Rodak's Hematology: Clinical Principles and Applications, 5th Ed.

CHAPTER 18. Body fluid analysis in the hematology laboratory

Bernadette F. Rodak,*


Performing Cell Counts on Body Fluids

Preparing Cytocentrifuge Slides

Cerebrospinal Fluid

Gross Examination

Cell Counts

Differential Cell Counts

Serous Fluid

Transudates Versus Exudates

Gross Examination

Differential Cell Counts

Synovial Fluid

Gross Examination

Differential Cell Counts


Bronchoalveolar Lavage Specimens

Procedure and Precautions

Differential Cell Counts


After completion of this chapter, the reader will be able to:

1. Describe the method for performing cell counts on body fluids.

2. Given a description of a body fluid for cell counting, choose the appropriate diluting fluid, select a counting area, and calculate and correct (if necessary) the counts.

3. Using the proper terminology, discuss the gross appearance of body fluids, including its significance and its practical use in determining cell count dilutions.

4. Discuss the advantages and disadvantages of cytocentrifuge preparations.

5. Differentiate between traumatic spinal tap and cerebral hemorrhage on the basis of cell counts and the appearance of uncentrifuged and centrifuged specimens.

6. Identify from written descriptions normal cells found in cerebrospinal, serous, and synovial fluids.

7. Describe the characteristics of benign versus malignant cells in body fluids, and recognize written descriptions of each.

8. Differentiate exudates and transudates based on formation (cause), specific gravity, protein concentration, appearance, and cell concentration.

9. Identify crystals in synovial fluids from written descriptions, including polarization characteristics.

10. Describe the process of obtaining bronchoalveolar lavage (BAL) samples, including safety precautions for analysis; state the purpose of BAL; and recognize types of cells that normally would be found in BAL specimens.


After studying the material in this chapter, the reader should be able to respond to the following case study:

A 33-year-old semiconscious woman was brought to the emergency department by her husband. The previous day she had complained of a headache and had left work early. When she got home, she took aspirin and took a brief nap, and she reported she felt better that evening. Her husband stated that the next morning “she couldn’t talk,” so he brought her to the emergency department. A spinal tap was performed. The fluid that arrived in the laboratory was cloudy. The WBC count was 10.6 × 109/L. Most of the cells seen on the cytocentrifuge slide were neutrophils.

1. When multiple tubes of CSF are obtained, which tube should be used for cell counts?

2. What dilution should be made to obtain a satisfactory cytocentrifuge slide?

3. What should you look for on the cytocentrifuge slide?

4. What is the most likely diagnosis for this patient?

The analysis of body fluids, including nucleated blood cell count and differential count, can provide valuable diagnostic information. This chapter is not intended as a comprehensive treatment of all body fluids, but it covers cell counting and morphologic hematology. The fluids discussed in this chapter include cerebrospinal fluid (CSF), serous or body cavity fluids (pleural, pericardial, and peritoneal fluids), and synovial (joint) fluids. Bronchoalveolar lavage (BAL) specimens are discussed briefly.

Performing cell counts on body fluids

Examination of all fluids should include observation of color and turbidity, determination of cell counts, and white blood cell (WBC) evaluation. Blood cell counts should be performed and cytocentrifuge slides should be prepared as quickly as possible after collection of the specimen, because WBCs begin to deteriorate within 30 minutes after collection.1 It is important to mix the specimen gently but thoroughly before every manipulation (i.e., counting cells, preparing any dilution, and preparing cytocentrifuge slides). Cell counts on fluids usually are performed using a hemacytometer (Chapter 14); however, many automated instruments now are capable of performing blood cell counts on fluids. Even specimens with low counts can be reported as long as the WBC count is above the limits of linearity of the instrument and no cellular interference flags are noted.2-5 Each instrument manufacturer should provide a statement of intended use that defines which body fluids have been approved by a regulatory agency for testing on the instrument.24Care should be taken to observe the operating limits of these instruments—that is, the analytical measurement range (AMR) and the volume limits for the fluid received. Red blood cell (RBC) counts on serous and synovial fluids have little clinical value;6 relevant clinical information is obtained merely from the appearance of the fluid (grossly bloody, bloody, slightly bloody).

Cell counts are performed with undiluted fluid if the fluid is clear. If the fluid is hazy or bloody, appropriate dilutions should be made to permit accurate counts of WBCs and RBCs. The smallest reasonable dilution should be made. The diluting fluid for RBCs is isotonic saline. Diluting fluids for WBCs include glacial acetic acid to lyse the RBCs, or Türk solution, which contains glacial acetic acid and methylene blue to stain the nuclei of the WBCs. Acetic acid cannot be used for synovial fluids because synovial fluid contains hyaluronic acid, which coagulates in acetic acid. A small amount of hyaluronidase powder (a pinch, or what can be picked up between two wooden sticks) or one drop of 0.05% hyaluronidase in phosphate buffer per milliliter of fluid should be added to the synovial fluid sample to liquefy it before performing cell counts or preparing cytocentrifuge slides. Dilutions should be based on the turbidity of the fluid or on the number of cells seen on the hemacytometer when using an undiluted sample.

