Approximately 80% of cases of tuberculous pericarditis are bloodstained effusions.
  • Idiopathic

If the fluid is milky, consider the involvement of the lymphatic system (ie, chylopericardium). If the fluid is cloudy and turbulent, it is suggestive of signs of increased capillary leakage and leukocytosis and is concerning for infectious effusion.

Routine tests

The following conditions are associated with elevated WBC counts:

  • Elevated levels of leukocytes (>10,000/mcl) with neutrophil predominance suggests a bacterial or rheumatic cause.
  • The monocyte count is noted to be highest in malignant effusions.

Myxedema is associated with low WBC count.

At this time, no standardized biochemical or cell count criteria meets a statistical relationship between specific causes of effusion. However, an elevated WBC level is suggestive of an inflammation within the pericardium.

A low ratio of pericardial effusion (PE) and serum glucose suggests infection. This low ratio, along with an elevated neutrophil count in pericardial fluid, is suggestive for bacterial pericardial effusion. [2]

According Light et al’s criteria for pleural fluid, an exudate is considered if one of the following parameters are met: the total protein fluid-to-serum ratio is higher than 0.5, the LDH fluid-to-serum ratio is more than 0.6, or the LDH fluid level exceeds two thirds of upper limit of normal serum level. [3] According to Burgess et al, Light et al’s criteria was applied to pericardial effusion and sensitivity and specificity were reported to be 98% and 72%, respectively. [4]

However, debate still surrounds the inconsistent data regarding the adaptation of Light et al’s criteria of pleural effusion to pericardial effusion. [5, 2] Dissimilar to pleural fluid, the establishment of exudative versus transudative effusion cannot be implied with the use of this criteria. A study by Ben-Horin et al noted that 118 of 120 patients with pericardial fluid fit the category for exudative under Light et al’s criteria. [2] Spodick also demonstrates that patients with improving congestive heart failure present with pseudoexduative effusion, likely secondary to rapid reabsorption of water compared with proteins and LDH in uncomplicated heart failure. [6]

In conclusion, analyses of pericardial fluid protein level (>0.5) and LDH (>0.6) can suggest between exudates from transudates pericardial fluid, [7] but should not be used as a sole diagnostic framework. Interpretation of the results should be applied with patient’s clinical presentation and other relevant laboratory results.

Complement levels, ANA and anti-ds DNA can be measured in the setting of pericardial effusion and systemic lupus erythematous to help identify pericardial membrane involvement.

If bacterial infection is suspected, at least 3 cultures of pericardial fluid for aerobes and anaerobes as well as blood cultures are required. [7] Gram stains in bacterial infection pericardial fluid have a specificity of 99% but a sensitivity of only 38% compared with bacterial cultures. [8] The use of Gram stain and culture is generally limited in mycobacterial, viral, and fungal infection.

Special tests

Cytologic study of pericardial fluid helps identify malignancy as the cause of pericardial effusion by detecting neoplastic cells within the fluid. However, it is not always straightforward. Nonmalignant cells can be morphologically indistinguishable from malignant cells. For example, mesothelial cell’s morphology can be benign, hyperplastic, reactive, or malignant. Data on pericardial fluid analysis are limited; however, a study by Rakha et al found the overall sensitivity of cytology evaluation of pleural fluid in the diagnosis of malignant mesothelioma was 53%. It appears to be useful in patients with epithelioid and biphasic pleural malignant mesothelioma; however, the sensitivity could be as low as 20% in sarcomatoid malignancy mesothelioma cases. [9]

A similar issue was described in cases of lymphoma. Ancillary studies, including immunocytochemistry, morphometry, flow cytometry, and cytogenetics/molecular genetics on effusion specimens may be helpful in differentiating lymphoma from reactive lymphocytoses with higher sensitivity. [10]

Cytologically negative effusion does not exclude malignancy as the cause. In a retrospective study of 82 patients with nonsmall cell lung cancer (NSCLC) and cardiac tamponade, no survival difference was noted between patients with positive pericardial fluid cytology findings for cancer (60 patients) compared with those with negative cytology findings (22 patients). Significant survival difference was reported after systemic chemotherapy was initiated. The authors concluded that in patients with advanced NSCLC and cardiac tamponade, the most likely cause of the pericardial effusion is the cancer itself, regardless of the cytology results. [11, 12]

