In-Clinic Hematology: The Blood Film Review

A complete blood count (CBC) is a critical component of the minimum laboratory database for evaluating veterinary patients. The proliferation of in-clinic analyzers facilitates rapid turnaround time that can improve patient care.

However, microscopic evaluation of a blood film is required to not only verify analyzer results but identify critical diagnostic features that analyzers cannot evaluate. Diagnostically essential morphologic abnormalities can be present even in patients with quantitatively normal results for all hematologic parameters.

Evaluating Blood Films

Prepare blood films immediately after an atraumatic sample is collected to avoid the potential for pre-analytical error and in vitro artifacts. Hand dipping using well-maintained Diff-Quik type stains is used with success for in-house staining, with a limitation of poor staining of some mast cell granules. New methylene blue should be on hand for highlighting reticulocytes and Heinz bodies. In reference laboratories, automated stainers are often utilized for a slightly more complex process that enhances some cytologic features, such as chromatin patterns, for more advanced diagnostic evaluation.

Scan the smear at low magnification (10×), and be sure to:

1. Note red and white blood cell densities in the counting area (Figure 1A), which is a few frames back from the feathered edge (Figure 1B), where cells occur in a monolayer; evaluation deeper in the smear can be more difficult (Figure 1C)
2. Note the presence of:
   1. Rouleaux (Figure 2): “Coin stacks” indicative of inflammation or hyperproteinemia that disperse with addition of saline. A mild degree of rouleaux formation is common in cats, and the amount present in Figure 2 would be considered normal in cats
   2. Agglutination (Figure 3): “Grape clusters” or doublets and triplets indicative of immune-mediated interactions that do not disperse with saline
3. Observe any bias in cell distribution, such as concentration of leukocytes at the feathered edge that may bias cell count estimates
4. Evaluate for the presence of microfilaria (Figure 4)
5. Identify platelet clumps that might artifactually decrease platelet numbers (Figure 5).

Next, move to a higher magnification (50× or 100×) within the counting area, and be sure to:

1. Perform a manual differential cell count to verify the analyzer results because this data can be misleading, especially when there are morphologic abnormalities
2. Assess red cell morphology, including the presence of inclusion bodies or parasites
3. Assess white cell morphology
4. Note platelet density to verify analyzer data
5. Note small platelet clumps and the presence of large platelets.

FIGURE 1. Low magnification canine blood film (200×, Wright-Giemsa stain) illustrating the counting area of the slide, which contains red blood cells (RBCs) in a monolayer with minimal overlap; leukocytes present are minimally distorted. The image is taken at 200× instead of 100× due to distortion caused by the microscope imaging program (A). Canine blood film at the feathered edge of the slide is too distorted to easily evaluate cell morphology. White blood cells (WBCs) can become distorted and RBCs can appear as spherocytes (B). Representative field of the body of a canine blood film, which is too thick to evaluate individual RBC and WBC morphology; RBCs are stacked on each other, with leukocytes compressed or distorted (C).

FIGURE 2. Feline blood film with rouleaux (arrows) present, which appear as stacked erythrocytes (similar to a column of coins); this amount of rouleaux is normal in cats. Wright-Giemsa stain; magnification, 200×.

FIGURE 3. Grapelike clusters of agglutinated RBCs (arrows) can be seen in canine patients with immune-mediated hemolytic anemia. Wright-Giemsa stain; magnification, 200×.

FIGURE 4. Dirofilaria immitis microfilaria at the feathered edge of a canine peripheral blood smear. This patient was positive on heartworm antigen testing. Note that many RBCs have lost central pallor along the feathered edge of the sample, making their appearance very similar to that of spherocytes; the leukocytes appear distorted (arrows). The loss of central pallor is likely secondary to thinning of the blood smear at the feathered edge, and leukocyte distortion is likely secondary to dragging of cells to the feathered edge that can occur during slide preparation. Wright-Giemsa stain; magnification, 200×.

FIGURE 5. Diffuse large platelet clumping along the feathered edge of a feline blood film. Occasionally, platelets stain extremely faint, making them difficult to identify. Wright-Giemsa stain; magnification, 200×.