A WBC count of approximately 200/μL or an RBC count of approximately 400/μL3 causes a fluid to be slightly hazy. If the fluid is blood-tinged to slightly bloody, the RBCs can be counted using undiluted fluid, but it is advisable to use a small (1:2) dilution with Türk solution (or similar) to lyse the RBCs and provide an accurate WBC or nucleated cell count. If the fluid is bloody, a 1:200 dilution with isotonic saline for RBCs and either a 1:2 or a 1:20 dilution with Türk solution for nucleated cells should be used to obtain an accurate count. When performing dilutions for blood cell counts, a calibrated pipette should be used, such as MLA pipettes or the Ovation BioNatural pipettes (VistaLab Technologies, Inc. Brewster, NY).

The number of squares to be counted on the hemacytometer should be determined on the basis of the number of cells present. In general, all nine squares on both sides of the hemacytometer should be counted. If the number of cells is high, however, fewer squares may be counted.7 Each square equals 1 mm2. The formula for calculating the number of cells (Chapter 14) is:


Guidelines for counting are summarized in . Table 18-1

TABLE 18-1

Guidelines for Counting Fluids











> 200/μL




Dilution for counting cells


1:2 Türk

1:2 Türk

1:20 in Türk solution

1:2 in Türk solution

No. squares to count on hemacytometer




9 or 4

9 or 4




> 400/μL


> 6000/μL

Dilution for counting cells





1:200 in isotonic saline

No. squares to count on hemacytometer

9 large

9 large

9 or 4 large

4 large or 5 small

5 small

Cytospin dilution (0.25 mL [5 drops] of fluid)*


Dilute with saline to 100–200/μL nucleated cell count

Straight or by nucleated cell count

Dilute with saline to 100–200/μL nucleated cell count

Dilute by nucleated cell count; if RBC count > 1 million/μL, make a push smear and differentiate cells that are pushed out on the end

* Expected cell yield (WBC count for number of cells recovered on slide): 0/μL for 0–70, 1-2/μL for 12–100, > 3/μL for > 100.

RBC, Red blood cell; WBC, white blood cell.

Preparing cytocentrifuge slides

The cytocentrifuge enhances the ability to identify the types of cells present in a fluid. This centrifuge spins at a low speed, which minimizes distortion of the cellular elements and provides a “button” of cells that are concentrated into a small area. The cytocentrifuge assembly consists of a cytofunnel, filter paper to absorb excess fluid, and a glass slide. These three components are fastened together in a clip assembly, a few drops of well-mixed specimen are dispensed into the cytofunnel, and the entire assembly is centrifuged slowly. The cells are deposited onto the slide, and excess fluid is absorbed into the filter paper, which produces a monolayer of cells in a small button (). Figure 18-1


FIGURE 18-1 Wright-stained cytocentrifuge slide showing a concentrated button of cells within a marked circle. Source: (Rodak BF, Carr JH: Clinical hematology atlas, ed 4, St. Louis, 2013, Saunders.)

Although there is some cell loss into the filter paper, this is not selective, and an accurate representation of the types of cells present in a fluid is provided. There also may be some distortion of cells as a result of the centrifugation process or crowding of cells when high cell counts are present. To minimize distortion resulting from overcrowding of cells, appropriate dilutions should be made with normal saline before centrifugation. The basis for this dilution should be the WBC count or the nucleated cell count. A nucleated cell count of 200/μL or fewer provides a good basis for the differential. If the RBC count is extremely elevated, a larger dilution may be necessary; however, an RBC count of 5000/μL would not cause significant nucleated cell distortion. If a fluid has a nucleated cell count of 2000/μL and an RBC count of 10,000/μL, a 1:10 dilution should be made, which produces a nucleated cell count of 200/μL and an RBC count of 1000/μL for the cytocentrifuge slide. If the RBC count of a fluid is greater than 1 million/μL3, it is best to make a “push” slide to perform the differential. In this case, the differential should be performed on the cells “pushed out” on the end of the smear instead of in the body of the smear, because that is where the larger, and possibly more significant, cells would be deposited.

If a consistent amount of fluid is used when cytocentrifuge slides are prepared, a consistent yield of cells can be expected; this can be used to confirm the WBC or nucleated cell count. For example, if five drops of fluid (undiluted or diluted) are always used to prepare cytocentrifuge slides, a 100-cell differential count should be obtainable if the WBC or nucleated cell count is equal to or greater than 3/μL3. In all cases, the entire cell button should be scanned before the differential count is performed to ensure that significant clumps of cells are not overlooked. The area of the cell button that is used for performing the differential count is not important, but if the number of nucleated cells present is small, use of a “systematic meander” starting at one side of the button and working toward the other side is best. In case the number of cells recovered is small, the area around the cell button should be marked on the back of the slide with a wax pencil, or premarked slides should be used to prepare cytocentrifuge slides (Figure 18-1).

