Cavitary Effusions in the Dog & Cat: Formation, Analysis, & Classification
ACVIM 2008
Michael A. Scott, DVM, PhD, DACVP; Steven L. Stockham, DVM, MS, DACVP
East Lansing, MI, USA


Pleural, peritoneal, and pericardial effusions form in a wide array of disorders and by several pathologic mechanisms. An understanding of these mechanisms and the analytical approach to recognizing them is helpful in tapping the diagnostic potential of cavitary fluid evaluation. Optimal use of fluid analysis requires the appropriate diagnostic samples and tests, informed diagnosticians, and appropriate interpretation in light of other clinical information (history, physical examination, imaging findings, or other laboratory data). Fluid analysis may then improve patient management by providing a specific diagnosis or clear evidence of a particular pathologic process. Nonspecific findings are common, and although they do not completely clarify the cause of the effusion, they may exclude several mechanisms and disorders from consideration.


 Ascites: fluid accumulated in a serous cavity (typically peritoneal cavity)

 Bilious: related to or containing bile

 Chylous: related to or containing chyle (i.e., intestinal lymph containing lipoproteins formed in enterocytes)

 Effusion: accumulation of fluid in a body space or cavity; fluid accumulates by the process of effusion

 Exudate: effusion produced by increased vascular permeability to plasma proteins because of inflammation

 Exudation: oozing out through pores

 Hemorrhage: escape or loss of blood from blood vessels or heart

 Hydraulic pressure: energy (pressure) of a fluid in motion

 Hydrostatic pressure: energy (pressure) of a fluid at rest

 Lymphorrhage: escape or loss of lymph from lymph vessels

 Modified transudate: a transudate that has been modified by the addition of protein and/or cells

 Transudate: effusion produced by changes in oncotic or hydraulic pressure or by decreased lymphatic drainage

 Transudation: passage of fluid or solute through a membrane because of hydraulic or oncotic pressure gradients

Physiologic Processes

In health, pleural, peritoneal, and pericardial cavities of most species contain small amounts of clear fluid that is enclosed by visceral and parietal layers of mesothelium. The fluid acts as a lubricant and transport medium. Because the volume of this fluid is very low in healthy dogs and cats, there is little information about its cellular and chemical nature. Generally speaking, the fluid is expected to have a low protein concentration and low cellularity with a mixture of mesothelial cells, neutrophils, macrophages, and lymphocytes. Successful collection of cavitary fluid in dogs and cats is indicative of a pathologic effusion, analysis of which may help determine the cause of its formation.

In health, plasma filtrate leaves capillaries, enters interstitial space, and diffuses through mesothelium into a serous cavity, from which fluid is removed by lymphatics and returned to plasma. Lymphatic stomata within the mesothelium, particularly of the parietal pleura and diaphragm, are important in fluid drainage from the pleural and peritoneal cavities. Formation of cavitary effusions is therefore dependent on vascular permeability, mesothelial permeability, lymphatic drainage, and the forces of Starling's law. According to Starling's law, net movement of fluid from capillary to interstitium is directly related to capillary permeability, surface area for fluid movement, and the difference between hydraulic and oncotic pressure gradients across the capillary wall (see Figure 1). Specifically,

Net filtration = LpS[Δ hydraulic pressure--Δ oncotic pressure] = LpS[(Pcap - Pif) - s(πcap - ππ)], where