High Magnification Evaluation By Lineage

Red Blood Cells

Numerous red blood cell (RBC) shape changes are described in textbooks, but clinicians should focus on identification and interpretation of the most diagnostically specific features that are occurring in significant numbers. These features are keys to screening for some of the most important red cell morphologic abnormalities, including:

Anisocytosis and polychromasia are indicative of regeneration. Reticulocyte counts are considered the reference standard for identification and quantitation of a regenerative response, but polychromasia can be used as an estimate, especially in dogs (Figure 6). A low level of polychromasia is normal.¹

Spherocytes, in large numbers, suggest immune-mediated anemia, zinc toxicity, or bee envenomation. Spherocytes are characterized by loss of central pallor and increased cytoplasm density, and appear smaller than other red cells due to their shape change (Figure 6). They are easier to identify and more common in dogs than in cats, and their presence in small numbers is usually nonspecific. Always evaluate for evidence of agglutination (Figure 3) and, if absent, a Coombs test may be indicated.

Heinz bodies in larger numbers (up to 75% of RBCs) are observed in feline metabolic stress, such as diabetes mellitus, hepatic lipidosis, and other conditions, generally without significant hemolysis; however, in dogs, even small numbers of Heinz bodies are considered pathologic. Large Heinz bodies are relatively easy to identify as knob-like extensions emerging from the margin of red cells (Figure 7), but smaller Heinz bodies can be difficult to identify and may appear as small refractile areas of cytoplasm. Staining with new methylene blue can highlight smaller Heinz bodies, as well as identify reticulocytes using a manual method.

Heinz bodies form as the result of oxidative damage and can be important indicators of exposure to toxic amounts of acetaminophen, zinc, onions, garlic, and leeks, among other toxic substances. A careful history that evaluates potential exposure is critical, and radiographs to identify metal foreign bodies may be indicated. Eccentrocytes—red cells with hemoglobin that has pulled away from the cell membrane—also indicate oxidative damage and may accompany Heinz bodies (Figure 8).

Schistocytes, in large numbers, are often reflective of vascular pathology, including potential for disseminated intravascular coagulation and hemangiosarcoma. These fragmented erythrocytes, which can occur nonspecifically in small numbers, are also associated with fragmentation anemia (Figure 9).

Ghost cells are observed with intravascular hemolysis. These cytoplasm-free membranes are more rarely seen as artifacts (Figure 10).

Acanthocytes, red cells with irregular projections, are associated with many underlying conditions, including metabolic derangements, vascular abnormalities (eg, hemangiosarcoma), and liver disease (Figure 9).

Echinocytes are often associated with drying artifact or electrolyte abnormalities. These cells have numerous, even sharp, cytoplasmic projections. They may be present in large numbers, but are not critical from a diagnostic perspective (Figure 7).

Nucleated RBC precursors are released from the marrow as part of a regenerative process, but may also signal endothelial damage (ie, sepsis, thermal), lead toxicity, architectural disruption of hematopoietic organs (eg, spleen), or hematopoietic neoplasia (Figure 11). Large numbers—in the absence of a regenerative response or obvious sepsis or hyperthermia—indicate that a pathologist should evaluate the blood film. Typically, these are metarubricytes, though earlier precursors can sometimes be observed.

Inclusions that may be present:

• Howell-Jolly bodies, dense, round purple inclusions, which are common and represent retained nuclear material normally observed in cats, or associated with increased red cell turnover or decreased splenic function.
• Red cell parasites, which may appear as pyriform (Babesia species), flat, or round inclusions on the cell surface that may detach with time if smears are not prepared immediately after sample collection (Mycoplasma species) (Figure 12).
• Viral inclusions, such as those occasionally seen in the acute phase of canine distemper, which are rare but diagnostically invaluable (Figure 13).

FIGURE 6. RBC morphology: Note anisocytosis due to presence of spherocytes (arrows) and polychromatophils (arrowheads). Normal erythrocytes have eosinophilic cytoplasm and central pallor. Spherocytes appear smaller, darker, and lack the central pallor typically noted in RBCs. Spherocytes can be difficult to identify in feline patients because their RBCs generally lack central pallor. Additionally, in some instances, all RBCs present appear as spherocytes in dogs, which makes identification difficult since typical RBCs are not present for comparison. Polychromatophils appear more basophilic and are frequently larger than typical RBCs. Wright-Giemsa stain; magnification, 1000×.

FIGURE 7. Heinz bodies (arrows) in a cat secondary to metabolic stress. Echinocytes (arrowheads) are spiked, regularly spaced projections off the borders of RBCs. Echinocytes most frequently represent drying artifact, but are also seen with electrolyte derangements, renal disease, or secondary to snake envenomation. Wright-Giemsa stain; magnification, 1000×.