Cerebrospinal fluid

CSF is the only fluid that exists in quantities sufficient to sample in healthy individuals. CSF is present in volumes of 100 to 150 mL in adults, 60 to 100 mL in children, and 10 to 60 mL in newborns.89 This fluid bathes the brain and spinal column and serves as a cushion to protect the brain, as a circulating nutrient medium, as an excretory channel for nervous tissue metabolism, and as lubrication for the central nervous system. CSF is collected by lumbar puncture using either the L3-4 or L4-5 interspace (Figure 18-2).6


FIGURE 18-2 Schematic representation of lumbar puncture procedure with spinal needle placed between vertebrae L4-5. Source: (From Brunzel NA: Fundamentals of urine and body fluid analysis, ed 4, St. Louis, 2013, Saunders.)

Gross examination

Normal CSF is nonviscous, clear, and colorless. A cloudy or hazy appearance may indicate the presence of WBCs (greater than 200/μL), RBCs (greater than 400/μL), or microorganisms.89 Bloody fluid may be caused by a traumatic tap, in which blood is acquired as the puncture is performed, or by a pathologic hemorrhage within the central nervous system. If more than one tube is received, the tubes can be observed for clearing from tube to tube. If the first tube contains blood but the remaining tubes are clear or progressively clearer, the blood is the result of a traumatic puncture. If all tubes are uniformly bloody, the probable cause is a subarachnoid hemorrhage. When a bloody sample is received, an aliquot should be centrifuged, and the color of the supernatant should be observed and reported. A clear, colorless supernatant indicates a traumatic tap, whereas a yellowish or pinkish yellow tinge may indicate a subarachnoid hemorrhage. This yellowish color sometimes is referred to as xanthochromia, but because not all xanthochromia is pathologic, the Clinical and Laboratory Standards Institute recommends avoiding the term and simply reporting the actual color of the supernatant (Figure 18-3 and Table 18-2).7


FIGURE 18-3 Flowchart for examination of cerebrospinal fluid (CSF). RBC, Red blood cell; WBC, white blood cell. Source: (Modified from Kjeldsberg CR, Knight JA: Body fluids, ed 3, Chicago, 1993, ASCP Press; reprinted with permission.)

TABLE 18-2

Characteristics of Cerebrospinal Fluid

Traumatic Tap

Pathologic Hemorrhage

Clear supernatant

Colored or hemolyzed supernatant

Clearing from tube to tube

Same appearance in all tubes

Bone marrow contamination


Cartilage cells

Siderophages (may have bilirubin crystals)

Cell counts

When multiple tubes of spinal fluid are collected, the cell count is generally performed on tube 3, or the tube with the lowest possibility of peripheral blood contamination. Tube 1 is used for chemistry and immunology, and tube 2 is used for microbiology. Normal cell counts in CSF are 0 to 5 WBCs/μL and 0 RBCs/μL in adults, and 0 to 30 WBCs/μL and 0 RBCs/μL in neonates. If a high RBC count is obtained, one may determine whether the source of WBCs is peripheral blood contamination by using the peripheral blood ratio of 1 WBC per 500 to 900 RBCs. If peripheral blood cell counts are known, the number of blood WBCs added to the CSF sample can be calculated using the following formula:


where WBCB is the WBC count for peripheral blood, RBCCSF is the RBC count for CSF, and RBCB is the RBC count for peripheral blood. The corrected or true CSF WBC count (WBCCSF) is calculated as follows:7


Some laboratories have questioned the value of an RBC count on CSF and report only the WBC count.

A high WBC count may be found in fluid from patients with infective processes, such as meningitis. In general, WBC counts are much higher (in the thousands) in patients with bacterial meningitis than in patients with viral meningitis (in the hundreds).10-12 The predominant cell type present on the cytocentrifuge slide (neutrophils or lymphocytes), however, is a better indicator of the type of meningitis—bacterial or viral. Elevated WBC or nucleated cell counts also may be obtained in patients with inflammatory processes and malignancies.

Differential cell counts

The cells normally seen in CSF are lymphocytes and monocytes (). In adults, the predominant cells are lymphocytes, and in newborns, the predominant cells are monocytes.Figure 18-489 Neutrophils are not normal in CSF but may be seen in small numbers because of concentration techniques. When the WBC count is elevated and large numbers of neutrophils are seen, a thorough and careful search should be made for bacteria because organisms may be present in very small numbers early in bacterial meningitis (Figure 18-5). In viral meningitis, the predominant cells seen are lymphocytes, including reactive or viral lymphocytes and plasmacytoid lymphocytes (Figure 18-6). However, early in the course of the illness, neutrophils may predominate.10-12 Eosinophils and basophils may be seen in response to the presence of foreign materials such as shunts, in parasitic infections, or in allergic reactions (Figure 18-7).89 When nucleated RBCs are seen, bone marrow contamination resulting from accidental puncture of the vertebral body during spinal tap should be suspected and reported. In the case of bone marrow contamination, other immature neutrophils and megakaryocytes also may be seen. When there is obvious bone marrow contamination, the WBC differential is likely to be equivalent to that of the bone marrow and not that of the CSF.