 Net filtration = net flux of fluid from capillary to interstitium

 Lp = permeability (porosity) of the capillary

 S = surface area available for fluid movement

 s = reflection coefficient for proteins across the capillary wall

 Pcap = hydraulic pressure of plasma

 Pif = hydraulic pressure of interstitial fluid

 πcap = oncotic pressure (colloidal osmotic pressure) of plasma

 ππ = oncotic pressure (colloidal osmotic pressure) of interstitial fluid

Starling's law shows that changes in blood pressure and plasma protein concentration may affect fluid flux, and that fluid flux may vary with tissues because the protein permeability of capillaries varies among tissues. The reflection coefficient expresses the protein permeability of capillaries as a number from 0 (completely permeable to proteins) to 1 (impermeable to proteins), and it approaches 1 in most capillary beds. It is lower in hepatic and pulmonary tissue. Because of differences in permeability to proteins, the interstitial fluid protein concentrations in people are near 1.5 g/dL in skeletal muscle, near 2.0 g/dL in subcutaneous tissue, near 4 g/dL in the intestines, and near 6 g/dL in liver.1 Interstitial oncotic pressures vary accordingly, being nearly equal to plasma oncotic pressure in hepatic interstitium compared to about 30 % of the plasma oncotic pressure in skeletal muscle.2

Figure 1. Illustration of fluid movements because of Starling's law in a typical capillary bed.
Figure 1. Illustration of fluid movements because of Starling's law in a typical capillary bed.


The hydraulic pressure gradient averages about 20.3 mmHg outward over a capillary bed. The oncotic pressure gradient remains relatively constant throughout the capillary bed (about 20 mmHg in most), generating an inward pressure that almost balances the outward hydraulic pressure. The difference in hydraulic and oncotic pressure gradients is actually about 13 mmHg on the arterial side of a capillary bed, thus leading to fluid efflux from the capillary and to the interstitial space. On the venous side, the difference is about -7 mmHg, and thus fluid leaves the interstitial space and enters the capillary. The small net outward pressure of about 0.3 mmHg is less than these values might suggest, because there are more venous capillaries than arterial capillaries, but enough that a transudate would accumulate if lymphatic drainage were impaired. Lymphatic vessels return fluid to blood via the thoracic duct (most of the body) or the right lymphatic duct (parts of head, neck, and thorax). Capillaries are permeable to electrolytes, glucose, urea, creatinine and other nonprotein solutes, so the concentrations of these substances in interstitial fluid and plasma are similar. Modified from Fundamentals of Veterinary Clinical Pathology, SL Stockham and MA Scott, Blackwell, 2008; printed with permission.

Pathologic Processes

Five pathologic processes are responsible for most effusions, either singly or in combination. These are 1) transudation, 2) exudation, 3) hemorrhage, 4) lymphorrhage, and 5) rupture of a hollow organ or viscous.


Transudates typically accumulate when there is increased plasma hydraulic pressure and/or decreased plasma oncotic pressure. They may also accumulate when lymphatic drainage from a cavity is impaired. Transudates are usually protein poor (< 2.0 g/dL), particularly if hypoproteinemia is present, but transudation can create a protein-rich transudate.

Protein-poor transudates are most common in dogs with hepatic cirrhosis or protein-losing nephropathy. An increased hydraulic pressure gradient caused by retention of Na+ and water contributes to transudation in these patients, and presinusoidal portal hypertension contributes in cirrhosis. Protein-poor transudation tends to occur most when hypoproteinemia and hypoalbuminemia are marked (i.e., serum albumin concentration < 1.5 g/dL). Decreased oncotic pressure in these conditions may eventually lead to a decreased oncotic pressure gradient, thus contributing to movement of fluids from plasma to the interstitial space. If lymphatic drainage cannot increase enough to remove the fluid, a transudate will form. It should be noted that people with analbuminemia typically do not have transudative effusions. This is because of compensatory processes including 1) increased synthesis of globulins, 2) reduced intravascular hydraulic pressure, 3) reduced interstitial oncotic pressure, and 4) increased lymphatic drainage (by at least a factor of 10 in people2). However, an acute onset of marked hypoproteinemia may cause transudation because of an insufficient time for adjustments in the oncotic pressure gradient and lymphatic drainage.