FIGURE 8. Eccentrocytes (arrows) present in a patient with acetaminophen toxicity. Wright-Giemsa stain; magnification, 1000×.

FIGURE 9. Schistocytes (arrows) and acanthocytes (arrowheads), along with 2 Howell-Jolly bodies, which are circular, basophilic inclusions. Wright-Giemsa stain; magnification, 1000×.

FIGURE 10. Two ghost cells (arrowheads) and numerous spherocytes are revealed in a patient with immune-mediated hemolytic anemia; polychromasia is also present. Wright-Giemsa stain; magnification, 1000×.

FIGURE 11. A single metarubricyte (arrow) and reactive lymphocyte (arrowhead); the metarubricyte has much darker and compact chromatin than the lymphocyte. Wright-Giemsa stain; magnification, 1000×.

FIGURE 12. Numerous small, basophilic, Mycoplasma organisms (black arrows) on the periphery of RBCs and free in the background; a single Howell-Jolly body (arrowhead) is also present, which is much larger and more basophilic than Mycoplasma organisms. Note the single polychromatophil (white arrow) and ghost cell. Wright-Giemsa stain; magnification, 1000×.

FIGURE 13. Distemper inclusions in RBCs (arrows) and neutrophils (arrowheads) are displayed. This example shows distemper inclusions visible in both Wright-Giemsa (A) and Diff-Quik (B) stained samples. The inclusions stain more faintly on the Wright-Giemsa stain than the Diff-Quik stain. Be aware that distemper inclusions are more easily visible with Diff-Quik preparations, and can be extremely difficult to identify in Wright-Giemsa stained preparations due to poor staining characteristics. Magnification, 1000×.

View images of normal leukocyte morphology by visiting the American Society for Veterinary Clinical Pathology’s website (asvcp.org): Click on the Students/Residents tab and select “Download Visual Hematology Guide — Leukocytes as a PDF file” from the new page that opens.

White Blood Cells

Morphologic observations of white blood cells (WBCs) are made while performing a 100-cell differential cell count at high magnification within the counting area.

Toxic change of neutrophils is a common and diagnostically critical morphologic abnormality indicative of inflammation that may be observed even with normal cell counts (Figure 14). Common components include cytoplasmic basophilia, vacuolization, and presence of Döhle bodies—small irregular inclusions that, in small numbers, can be present in healthy cats.

Left shift indicates the presence of granulocyte precursors, mostly band forms in which nuclear segmentation is incomplete (Figure 15). Left shift often occurs along with toxic change, which indicates the release of granulocyte precursors due to an intense demand for inflammatory cells in peripheral tissues.

Reactive lymphocytes are nonspecific indicators of antigenic stimulation. They are characterized by a slightly larger size (in some cases, approximately the size of a neutrophil), increased amounts of cytoplasm that can have enhanced cytoplasmic basophilia, prominent perinuclear clear zone, a few small clear punctate vacuoles, and/or small magenta granules (Figure 16); nucleoli should not be present.

WBC inclusions may indicate an infectious agent. Distemper inclusion bodies can be visualized in leukocytes and erythrocytes (Figure 13); other relevant infectious agents include tick-borne diseases caused by Anaplasma species (Figure 17). Rarely bacteria can be seen within neutrophils or monocytes in septic patients.

High power examination of the feathered edge is recommended to optimize identification of infectious agents. Although leukocyte morphology is often distorted, the concentration of cells usually allows visualization of rare agents. Infectious agents can be identified in patients with quantitatively normal CBC results.

Lymphoblasts (Figure 18), mast cells (Figure 19), and malignant histiocytes are abnormal cells seen with some frequency but not specifically identified by analyzers.

• Immature lymphocytes, in small numbers, are occasionally seen in septic patients; therefore, a pathologist should always review smears containing immature lymphocytes.

• Mast cells are often easiest to identify by examining the feathered edge of the smear, but poorly granulated forms can be challenging to identify. In dogs, mast cells are associated with a number of diseases as well as mast cell neoplasia²; in cats, the presence of mast cells typically indicates visceral mast cell disease, increasing the concern for neoplasia.³

FIGURE 14. WBC morphology: Toxic change revealed in 2 segmented neutrophils; toxic change appears as increased cytoplasmic basophilia, foaminess, and Döhle bodies (arrows). Wright-Giemsa stain; magnification, 1000×.