FIGURE 18-4 Monocyte (left) and lymphocyte (right) seen in normal cerebrospinal fluid (×1000).


FIGURE 18-5 Neutrophils with bacteria in cerebrospinal fluid from a patient with bacterial meningitis (Wright stain, ×1000). Source: (From Rodak BF, Carr JH: Clinical hematology atlas, ed 4, St. Louis, 2013, Saunders.)


FIGURE 18-6 Reactive (viral) lymphocytes in cerebrospinal fluid from a patient with viral meningitis (×1000).


FIGURE 18-7 Eosinophil (A), lymphocytes (B), monocyte (C), and neutrophil (D) in cerebrospinal fluid from a patient with a shunt (×1000).

Ependymal and choroid plexus cells, lining cells of the central nervous system, may be seen. These are large cells with abundant cytoplasm that stains lavender with Wright stain. They most often appear in clumps, and although they are not diagnostically significant, it is important not to confuse them with malignant cells (). Figure 18-8


FIGURE 18-8  Clump of ependymal cells in cerebrospinal fluid (×200).

Cartilage cells may be seen if the vertebral body is accidentally punctured. These cells usually occur singly, are medium to large, and have cytoplasm that stains wine red with a deep wine red nucleus with Wright stain (). Figure 18-9


FIGURE 18-9 Cartilage cells in cerebrospinal fluid (×400).

Siderophages are macrophages (i.e., monocytes or histiocytes) that have ingested RBCs and, as a result of the breakdown of the RBCs, contain hemosiderin. Hemosiderin appears as large, rough-shaped, dark blue or black granules in the cytoplasm of the macrophage. These cells also may contain bilirubin or hematoidin crystals, which are golden yellow and are a result of further breakdown of the ingested RBCs. The presence of siderophages indicates a pathologic hemorrhage. Siderophages appear approximately 48 hours after hemorrhage and may persist for 2 to 8 weeks after the hemorrhage has occurred (Figure 18-10).


FIGURE 18-10 Siderophage with bilirubin crystals (hematoidin) in cerebrospinal fluid (×400).

A high percentage of patients with acute lymphoblastic leukemia or acute myeloid leukemia have central nervous system involvement.89 It is always important to look carefully for leukemic cells (i.e., blast forms) in the CSF of patients with leukemia. Patients with lymphoma, myeloma, and chronic myelogenous leukemia in blast crisis also may have blast cells in the CSF. These blast cells have the characteristics of blast forms in the peripheral blood, including a high nucleus-to-cytoplasm ratio, a fine stippled nuclear chromatin pattern, and prominent nucleoli. They are usually large cells that stain basophilic with Wright stain and have a fairly uniform appearance (Figure 18-11). If a traumatic tap has occurred and the patient has a high blast count in the peripheral blood, the blasts seen in the CSF may be the result of peripheral blood contamination and not central nervous system involvement. The possibility of peripheral blood contamination should be reported and the tap should be repeated in a few days.


FIGURE 18-11 Lymphoblasts in cerebrospinal fluid (×1000). Source: (From Rodak BF, Carr JH: Clinical hematology atlas, ed 4, St. Louis, 2013, Saunders.)

Malignant cells resulting from metastases to the central nervous system may be found. The most common primary tumors that metastasize to the central nervous system in adults are breast, lung, and gastrointestinal tract tumors and melanoma.89 In children, metastases to the central nervous system are related to Wilms tumor, Ewing sarcoma, neuroblastoma, and embryonal rhabdomyosarcoma.9 Malignant cells are usually large with a high nucleus-to-cytoplasm ratio and are often basophilic or hyperchromic. They often occur in clumps but may occur singly. Within clumps of malignant cells, there is dissimilarity between cells, and in multinucleated cells, there may be variation in nuclear size. Clumps of malignant cells may appear three-dimensional, requiring up-and-down focusing to see the cells on different planes, and there are usually no “windows” (clear spaces) between the cells. The nuclei of these cells are usually large, often with abnormal distribution of chromatin, and they may have an indistinct or jagged border, or there may be “blebbing” at the border. Increased mitosis may be shown by the presence of several mitotic figures in the cell button. Malignant cells frequently have a pleomorphic appearance (Figure 18-12and Table 18-3).


FIGURE 18-12 Clump of breast tumor cells in cerebrospinal fluid (×400).

TABLE 18-3

Characteristics of Benign and Malignant Cells



Occasional large cells

Many cells may be very large.

Light to dark staining

May be very basophilic.

Rare mitotic figures

May have several mitotic figures.

Round to oval nucleus; nuclei are uniform in size with varying amounts of cytoplasm.

May have irregular or jagged nuclear shape.

Nuclear edge is smooth.

Edges of nucleus may be indistinct and irregular.

Nucleus is intact.

Nucleus may be disintegrated at edges.

Nucleoli are small, if present.

Nucleoli may be large and prominent.

In multinuclear cells (mesothelial), all nuclei have similar appearance (size and shape).

Multinuclear cells have varying sizes and shapes of nuclei.

Moderate to small N:C ratio

May have high N:C ratio.