Protein-rich transudates typically occur when there is increased plasma hydraulic pressure in the liver or lungs because of venous congestion. In congestive heart failure, decreased cardiac output, increased hydraulic pressure in veins, and increased Na+ and water retention increases the rate of plasma entering the space of Disse because of increased hydraulic pressure in hepatic sinusoids. If there is not a corresponding increase in lymphatic drainage, the protein-rich fluid may move from the liver to the peritoneal cavity. Similarly, hydraulic pressure in portal blood vessels is increased in portal hypertension. Blood enters the abdominal portal venous system from capillary beds in viscera, travels to the liver via the portal vein, flows through the hepatic sinusoids of the hepatic lobule to the central vein, exits the liver via the hepatic vein, and enters the vena cava. Lesions within or adjacent to these vessels or the heart may impair portal blood flow and increase hydraulic pressure within the portal system. When lesions are postsinusoidal with respect to the direction of portal blood flow, as in right heart failure, protein-rich transudates may form.


Exudates form when inflammatory mediators cause increased vascular permeability that allows plasma to ooze from the blood. Inflammatory mediators may also cause vasodilation, allowing more blood to enter the inflamed tissues and thus increase capillary hydraulic pressure. The leakage of proteins into the interstitium reduces the oncotic pressure gradient between plasma and interstitial fluid, so plasma has less ability to retain water. Additionally, pleuritis, peritonitis, or pericarditis may damage mesothelium so that the protein-rich interstitial fluid more easily moves into the cavities. Exudation of protein-rich fluid is accompanied by the migration of leukocytes (mostly neutrophils) into the effusion when chemotactic substances are in the fluid. If inflammation directly involves blood vessels (vasculitis), vessels may become very permeable to plasma proteins. Exudates may be infectious (septic) and caused by bacteria, fungi, viruses (FIP), or protozoa, or they may be noninfectious (nonseptic) and caused by necrotic tissue, sterile foreign material, or the presence of an irritating body fluid such as bile or urine.


Mild hemorrhage is a common component of many exudates because of the vascular damage associated with inflammation, but when the primary reason for an effusion is hemorrhage, the fluid is a hemorrhagic effusion. An effusion Hct of 5 % and a blood Hct of 25 % suggests that 20 % of the effusion's volume is blood. Lymphatic vessels attempt to resorb the effusion's H2O, solutes, and erythrocytes, so the effusion volume and composition change with time.


Lymphorrhage is a term used for the escape of lymph from lymphatic vessels, just as hemorrhage is the escape of blood from blood vessels. Lymphorrhage may be traumatic or nontraumatic. Nontraumatic lymphorrhagic effusions occur when there is 1) lymph stasis, 2) lymphatic hypertension, 3) defective lymphatic valve function because of dilated lymphatic vessels, and/or 4) increased permeability of lymphatic vessels. Lymphorrhagic effusions can be classified as chylous or nonchylous based on the presence or absence of chylomicrons in the effusion. A chylous effusion is produced when chylomicron-rich lymph leaks from lymphatic vessels and enters the pleural cavity to form a chylothorax or the peritoneal cavity to form a chyloabdomen. Chylomicrons are formed in intestinal mucosal cells, enter intestinal lymphatic vessels, and enter peripheral blood via the thoracic duct. When a chylous effusion is present, it indicates that lymphatic vessels somewhere between the small intestine and the thoracic vena cava are damaged. Most common in cats, chylothorax may be caused by neoplasms, cardiomyopathy, heart failure, trauma, lung lobe torsion, and infections, but frequently it is idiopathic. A nonchylous lymphatic effusion is produced when lymph without chylomicrons leaks from lymphatic vessels and enters the pleural cavity or peritoneal cavity. The lesion may involve lymphatic vessels that are not in the drainage path from intestine to thoracic duct and thus do not contain chylomicrons. However, the absence of chylomicrons might indicate a lack of dietary intake of lipids or that the chylomicrons did not persist in the effusion.

