Pleura and peritoneum are serous membranes that line the thoracic and abdominal cavities, and moreover cover the surfaces of organs in situ, as parietal and visceral parts respectively. Their function is based on the abilities of absorption, transudation and exudation, as well as on their ability to form adhesions in the presence of an inflammatory process.

Normally, a small amount of free fluid is present in the cavities. The abnormal accumulation of fluid is an indication of a pathologic process involved in it's production and/or removal and is not a disease itself. Effusion is produced as a result of alterations in the homeostatic mechanisms involved, such as increased vascular permeability, decreased lymphatic removal or obstruction, increased systemic venous hydrostatic pressure and decreased colloid oncotic pressure.

The laboratory evaluation of thoracic or/and abdominal effusion is a useful diagnostic tool for assessment of the diseases involved. Understanding disease pathophysiology helps the veterinarian maximizing the use of laboratory data. In small animals, diseases associated with peritoneal effusions include septic peritonitis, nonseptic peritonitis, hemoabdomen, uroabdomen, pancreatitis, bile peritonitis, chylous effusion, neoplasia, etc. Diseases associated with pleural effusions are heart failure, ruptured lymphatics, lung lobe torsion, trauma and hemothorax, diaphragmatic hernia, FIP, bacterial or fungal infections, heartworm, aelurostrongylosis, intrathoracic neoplasia, etc.

Pleural or/and peritoneal effusions are commonly encountered in veterinary practice, are usually suspected initially from clinical signs--physical findings and confirmed by thoracic radiography.

Practitioners can collect effusion samples quickly, inexpensively and safely. Although the appropriate sample collection by thoracocentesis or abdominocentesis and smear preparation are essential for an accurate diagnosis, such techniques are not described in this article which focuses on data interpretation and mainly on cytopathology.

"In-house laboratory" analysis of fluid samples should include the following parameters: gross examination of the effusion and physical characteristics (such as transparency or turbidity, color, odor, clots, fibrin), protein concentration and specific gravity, measurement of total nucleated cell count, packed red blood cell volume, as well as examination for the presence of other cells, bacteria, fungi, food particles, or plant material.

In small animal practice, determination of total protein concentration is one of the first examination of fluids. Protein concentration can be determined by use of urine test strips and refractometry. The Bradford and biuret spectrophotometric methods are also used by special laboratories. Care should be given to the reliability of strips and to refractometry which underestimates samples with very low protein concentration (< 2 g/dL). Samples collected into EDTA tubes, should be centrifuged and stored at -20° C, to be possibly assessed later. Cells can be enumerated using Unopette system (same used to make white blood cell counts) by the practitioner or an automated cell counter by external laboratories. An estimation of cell count may be made on examination of a stained film on a slide. The slide should be made the same way just as a blood film is made.

In general, biochemical evaluation should proceed based on the clinician's index of suspicion for a particular disease process. Centesis fluids are stable for up to 24 hours after collection, if they are refrigerated. If reference or commercial laboratories are used, fluids should be added directly to EDTA or non-anticoagulant tubes and then submitted. It is also essential to make fresh unfixed, rapidly air-dried and unstained smears from either concentrated fluid (sediment smears) or unconcentrated fluid (direct smears), because changes occur in vitro (e.g., phagocytosis of erythrocytes and bacteria) complicating result interpretation. If samples are hypocellular or hemorrhagic, cytospin centrifuge can be used. Portions of specimen should be cultured, since common bacteria (anaerobes or/and aerobes) such as Bacteroides, Fusobacterium, Escherichia coli, Enterobacter, Pasteurella, Streptococcus, Actinomyces, Nocardia could be involved.

If possible, chemistry tests performed from the fluid should include measurements of creatinine, urea nitrogen, bilirubin, lipase/amylase, triglycerides and cholesterol. These ancillary tests may help to determine a specific cause for fluid accumulation, for example uroperitoneum, bile peritonitis, pancreatitis, etc. Comparison of concurrent serum and pleural fluid cholesterol and triglyceride concentrations and cholesterol to triglyceride ratios of the effusion can be helpful in diagnosing chylous effusion and differentiate chylous from pseudochylous effusion.

Pleural and peritoneal fluids are categorized as transudates, modified transudates and exudates (septic and nonseptic) depending on the protein content, specific gravity, total nucleated cell count and diverse cell types. However, some times it is difficult to place samples in one of these categories. Fluid patterns also include chylous and hemorrhagic effusions.

Transudates are clear, colorless, effusions with low protein (< 1.5 g/dL) concentration, low number of nucleated cells (< 1500/μL) and specific gravity less than 1013. Common cell types found are few nondegenerate neutrophils and large mononuclear cells (macrophages and mesothelial cells). Possible causes of transudates include: right-side heart failure, pericardial disease, lymphatic obstruction, hypoalbuminemia (e.g., due to renal glomerular disease, hepatic insufficiency, protein-losing enteropathy etc), leakage of low-protein lymph from intestinal lymphatics (secondary to obstruction of intestinal lymph flow by masses), etc.