Clumps of cells have similar appearance among cells, are in the same plane of focus, and may have “windows” between cells.

Clumps of cells contain cells of varying sizes and shapes, are “three-dimensional” (require focusing up and down to see all cells), and have dark-staining borders.

N:C, Nucleus-to-cytoplasm.

Serous fluid

Serous fluids, including pleural, pericardial, and peritoneal fluids, normally exist in very small quantities and serve as lubricant between the membranes of an organ and the sac in which it is housed. Pleural fluid is found in the space between the lungs and the pleural sac; pericardial fluid, in the space between the heart and the pericardial sac; and peritoneal fluid, between the intestine and the peritoneal sac (). An accumulation of fluid in a cavity is termed an Figure 18-13effusion. When an effusion is in the peritoneal cavity, it also may be referred to as ascites or ascitic fluid.6 It would be difficult to remove these fluids from a healthy individual; the presence of these fluids in detectable amounts indicates a disease state.


FIGURE 18-13 Parietal and visceral membranes of the pleural, pericardial, and peritoneal cavities. Parietal membranes line the body wall, whereas visceral membranes enclose organs. The two membranes are actually one continuous membrane. The space between opposing surfaces is identified as the body cavity (i.e., pleural, pericardial, and peritoneal cavities). Source: (From Brunzel NA: Fundamentals of urine and body fluid analysis, ed 3, St. Louis, 2013, Saunders.)

Transudates versus exudates

As noted, the accumulation of a large amount of fluid in a cavity is called an effusion. Effusions are subdivided further into transudates and exudates to distinguish whether disease is present within or outside the body cavity. In general, transudates develop as part of systemic disease processes, such as congestive heart failure, whereas exudates indicate disorders associated with bacterial or viral infections, malignancy, pulmonary embolism, or systemic lupus erythematosus. Several parameters can be measured to determine whether an effusion is a transudate or an exudate (Table 18-4).

TABLE 18-4

Serous Fluid: Transudates Versus Exudates




Specific gravity

< 1.016

> 1.016


< 3 g/dL

> 3 g/dL

Lactate dehydrogenase

< 200 IU

> 200 IU

White blood cells

< 1000/μL (predominant cell type mononuclear)

> 1000/μL

Protein: fluid-to-serum ratio

< 0.5

> 0.5

Lactate dehydrogenase: fluid-to-serum ratio

< 0.6

> 0.6


Clear or straw-colored

Cloudy or yellow, amber, or grossly bloody


Extremely large

Gross examination

Transudates should appear straw-colored and clear. A cloudy or hazy fluid may indicate an exudate from an infectious process; a bloody fluid, trauma, or malignancy; and a milky fluid, effused chyle in the pleural cavity.

Differential cell counts

The cells found in normal serous fluid are lymphocytes, histiocytes (macrophages), and mesothelial cells. Neutrophils commonly are seen in the fluid sent to the laboratory for analysis but would not be present in normal fluid. When neutrophils are seen, they have more segments and longer filaments than in peripheral blood ().Figure 18-14


FIGURE 18-14 Hypersegmented neutrophil with prominent filaments. Normal appearance of neutrophils in body fluids (×1000). Source: (From Rodak BF, Carr JH: Clinical hematology atlas, ed 4, St. Louis, 2013, Saunders.)

Mesothelial cells are the lining cells of body cavities and are shed into these cavities constantly. These are large (12- to 30-μm) cells and have a “fried egg” appearance with basophilic cytoplasm, oval nucleus with smooth nuclear borders, stippled nuclear chromatin pattern, and one to three nucleoli.89 Mesothelial cells may vary in size, may be multinucleated (including giant cells with 20 to 25 nuclei), and may have frayed cytoplasmic borders, cytoplasmic vacuoles, or both. They may occur singly, in small or large clumps, or in sheets. When they occur in clumps, there are usually “windows” between the cells. The nucleus-to-cytoplasm ratio is 1:2 to 1:3, and this is generally consistent despite the variability in cell size.6 They tend to have a similar appearance to each other on a slide. Mesothelial cells are seen in most effusions, and their numbers are increased in sterile inflammations and decreased in tuberculous pleurisy and bacterial infections (Figure 18-15).8


FIGURE 18-15 A, Mesothelial cells in peritoneal fluid (×200). Note “fried egg” appearance. B, Mesothelial cell with 21 nuclei in pleural fluid (×400).

Macrophages appear as monocytes or histiocytes in serous fluids and may contain RBCs (erythrophages) or siderotic granules (siderophages), or they may appear as signet ring cells when lipid has been ingested and the resulting large vacuole pushes the nucleus to the periphery of the cell (). Figure 18-16


FIGURE 18-16 A, Erythrophage in peritoneal fluid (×1000). B, Signet ring cell (arrow) in peritoneal fluid (×200).

Eosinophils and basophils are not normally seen. These may be present in large numbers, however, as a result of allergic reaction or sensitivity to foreign material.