The use of tumor marker measurement in pericardial fluid may be helpful in cytologically negative pericardial fluid. The interpretation of these tumor marker levels is similar to that in serum. Malignant pericardial effusion is associated with high levels of tumor markers. However, low levels of these markers do not exclude neoplastic involvement of the pericardial membrane. The cut-off of these markers in pericardial fluid is not yet clear at this point. Moreover, different types of cancer can have different tumor markers. Various tumor markers have been tested, such as carcinoembryonic antigen (CEA), carbohydrate antigen (CA) 19-9, carbohydrate antigen (CA) 72-4, squamous cell carcinoma (SCC) antigen, neuron-specific enolase (NSE), serum cytokeratin 19 fragments (CYFRA 21-1), BerEp4, and hyaluronan.

Szturmowicz reported considerably higher median CEA and CYFRA 21-1 concentrations in malignant pericardial effusion compared with nonmalignant pericardial effusion (80 ng/mL [0-317] vs. 0.5 ng/mL [0-18.4], and 260 ng/mL [5.3-10080] vs 22.4 ng/mL, respectively). In this study, the optimal cutoff value for CYFRA 21-1 in pericardial effusion was 100 ng/mL, and CEA 5 ng/mL. CYFRA 21-1 of more than 100 ng/mL or CEA of more than 5 ng/mL were found in 14 of 15 patients with malignant pericardial effusion and negative pericardial fluid cytology findings. [13]

Similar results were found in a study by Karatolios et al that included 29 patients with proven malignant pericardial effusion and 25 patients with nonmalignant causes. The mean concentrations of the CEA, CA 72-4, and CA 19-9 were significantly higher in malignant pericardial effusions than in nonmalignant effusions. Pericardial fluid CA 72-4 levels >1 kU/L had 72% sensitivity and 96% specificity in differentiating malignant pericardial effusions from effusions due to benign conditions. [14]

Elevated pericardial ADA activity is suggestive of TB pericarditis. The test is a valid diagnostic tool applicable regardless of HIV status. A lower ADA level may be observed in patients with HIV who have a low CD4 count. ADA levels of more than 40 U/L are diagnostic for TB pericarditis (sensitivity 88%, specificity 83%). [15] Elevated levels of more than 200 pg/L are suggestive of tuberculous pericarditis. (sensitivity 92%, specificity 100%). [16] A definite diagnosis of TB pericarditis is the presence of tubercle bacilli in pericardial fluid or on histological section of a pericardium. However, conventional culture for TB has a lower sensitivity, and inoculation of fluid into double-strength liquid Kirchner culture medium is recommended because this may increase the yield from 53% to 75%. [15, 16]

Polymerase chain reaction (PCR) has been used to detect M tuberculosis using nucleic acid amplification. An advantage of the test is its ability to rapidly identify M tuberculosis with as small as 1 mcL of pericardial fluid. The technique, however, is less sensitive than established methods and is prone to contamination and false-positive results. At this time, PCR is not an ideal diagnostic tool to detect tuberculous pericarditis. Other methods as suggested above should be considered beforehand.

Molecular procedures involving direct amplification from sterile sites is an alternative approach in identification of pathogens associated with pericardial effusion. Etiological diagnosis have been shown to be significantly higher with use of PCR-based diagnosis than use of culture only, increasing the number of cases from 13.9% to 39.5% (p< 0.01). [17] The cost-effectiveness and accessibility of these molecular procedures, however, may be a limiting factor in the complementary method to the systemic approach to the diagnosis of pericardial effusion.

Elevation of B-type natriuretic peptide (BNP) levels in pericardial fluid is noted in patients with postmyocardial infarction, reflecting the stretching of ventricular cardiomyocytes after an injury to the myocardium. No statistical association is noted between elevated pericardial BNP levels and left ventricular systolic function. [18]

Other considerations

The accuracy and diagnostic value of biochemical and cell count analysis of pericardial effusions in distinguishing the various etiology of the effusion is lacking. A considerable overlap of several laboratory parameters creates a difficult approach in determining the causative agent to the effusion. [1, 2] Drainage of the pericardium in patients without hemodynamic compromise also has low diagnostic yield. [15, 16]

With the limited validity and diagnostic value of routine hematological and biochemical tests of pericardial fluid, results should be interpreted cautiously and used only as clinical correlates to patient’s overall presentation.