Rupture of a Hollow Organ or Viscous

Peritoneal leakage of fluid from the urinary tract, the biliary tract, or the alimentary tract may contribute to peritoneal effusions and induce an inflammatory reaction and exudation that may be septic or nonseptic. The effusions change over time.

Multiple Processes

Many cavitary effusions accumulate because of multiple concurrent processes. For example, neoplasia may induce 1) hemorrhage, 2) necrosis and associated inflammation and exudation, and 3) impaired lymphatic drainage.

Routine Diagnostics

The composition of pleural, peritoneal, and pericardial effusions provides evidence for the pathologic process(es) that caused them. Analysis may reveal a cause for an effusion, but analysis more often provides possible explanations for its formation. If appropriate historical and physical examination findings are known, results of most fluid analyses can be appropriately interpreted by using microscopic examination, the total protein concentration, and the total nucleated cell concentration (TNCC). Other tests may further characterize and clarify the cause of an effusion.

Sample Collection and Processing

As with any laboratory test, the quality of the sample affects the quality of the results. Fluid should be placed into two tubes: one tube that contains EDTA to inhibit clotting, and one sterile tube for possible submission for culturing or chemical analysis. Tubes that are not immediately submitted to a laboratory should be kept cool and delivered to the laboratory within 36 h. Concurrent submission of stained and/or unstained cytologic preparations made from an aliquot of the fresh fluid is recommended to allow evaluation with minimal cell deterioration, organism proliferation, or in vitro phagocytosis. Methods of preparing specimens for microscopic examination depend on the cellularity of the fluid, but all specimens should be air-dried prior to staining.

Direct smear: A small drop of fluid is spread on a glass slide using the wedge technique for making a blood film, being sure that the feathered edge is in the stainable area of the slide. A concentration method is desirable for fluids with a TNCC < 20,000/µL.

Line (stop-flow) preparation: This method is similar to the direct smear, but the push slide is abruptly stopped and lifted so that cells are concentrated in a terminal line. Tipping the slide up to allow the terminal line to flow back often improves cell visibility.

Sediment smear: This concentration method may be useful for effusions that have a TNCC < 20,000/µL. Fluid is centrifuged by methods similar to preparing urine sediment. After removal of most supernatant, the sediment is resuspended in a small amount of fluid, and then a direct smear method is used to distribute cells on a glass slide.

Cytocentrifuge preparation: Cytocentrifuges are routinely used to concentrate fluid constituents onto a glass slide in clinical pathology laboratories, producing excellent samples for fluids of low or moderate cellularity.

Physical Analysis

Color and transparency of a fluid pre- and post-centrifugation reflects its content. Color is affected by pigmented solutes such as bilirubin (yellow to orange), hemoglobin (pink to red to brown), stercobilinogen (brown), and chlorophyll (green). A sediment's color reflects pigments in the fluid's cells or particles (e.g., red for erythrocytes, cream to tan for nucleated cells, and brown for fecal contents). A nontransparent, white, cloudy supernatant indicates the presence of lipoproteins (e.g., chylomicrons).

Microscopic Examination

Microscopic examination of stained cytopreparations of cavitary fluids is frequently the most important part of the fluid analysis, although it often yields little useful information when cavitary effusions are clear and colorless. A subjective estimate or objective enumeration of the nucleated cell differential count is done to determine percentages of each nucleated cell. Cell clumping and irregular cell distribution are common, thus differential counts are estimates. Diagnostic features of cells and extracellular structures such as microorganisms and other material are identified.

Total Nucleated Cell Concentration (TNCC)

Because nucleated cells in pleural and peritoneal fluids include mesothelial cells and potentially other nucleated cells in addition to leukocytes, the TNCC is not a leukocyte concentration. However, the electronic and manual methods of determining leukocyte concentrations will provide adequate TNCCs for most fluids. The hemocytometer method of measuring TNCC is usually recommended if the fluid is grossly abnormal, because the fluid may contain clumps of cells, debris, or other material that could plug the small tubing or orifices of electronic cell counters. The lowest TNCC values (< 1,000/µL) are found in protein-poor transudates. The greatest TNCC values (> 100,000/µL) are found in exudates and neoplastic lymphoid effusions. If the nucleated cells are present in clumps, the measured TNCC will be less than the true concentration.