When large numbers of neutrophils are seen, a thorough search should be made for bacteria. If possible, Gram staining should be performed on a second cytocentrifuge slide to aid in rapid identification if bacteria are found. lists Gram-stained organisms most commonly seen in body fluids. Table 18-5

TABLE 18-5

Gram-Stained Organisms Most Commonly Seen in Body Fluids




Gram-negative diplococci 

Gram-positive cocci 

Gram-negative coccobacilli 

Yeast—stains gram-positive 

Cryptococcus—look for capsule

Serous (peritoneal, pleural, or pericardial)

Gram-positive cocci 

Gram-negative bacilli 

Gram-positive bacilli 

Yeast—stains gram-positive

Synovial (joint)

Gram-positive cocci

Gram-negative bacilli

Gram-negative diplococci

Gram-negative coccobacilli

Note: If the Gram-stained organisms seen in a fluid are not listed above for that fluid, do not report Gram stain results. Save the slide for review.

Lupus erythematosus cells may be seen in serous fluids of patients with systemic lupus erythematosus, because all the factors necessary for the formation of these cells—presence of the lupus erythematosus factor, incubation, and trauma to the cells—exist in vivo. A lupus erythematosus cell is an intact neutrophil that has engulfed a homogeneous mass of degenerated nuclear material, which displaces the normal nucleus. Lupus erythematosus cells can form in vivo and in vitro in serous and synovial fluids and should be reported ().Figure 18-17


FIGURE 18-17 Lupus erythematosus cell (arrow) in pleural fluid (×1000).

Malignant cells are seen in serous fluids from primary or metastatic tumors. They have the characteristics of malignant cells found in CSF (). Figure 18-18Figure 18-19 presents a flow chart for examination of serous fluids.


FIGURE 18-18 A, Clump of tumor cells in pleural fluid (×200). B, Tumor cells and mitotic figure in pleural fluid (×1000). C, Adenocarcinoma cells in pleural fluid (×200). D, Tumor cells in peritoneal fluid (×200). Note cannibalism.


FIGURE 18-19 Flowchart for examination of serous fluid. NCC, Nucleated cell count; RBC, red blood cell; WBC, white blood cell.

Synovial fluid

Gross examination

Synovial fluid is normally present in very small amounts in the synovial cavity surrounding joints. When fluid is present in amounts large enough to aspirate, there is a disease process in the joint. demonstrates placement of the needle for synovial fluid collection from a knee. Normally this fluid is straw-colored and clear. Synovial fluid contains hyaluronic acid, which makes it very viscous. A small amount (pinch) of hyaluronidase powder should be added to all joint fluids to liquefy them before cell counts are performed or cytocentrifuge slides are prepared. If a crystal analysis is to be performed, an aliquot of fluid should be removed for this purpose Figure 18-20before the hyaluronidase is added.


FIGURE 18-20 Schematic of knee demonstrating placement of the needle for synovial fluid aspiration. (From Applegate E: The anatomy and physiology learning system, ed 4, Philadlephia, 2011, Saunders.)

Differential cell counts

Cells found in normal synovial fluid are lymphocytes, monocytes/histiocytes, and synovial cells. Synovial cells line the synovial cavity and are shed into the cavity. They resemble mesothelial cells but are usually present in smaller numbers (). Figure 18-21


FIGURE 18-21 Synovial cells in synovial fluid (×400). Note similarity to mesothelial cells.

Lupus erythematosus cells may be present in synovial fluid just as in serous fluid. Malignant cells are rarely seen in synovial fluid, but when present resemble tumor cells seen in serous fluids or CSF.

Many neutrophils are present in synovial fluid in acute inflammation of joints. As always, a careful search should be made for bacteria when many neutrophils are seen.


Intracellular and extracellular crystals may be present in synovial fluid. Crystal examination may be performed by placing a drop of fluid on a slide and adding a coverslip or by examining a cytocentrifuge preparation. However, the specimen should be fresh, without hyaluronidase added. All synovial fluids should be examined carefully for crystals using a polarizing microscope with a red compensator. The crystals most commonly seen in synovial fluids are cholesterol, calcium pyrophosphate, and monosodium urate.

Cholesterol crystals are large, flat, extracellular crystals with a notched corner.13 They are seen in patients with chronic effusions, particularly patients with rheumatoid arthritis.

Calcium pyrophosphate crystals are seen in pseudogout. These crystals are intracellular and are small rhomboid, platelike, or rodlike crystals.13 The crystals are weakly birefringent when polarized (i.e., they do not appear bright when polarized). When the red compensator is used, calcium pyrophosphate crystals appear blue when the longitudinal axis of the crystal is parallel to the y-axis (Figure 18-22).13


FIGURE 18-22 Intracellular calcium pyrophosphate crystals in synovial fluid (×1000). A, Wright stain. B, Polarized with red compensator. Source: (B courtesy of George Girgis, MT[ASCP], Indiana University Health, Indianapolis, IN.)

Monosodium urate crystals are seen in gout. They are large needlelike crystals that may be intracellular or extracellular. These crystals are strongly birefringent when polarized. When the red compensator is used, monosodium urate crystals appear yellow when the longitudinal axis of the crystal is parallel to the y-axis (Figure 18-23).13 Figure 18-24 presents a flowchart for synovial fluid analysis.