Modified transudates are cellular fluids that vary in color (from colorless to milky white, red, reddish brown, or even yellow depending on the cause), have high protein concentration (2.5-7.5 g/dL) and low to moderate number of cells (1000-7000 nucleated cells/μL). Common cell types present are nondegenerate neutrophils, large mononuclear cells, small lymphocytes. Causes of modified transudates are many including fluid leakage from lymphatics with high-protein lymph or blood vessels, cardiovascular diseases, liver diseases, neoplasia, urinary bladder rupture, peripheral blood contamination, etc. In many cases, modified transudates may represent a transition stage to exudates as the disease develops.

Chylous effusion classically has been described as a milky-white pleural fluid that fails to become clear upon centrifugation, occurring secondary to many causes (as a result of lymph leakage mainly into the pleural or the peritoneal space). Pseudochylous effusion is typically associated with chronic inflammatory thoracic disease or heart failure or malignancy, resembles chylous effusion grossly, but does not contain chylomicra, clears upon centrifugation and contains cholesterol and protein-lecithin compounds (from cell membrane degeneration).

Exudates vary from amber to white to red and turbid to cloudy, have high protein (> 3.0 g/dL) and very high level of nucleated cells (> 7000 cells/μL). Common cell types seen are neutrophils which may be mainly healthy (non septic exudates) or degenerate and toxic (septic exudates) for example in bacterial infections (unless there are small amounts of toxin). Yet, a bacterial infection should be considered even when the neutrophils are nondegenerate. Moreover, neutrophils may develop because of abundant exfoliation of cancer cells typically found in neoplastic effusions. Causes of exudates include inflammatory processes seen during gut perforation, ruptured abscess of liver or prostate, perforated pyometra, focal necrosis of intestine due to intussusception, feline infectious peritonitis, etc.

Hemorrhagic effusions are seen during trauma, hemostatic defects, heartworm infection, neoplasia, as a postsurgical complication, etc. Such fluids must be distinguished from iatrogenic blood contamination. In cytology smears the fresh blood is accompanied by platelets, whereas long standing hemorrhage has no platelets and may contain phagocytized red blood cells.

Although numeric values are important, the systematically performed microscopic evaluation is a critical aspect of the diagnostic procedure; not only does it allow complete classification of the fluid but also the identification of specific cell types (morphology of cells) and moreover microorganisms that might be responsible for the fluid accumulation. Common cells seen in smears from body cavity effusions include mesothelial cells (non reactive & reactive), histiocytes (macrophages), neutrophils, eosinophils, lymphocytes, plasma cells, red blood cells and occasionally other benign cells such as fat cells, pulmonary cells, liver cells, muscle cells, etc. Miscellaneous findings include microfilariae, glove powder, nuclei from rupture cells, etc.

The mesothelial cells line the pleural, peritoneal, and visceral surfaces and are very often found in effusions as single cells or in clusters. Their size is variable but their morphology is similar. They have usually one centrally located round nucleus with a prominent nuclear membrane and a fine reticular chromatin pattern and moreover nucleoli may be evident in reactive cells. The activated mesothelial cells have a characteristic basophilic cytoplasm, a blue or red corona and show possibly phagocytic activity. Binucleated and multinucleated forms are often seen, but mitotic figures are occasional and normal. Pseudoglandular structures or rosettes can be observed. The macrophages have single oval to bean shaped nucleus, but may be variable or even multinucleated, lacy nuclear chromatin and they have frequently vacuolated cytoplasm indicating phagocytosis.

Forms of neutrophils which could be found include: non-degenerate (marked basophilic chromatin), degenerate (with swollen nucleus that occupies most of the cytoplasm, or swollen-loose-homogeneous eosinophilic nuclear chromatin), with toxic changes (Döhle bodies, toxic granulation, cytoplasmic basophilia, foamy cytoplasm).

The lymphocytes are usually in small numbers except in chylous (mature lymphocytes) or lymphosarcomatous (neoplastic lymphoid cells) effusions. Mature lymphocytes are smaller than neutrophils, have small amount of clear-blue cytoplasm, oval to bean-shaped nucleus, clumpy nuclear chromatin and no visible nucleoli. Whereas, reactive lymphocytes are larger than small lymphocytes and have small to moderate amount of very blue cytoplasm. Immature lymphocytes are much larger than neutrophils, have a moderate amount of clear to blue cytoplasm, variably shaped nuclei and one to three nucleoli (e.g., centroblasts, centrocytes, immunoblasts, etc).

Eosinophils have in dogs variably sized round granules whereas in cats rod-shaped granules. Eosinophilia (moderate to large cell numbers) in fluid has been associated with fungal infections, neoplasia (including mast cell tumors, lymphoma, and thymoma), heartworm disease, allergic conditions, and hypersensitivity reactions. Mast cells are seen occasionally in many inflammatory disorders and large numbers could be associated with mast cell tumors within the body cavities. Erythrocytes are often seen secondary to hemorrhage (erythrophagia and hemosiderin granules seen) or peripheral blood contamination (platelets seen in the effusion).