FIGURE 15. WBC morphology: Two band neutrophils that exhibit toxic change. Wright-Giemsa stain; magnification, 1000×.

FIGURE 16. Canine blood film with 2 reactive lymphocytes with increased size, cytoplasmic basophilia, and perinuclear clearing. Wright-Giemsa stain; magnification, 1000×.

FIGURE 17. Variably shaped Anaplasma morulae (arrows) within segmented neutrophils are present. The patient was antibody negative at time of organism identification. Note that morulae are lighter than neutrophil chromatin. Wright-Giemsa stain; magnification, 1000×.

FIGURE 18. Numerous large lymphocytes with occasional visible nucleoli (arrows) present in a canine patient with acute lymphoblastic lymphoma. Wright-Giemsa stain; magnification, 1000×.

FIGURE 19. Feline blood film with 2 mast cells and a basophil on the feathered edge; a distorted cell lacking discernable margins is also present. Note presence of dark granular debris that is artifact from slide preparation (arrow). Wright-Giemsa stain; magnification, 1000×.

Platelets

Identification of platelet clumps, including visual estimation by examination of a blood film, is essential to verify the accuracy of automated platelet counts because technical problems can interfere with accurate platelet counting.

Large platelets are evidence of platelet turnover, which can reflect destructive or consumptive processes, such as immune-mediated thrombocytopenia, inflammation, and disseminated intravascular coagulation. Keep in mind that an expanding number of breeds have congenital macrothrombocytopenia initially characterized in Cavalier King Charles spaniels; these breeds include Norfolk and Cairn terriers, Chihuahuas, Labrador retrievers, poodles, English toy spaniels, shih tzus, Maltese, Jack Russell terriers, Havanese, boxers, cocker spaniels, bichons frises, and some mixed breeds.

• Due to the fact that the platelet mass in these breeds appears to be relatively normal, with platelet size compensating for the low number of platelets, clinical bleeding is not a characteristic of the syndrome.
• Presence of persistent moderate thrombocytopenia, appearance of large platelets, and lack of history of unusual bleeding likely indicates this syndrome.
• When a plateletcrit is available, it should be normal or just below the reference interval.⁴
• Genetic testing can be performed at Auburn University to confirm this diagnosis in breeds in which this genetic mutation has been characterized.⁵

Common Patterns To Recognize

Regenerative Anemia

Regenerative anemia in the absence of clinical evidence of hemorrhage should prompt careful evaluation of red cell morphology for causes of hemolysis. There is a tendency to presume that hemolysis is the result of immune-mediated hemolytic anemia in dogs, which causes clinicians to overlook less common, but important and treatable, causes that are identified with blood film review, such as zinc toxicity and oxidative and Heinz body anemias (induced by, for example, onions, garlic, leeks, skunk musk, and RBC parasites). Inflammatory leukograms often accompany hemolytic anemia, and platelet numbers should be scrutinized to identify concurrent immune-mediated thrombocytopenia or the potential for disseminated intravascular coagulation.

Nonregenerative Anemia

While mild nonregenerative anemia is frequently a nonspecific response to chronic disease, more severe nonregenerative anemia can indicate bone marrow disease. In addition to evaluating the patient for other cytopenias, it is important to examine the blood film carefully for morphologically abnormal cells that may signal malignancy.

Other Morphologic Abnormalities

Patients with significant inflammation and circulating neoplastic cells may have normal total white cell counts, and automated analyzers perform poorly when identifying morphologic abnormalities. Be sure to avoid the pitfall of assuming normal numbers equate with normal cells and look carefully for:

• Toxic change
• Left shift
• Reactive cells
• Presence of cells, such as lymphoblasts, mast cells, and histiocytes, that signify the need for more diagnostic investigation.

Conclusions and Recommendations

Blood film review is a critical component of in-clinic hematology. A cursory review may be sufficient for healthy animals, while more detailed analysis is required for sick patients. Key components every practice should have in place for in-clinic hematology include:

1. Written guidelines to determine when samples should be sent to a reference laboratory for evaluation
2. Staff training in blood film review, with attention to ongoing educational activities
3. Quality assurance checks, performed by routinely sending samples to a reference laboratory to compare their results with those generated in the practice. Address any significant discrepancies with additional training and re-evaluation of guidelines for reference laboratory verification of results.

Following the guidelines outlined in this article will ensure that in-clinic hematology is a safe and effective tool in your practice.

Key to Terms

CBC = complete blood count; RBC = red blood cell; WBC = white blood cell