FIGURE 18-23 Intracellular (A) and extracellular (B) monosodium urate crystals in synovial fluid (×1000). A, Wright stain. B, Polarized with red compensator. Source: (A from Rodak BF, Carr JH: Clinical hematology atlas, ed 4, St. Louis, 2013, Saunders. B courtesy of George Girgis, MT[ASCP], Indiana University Health, Indianapolis, IN.)


FIGURE 18-24 Flowchart for examination of synovial (joint) fluid. LE, Lupus erythematosus; NCC, nucleated cell count; RBC, red blood cell; WBC, white blood cell.

Bronchoalveolar lavage specimens

Procedure and precautions

BAL specimens are not naturally occurring fluids; they are produced when the BAL procedure is performed. The procedure consists of introducing warmed saline into the lungs in 50-mL aliquots and then withdrawing it. The specimen received in the laboratory is the withdrawn fluid. The purpose of the procedure is to determine types of organisms and cells that are present in areas of the lung that are otherwise inaccessible. This procedure is performed on patients with severe lung dysfunction. The specimen should always undergo an extensive microbiologic workup and often cytologic examination. It is common to see bacteria, yeast, or both on cytocentrifuge slides prepared from these specimens. Because samples are obtained from the interior of the lung and may contain airborne organisms, care should be taken to avoid aerosol production. Samples should be mixed and containers opened under a biologic safety hood, and a mask should be worn when performing cell counts. Because the risk of performing cell counts and preparing cytocentrifuge slides on BAL specimens outweighs the clinical relevance of the information obtained, some hematology laboratories no longer perform this procedure and defer to information reported from the microbiology laboratory.

Cell counts and cytocentrifuge preparations are performed as with any body fluid. Significant cell deterioration occurs within 30 minutes of collection, with the neutrophils disintegrating most rapidly.

Differential cell counts

The cell types most commonly seen in BAL specimens are neutrophils, monohistiocytes (macrophages), and lymphocytes. Mesothelial cells are not seen in BAL specimens because these cells line the body cavities and not the interior of the lung. Pneumocytes, which can resemble mesothelial cells or adenocarcinoma, may be seen in patients with adult respiratory distress syndrome.

Ciliated epithelial cells can be seen and should be reported because they indicate that the sample was obtained from the upper respiratory tract instead of deeper in the lung. These are columnar cells, with the nucleus at one end of the cell, elongated cytoplasm, and cilia at the opposite end of the cell from the nucleus. They can occur in clusters. If the sample is not aged when the cell count is performed, these cells are in motion in the hemacytometer, because they can be propelled by their cilia ().Figure 18-25


FIGURE 18-25 Ciliated epithelial cells in bronchoalveolar lavage fluid (×100).

Histiocytes laden with carbonaceous material are seen in patients who use tobacco. These cells resemble siderophages in other fluids, but the carbonaceous material is black, brown, or blue-black and is more dropletlike (). Figure 18-26


FIGURE 18-26 Histiocytes with carbonaceous material in bronchoalveolar lavage fluid (×40).

Pneumocystis jiroveci (formerly Pneumocystis carinii) may be seen in specimens from patients infected with human immunodeficiency virus. The P. jiroveci organisms appear as clumps of amorphous material. Close examination of the clumps may reveal cysts (Figure 18-27).


FIGURE 18-27 Cyst of Pneumocystis jiroveci (formerly Pneumocystis carinii) in bronchoalveolar lavage fluid (×100).


• Cell counts and differential counts performed on body fluid specimens are valuable diagnostic tools.

• Calibrated methods must be used when performing cell counts to provide accurate counts.

• To optimize cell morphologic features, specimens should not be overdiluted or underdiluted when cytocentrifuge slides are prepared.

• Normal cell types in any fluid are lymphocytes, macrophages (monocytes, histiocytes), and lining cells (ependymal cells in CSF, mesothelial cells in serous fluids, synovial cells in joint fluids).

• Bacteria and yeast may be seen in any fluid.

• Malignant cells may be seen in any fluid but are rare in synovial fluid.

• Synovial fluid should be examined for crystals using a compensated polarizing microscope.

• BAL specimens are not a true body fluid, but examination of cells present may provide diagnostic information.

Now that you have completed this chapter, go back and read again the case study at the beginning and respond to the questions presented.

Review questions

Answers can be found in the Appendix.

Refer to the following scenario to answer questions 1 and 2: A spinal fluid specimen is diluted 1:2 with Türk solution to perform the nucleated cell count. A total of 6 nucleated cells are counted on both sides of the hemacytometer, with all nine squares counted on both sides. Undiluted fluid is used to perform the RBC count. A total of 105 RBCs is counted on both sides of the hemacytometer, with four large squares on both sides counted.