In some instances, malignant cells are found during the course of the disease in effusions from animals with a known history of malignant neoplasia, however in other cases, a malignant effusion is the first indication of cancer. Tumors are primary or metastatic. Most often primary tumors are mesotheliomas, which can be either malignant or benign (less cases). Malignant mesotheliomas present three main histological patterns: a sarcoma-like with spindle-shape cells, a sclerosing pattern and, the most common epithelioid type. Concerning the epithelioid pattern, differential diagnosis must be made with metastasis from an adenocarcinoma located in another site, e.g., mammary gland or ovary. Yet, neoplastic mesothelial cells are difficult to identify cytologically even by expert cytopathologists. In general, mesotheliomas cytologically are characterized by atypical cells having a mesothelial appearance (they show dense cytoplasm, distinct cell borders, increased binucleation and multinucleation and maybe a microvillus edge), increased mitotic figures and increased cellularity. Rare tumors such as lipomas, liposarcomas and ganglioneuromas can also develop from the serous membranes.

Most common primary cancers that metastasize to the pleural and peritoneal cavity are from the lung, thyroid, gastrointestinal tract, mammary gland, ovary, kidney, liver, pancreas, spleen, etc. They are usually epithelial in origin and less mesenchymal (e.g., hemangiosarcomas) or malignant lymphomas and less often mast cell tumors, TVTs, etc.

Morphologic criteria (at least 4) for identifying malignancy of neoplastic cells include the following: variability in size of markedly enlarged nuclei and cells (macrokaryosis, anisokaryosis), increased N:C ratio, macronucleoli, anisonucleoliosis, angular nucleoli, nuclear molding, coarse chromatin pattern, multinucleation, increased mitotic figures, abnormal mitoses, cytoplasmic criteria of malignancy (basophilia, vacuolization, ill-defined margins), presence of cell products such as psammoma bodies, melanin or mucus not produced by mesothelial cells, formation of cell aggregates (such as large papillary or glandular clusters) that are not usually formed by mesothelial cells. In a study, the sensitivity of cytologic evaluation for the detection of malignant tumors in body cavity effusions was 64% for dogs and 61% for cats. Specificity was 99% for canine and 100% for feline effusions.

In conclusion, although cytopathology is of high diagnostic value, cytology findings should always be interpreted in conjunction with the history, signalment, physical findings, and other diagnostic aids in making a definitive diagnosis.

References

1.  Alleman AR. Abdominal, thoracic, and pericardial effusions. Vet Clin North Am Small Anim Pract, 2003, 33(1):89-118.

2.  Baker R and Lumsden JH. Color atlas of cytology of the dog and cat. Mosby, St. Louis, 1st ed, 2000, 159-165.

3.  Braun JP, Guelfi JF, Pages JP. Comparison of four methods for determination of total protein concentrations in pleural and peritoneal fluid from dogs. Am J Vet Res, 2001, 62(3):294-6.

4.  Clinkenbeard KD. Diagnostic cytology: carcinomas in pleural effusions. Comp Cont Educ, 1992, 14(2): 187-195.

5.  Cowell RL, Tyler RD, Meinkoth JH. Abdominal and thoracic fluid. In: Cowell R and Tyler R (eds). Diagnostic cytology and hematology of the dog and cat. Mosby, St. Louis, 2nd ed, 1999, 142-158.

6.  Else RW, Simpson JW. Diagnostic value of exfoliative cytology of body fluids in dogs and cats. Vet Rec, 1988, 123(3):70-76.

7.  Fournel-Fleury C, Magnol JP, Guelfi JF (eds). Color atlas of cancer cytology of the dog and cat. Prat Med Chir Anim Comp, Paris, 1st ed, 1994, 69-141.

8.  Hirschberger J, DeNicola DB, Hermanns W, Kraft W. Sensitivity and specificity of cytologic evaluation in the diagnosis of neoplasia in body fluids from dogs and cats. Vet Clin Path, 1999, 28 (4):142-146.

9.  Meyer DJ and Franks PT. Effusion: classification and cytologic examination. Comp Cont Educ 1987, 9(2): 123-128.

10. Perman V, Osborne CA, Stevens JB. Laboratory evaluation of abnormal body fluids. Vet Clin North Am, 1974, 4(2):255-268.

11. Prasse KW and Duncan JR. Laboratory diagnosis of pleural and peritoneal effusions. Vet Clin North Am, 1976, 6(4):625-636.

12. Shelly SM. Body cavity fluids. In: Raskin RE and Meyer DJ (eds). Atlas of canine and feline cytology. Saunders WB, Philadelphia, 1st ed, 2001, 187-205.