Refractometric Estimates of Total Protein Concentration

Clinical refractometers measure the refractive index of a fluid, which is then displayed on scales calibrated for total protein in human plasma, specific gravity of urine from healthy people or domestic animals, or total solids in human plasma. On a g/dL basis, most dissolved solids are proteins, but other solids include electrolytes, urea, and glucose. The total protein scale will approximate the true total protein concentration in protein-rich fluids because the altered refractive index is mostly due to proteins. High concentrations of lipids (as seen in chylous effusions) or other nonprotein solutes (e.g., urea in uroperitoneum) will increase a fluid's refractive index and produce falsely increased values. Some EDTA tubes contain an additive that refracts light and causes falsely increased values, particularly with small sample volumes.3 Values of 1.0-2.0 g/dL are typical of protein-poor transudates and early uroperitoneal effusions. Greater values generally reflect greater protein permeability of blood vessels (e.g., exudates and protein-rich transudates).

Hematocrit or Erythrocyte Concentration

The Hct or erythrocyte concentration should be measured when the effusion is pink to red, using the same analytic methods used for blood, but the values may be lower than the assays' detection limits. When hemorrhage is a major contributor to the formation of the effusion, the fluid Hct can approach the animal's blood Hct. The fluid Hct decreases over time because erythrocytes are absorbed by lymphatics and the altered oncotic pressure gradients promote fluid movement from interstitium to the body cavity.

Chemical Analysis

Chemical analysis is not routine, but is indicated in selected cases to assess the concentrations of urea, creatinine, cholesterol, triglycerides, electrolytes, proteins, and occasionally other substances.

Classification by Pathologic Mechanism

An approach has developed in veterinary medicine to classify fluids as transudates, modified transudates, or exudates based on numeric findings related to cell and protein concentrations rather than on pathologic processes responsible for the effusion. Values used for these classifications are varied and somewhat arbitrary. The modified transudate category (rarely used in human medicine) has become a catch-all for fluids with cell and protein concentrations somewhere between obvious protein-poor transudates and obvious exudates, encompassing fluids formed by a wide variety of pathologic processes. For example, neoplastic, chylous, bilious, low-grade exudative, uroperitoneal, and hemorrhagic effusions may all have numeric findings of a so-called modified transudate, but they do not accumulate primarily because of transudation and subsequent modification. Moreover, it is rarely possible to determine that a fluid is a modified transudate at the time it is evaluated. Consequently, diagnostic labels often misrepresent the main process responsible for the accumulation of fluid. We believe that a more desirable approach is to relate fluid analytical findings to the pathologic process(es) that may be responsible for the effusion. For example, interpreting a fluid as a bilious effusion, a uroperitoneum, or a neoplastic effusion is more valuable than classifying it as a transudate, an exudate, or a modified transudate. Specific classification may require special tests (immunostaining, culture, or analysis for lipid, bilirubin, or creatinine concentration) or it may be impossible (e.g., differentiating protein-rich transudates from low-grade exudates). When specific conclusions are not possible, a more general classification is appropriate. However, use of the modified transudate category should be reconsidered.


1.  Guyton AC, et al. Textbook of Medical Physiology, 10th ed. 2000.

2.  Rose BD, et al. Clinical Physiology of Acid-Base and Electrolyte Disorders, 5th ed. 2001.

3.  Estepa JC, et al. Res Vet Sci 2006;80:5.

Speaker Information
(click the speaker's name to view other papers and abstracts submitted by this speaker)

Michael Scott, DVM, PhD, DACVP
Michigan State University
East Lansing, MI

MAIN : ASVCP : Cavitary Effusions
Powered By VIN