1. The nucleated cell count is ___/μL.

a. 3

b. 7

c. 13

d. 66

2. The RBC count is ___/μL.

a. 131

b. 263

c. 1050

d. 5830

3. Based on the cell counts, the appearance of the fluid is:

a. Turbid

b. Hemolyzed

c. Clear

d. Cloudy

4. All of the following cells are normally seen in CSF, serous fluids, and synovial fluids except:

a. Lining cells

b. Neutrophils

c. Lymphocytes

d. Monocytes/histiocytes (macrophages)

5. Spinal fluid was obtained from a 56-year-old woman. On receipt in the laboratory, the fluid was noted to be slightly bloody. When a portion of the fluid was centrifuged, the supernatant was clear. The cell counts were 5200 RBCs/μL3 and 24 WBCs/μL. On the cytocentrifuge preparation, several nucleated RBCs were seen. The differential was 52% lymphocytes, 20% neutrophils, 22% monocytes, 4% myelocytes, and 2% blasts. What is the most likely explanation for these results?

a. Bone marrow contamination

b. Bacterial meningitis

c. Peripheral blood contamination

d. Leukemic infiltration in the central nervous system

6. A 34-year-old woman with a history of breast cancer developed a pleural effusion. The fluid obtained was bloody and had a nucleated cell count of 284/μL. On the cytocentrifuge preparation, there were several neutrophils and a few monocytes/histiocytes. There were also several clusters of large, dark-staining cells. These cell clumps appeared “three-dimensional” and contained some mitotic figures. What is the most likely identification of the cells in clusters?

a. Mesothelial cells

b. Metastatic tumor cells

c. Cartilage cells

d. Pneumocytes

7. A serous fluid with a clear appearance, specific gravity of 1.010, protein concentration of 1.5 g/dL, and fewer than 500 mononuclear cells/μL would be considered:

a. Infectious

b. An exudate

c. A transudate

d. Sterile

8. On the cytocentrifuge slide prepared from a peritoneal fluid sample, many large cells are seen, singly and in clumps. The cells have a “fried egg” appearance and basophilic cytoplasm, and some are multinucleated. These cells should be reported as:

a. Suspicious for malignancy

b. Macrophages

c. Large lymphocytes

d. Mesothelial cells

Refer to the following scenario to answer questions 9 and 10: A 56-year-old man came to the physician’s office with complaints of pain and swelling in his left big toe. Fluid aspirated from the toe was straw-colored and cloudy. The WBC count was 2543/μL. The differential consisted mainly of neutrophils and monocytes/histiocytes. Intracellular and extracellular crystals were seen on the cytocentrifuge slide. The crystals were needle-shaped and, when polarized with the use of the red compensator, appeared yellow on the y-axis.

9. The crystals are:

a. Cholesterol

b. Hyaluronidase

c. Monosodium urate

d. Calcium pyrophosphate

10. This patient’s painful toe was caused by:

a. Gout

b. Infection

c. Inflammation

d. Pseudogout


1.  Glasser L. Cells in cerebrospinal fluidDiagn Med; 1981; 4:33-50.

2.  Brown W, Keeney M, Chin Yee I, et al. Validation of body fluid analysis on the Coulter LH 750Lab Hematol; 2003; 9:155-159.

3.  Perné A, Hainfellner J. A, Womastek I, Haushofer A, Szekeres T, Schwarzinger I. Performance evaluation of the Sysmex XE-5000 hematology analyzer for white blood cell analysis in cerebrospinal fluidArch Pathol Lab Med; 2012; 136:194-198.

4.  Yang D, Zhou Y, & Chen B. Performance evaluation and result comparison of the automated hematology analyzers Abbott CD3700, Sysmex XE 2100 and Coulter LH 750 for cell counts in serous fluidsClinic Chimica Acta; 2013; 419:113-118.

5.  Boer K, Deufel T, Reinhoefer M. Evaluation of the XE-5000 for the automated analysis of blood cells in cerebrospinal fluidClin Biochem; 2009; 42:684-691.

6.  Brunzel N. A. Fundamentals of urine and body fluid analysis. 3rd ed. St Louis : Saunders 2013.

7.  Clinical and Laboratory Standards Institute (CLSI). Body fluid analysis for cellular composition proposed guideline. Wayne, PA : CLSI. CLSI document H56–A, vol. 26, no. 26 2006.

8.  Kjeldsberg C, Knight J. Body fluids. 3rd ed. Chicago : ASCP Press 1993.

9.  Galagan K, Blomberg D, Cornbleet P. J, et al. Color atlas of body fluids. Northfield, Ill : College of American Pathologists 2006.

10.  Bartt R. Acute bacterial and viral meningitisContinuum Lifelong Learning Neurol; 2012; 18:1255-1270.

11.  Rotbart H. A. Viral meningitisSemin Neurol; 2000; 20:277-292.

12.  Spanos A, Harrell F. E, Durack D. T. Differential diagnosis of acute meningitisJAMA; 1989; 262:2700-2707.

13.  Strasinger S. K, DiLorenzo M. S. Urinalysis and body fluids. 5th ed. Philadelphia : FA Davis 2008.

*The author would like to acknowledge Leilani Collins for her work on this chapter in the previous edition.