Pericardial fluid is an ultrafiltrate of plasma that lies within the pericardial sac, acting as a lubricant between the visceral and parietal layer of the pericardium. The space normally contains 15-50 mL of thin, clear, straw-colored fluid that is enriched in molecules from the myocardial interstitial fluid and lymphatic drainage. Molecules up to 40 kDA are commonly diffused through the ventricular myocytes. The composition of the fluid is believed to be a result of Starlings forces and the gradients between hydrostatic and osmotic pressure of the pericardial fluid and plasma. [6, 19]  The fluid may subsequently reflect any circumstances imposed onto the heart, ranging from trauma and infections to metabolic derangements.

Limited studies have been able to define the normal physiological composition of pericardial fluid. An analysis of 30 patients undergoing open heart surgery with no history of pericardial disease, showed the mean total protein within the serosal fluid was 0.6 times of serum level. Pericardial fluid lactate dehydrogenase (LDH) was also noted to be higher than expected, with an average of 2.4 times higher than serum LDH and 1.5 times greater than the normal upper limits of serum LDH. [5] The ratio of small soluble molecules (ie, urea, uric acid, glucose, creatinine) between the pericardium and serum, however, were noted to be equated to nearly 1, suggesting the “transudative” nature of pericardial effusion.

This suggests that while certain features of the composition of pericardial fluid resemble an exudate if Light et al’s criteria is applied (namely LDH and total protein levels), other factors in the composition of the fluid (ratio of small soluble molecules) suggest transudate. Therefore, Light et al’s criteria is unlikely to be applicable in the evaluation of pericardial fluid composition. [5, 3]

Abnormal accumulation of pericardial fluid can be secondary to obstruction of fluid drainage, injury or insult to the pericardium, infection, malignancy, systemic or autoimmune processes, drugs, or procedures. Numerous cases are idiopathic. [20]

The etiology of pericardial effusion includes the following:

  • Idiopathic pericarditis
  • Infection
  • Bacteria
  • Staphylococcus, Streptococcus, Haemophilus, Neisseria, Chlamydia
  • M tuberculosis (still prevalent in developing countries)
  • Viral (coxsackievirus A and B, echovirus, adenovirus, HIV)
  • Fungal – Aspergillus, Candida, Histoplasma, Blastomycosis, Coccidioidomycosis
  • Protozoan – Echinococcus, Amebiasis, Toxoplasmosis
  • Neoplasms
  • Metastatic/paraneoplastic (breast, lung, leukemia, lymphoma)
  • Primary – Teratoma, lipoma, angioma, rhabdomyosarcoma
  • Autoimmune connective tissue disease
  • Trauma (eg, blunt and penetrating trama, radiofrequency catheter ablation of atrial fibrillation)
  • Metabolic causes
  • Hypothyroidism
  • Anorexia nervosa
  • Uremia
  • Chylopericardium
  • Drugs (hydralazine, isoniazid, procainamide, phenytoin, anticoagulants)
  • Pericardial injury syndrome (postmyocardial infarction effusion, posttraumatic effusion, postcardiotomy) [21, 22]
  • Radiation
  • Hyperlipidemia
  • Severe pulmonary hypertension
  • Thoracic aortic disease – Leakage or rupture


Invasive pericardial drainage procedure (ie, pericardiocentesis or open surgical drainage) and the diagnostic analysis of pericardial fluid is warranted in the following cases: [23]

  • Patients with a strong suspicion of purulent or TB pericarditis
  • To determine if the pericardial effusion is secondary to neoplastic pericardial involvement
  • Pericardial effusion of unknown origin
  • Patients with massive idiopathic chronic pericardial effusion
  • Pericardial tamponade caused by uncontrolled pericardial effusion with hemodynamic instability

Considerations must be taken with coagulopathic patients with increased risk of bleeding. For diagnostic or nonemergent pericardiocentesis, imaging is imperative.