Donald J.

Meuten, DVM, PhD, Diplomate ACVP

College of Veterinary Medicine, North Carolina State University, Raleigh, NC

CYTOLOGIC DIAGNOSIS OF NEOPLASIA

Several preliminary considerations are important in the cytologic diagnosis of neoplasia.

Information regarding the location/size of the lesion and the method of specimen collection are important
in deciding if the specimen is representative of the lesion. Key to this diagnosis is to determine that the
"lesion" is not inflammatory and is not a normal structure. The more neutrophils you see, the more
likely the lesion is inflammatory and not a tumor. True, neutrophils can infiltrate tumors and be present,
but your thumb-rule is still the same -- the greater the inflammation, the less likely there is a neoplasm.
Another general rule is: Every time you get ready to diagnose a tumor with inflammation consider a
granulomatous lesion as the other differential. The large cells you are seeing that you think are neoplastic
may be macrophages. Macrophages in tissues can be large and appear “epithelial,” in fact these types of
macrophages are referred to as “epithelioid macrophages.” If some of these cells in question have
vacuoles of various sizes, and have phagocytosed something, they are probably macrophages and not
tumor cells. When there are numerous mononuclear cells with minimal or no inflammation, then it is
usually one of the following possibilities: a tumor, hyperplasia, or a normal structure. Normal structures
generally yield few cells vs. numerous cells exfoliate from a tumor (usually). The more uniform the
mononuclear cells, the more likely it is a normal structure, or at least “benign.” This is usually not a
problem; most of us do not go around aspirating normal structures! The cytologic features of the cells are
evaluated to classify the tumor as benign or malignant and as to its tissue type, i.e., epithelial,
mesenchymal, or round cell. An attempt can be made to assess the potential malignancy of the tumor by
identifying cytologic criteria of malignancy (discussed below), the most important of which is
variability. If the cells are numerous, but resemble a normal tissue/organ, their appearance must be
critically assessed to decide between normal (perhaps the lesion was missed), benign proliferation (focal
hyperplasia) or a benign tumor. This is really only a dilemma when aspirating internal organs. For
dermal lesions forget hyperplasia and normal structure, you are aspirating a lump and it is not normal.
Decide this: is the dermal lesion inflammatory or not inflammatory. If it’s not inflammatory then decide
if it is a round cell, mesenchymal or epithelial cell tumor, then try to decide if it is benign or malignant.

KEY = the more variable the cells and nuclei, the more likely it is neoplasia; the more uniform they are =
normal, hyperplasia, or adenoma.

Determination of cell type: The shape of the cells and their nuclei, and the presence or absence of
cohesive tendencies are the main criteria used in attempting to classify the cell type of a neoplasm.
Frequently, the cytologic diagnosis cannot be highly specific, but classification as to epithelial,
mesenchymal, or round cell origin and the biologic behavior of the tumor can usually be accomplished.
Three of the best criterion to determine origin of a tumor is location, location and location! Example: if
you aspirate a mass in the urinary bladder and you recognize neoplastic cells then it is a transitional cell
carcinoma until proven otherwise. If you see neoplastic cells in a mass in the distal radius it is an
osteosarcoma until proven otherwise.
Epithelial cells usually exfoliate easily (numerous cells seen) and tend to be shed in clusters.
Cell shape may reflect that of the specific epithelial type (squamous, cuboidal, columnar), but these
characteristics are often lost in poorly differentiated tumors. A single large cytoplasmic-vacuole, marked

1
distension of cytoplasm, and/or acinar-like formations (balls, morulae) indicate a glandular epithelial
origin. Cells tend to be roundish, polyhedral, large and nuclei tend to be round.
Mesenchymal (connective tissue) cells tend to exfoliate poorly (low cell numbers) and are more
prone to exfoliate individually (though poorly organized groups of cells may be closely apposed in highly
cellular specimens). Cell shape tends to be more elongated (spindle-shaped) and nuclei may be oval or
elongated. More specific characterization as to cell type may not be possible unless evidence of some
associated cell product can be found. Extracellular matrix material can provide a clue in myxosarcomas,
osteosarcomas, and chondrosarcomas.
Identification of intracellular pigment commonly allows a specific diagnosis of melanocytic
tumors. If intracellular pigment is seen, the diagnosis is usually easy to make. Purple granules are
characteristic of mast cells and black or greenish granules are features of melanomas (described later in
notes).
Round cell tumors also exfoliate discrete cells, usually lots of them, but they lack the elongate
shape common among connective tissue cells and they don’t form balls or morulae like epithelial tumors.
Lymphoma, mast cell tumor, histiocytoma, transmissible venereal tumor, and plasmacytoma are in
this group. Specimens from these tumors tend to be highly cellular, and characteristic morphologic
features often allow specific cytologic diagnosis (see section on cytology of the skin). Round cells may
lie close together on a slide and this is often misinterpreted as an epithelial cluster. Basal cell tumors are
epithelial but they may look like a round cell tumor if you do not appreciate how cells are arranged in
rows or in morulae.

Cytologic criteria of malignancy: Once a diagnosis of neoplasia is made, the next steps are: 1) Is it
epithelial, mesenchymal or round cell? and, more importantly; 2) Is it benign or malignant? The most
useful cytologic features of malignancy can be grouped as follows: 1) general, 2) nuclear.
General features of malignancy include: overall high cellularity of the specimen; and
heterogeneity or variability of cell features (i.e., generally large, but variably sized and shaped cells and
nuclei). As mentioned before, recognition that a population is “foreign” to the tissue in question is
important. The single-most important feature of malignancy is VARIABILITY. Variable numbers of
nuclei and nucleoli. Variable sizes of cells, nuclei, and nucleoli. Variable shape of cells, nuclei, and
nucleoli.

KEY = VARIABILITY OF CELL FEATURES = Size, Shape, Numbers = MALIGNANCY

Nuclear features of malignancy include: high/variable nuclear to cytoplasmic ratio; variation in

nuclear size and shape; abnormal multinucleation (i.e., variably sized within same cell); variable number
of nuclei/cell; prominent nucleoli of varying size, shape, and number; abnormal mitotic figures; and
irregular and/or coarse chromatin pattern.
No one feature can be singled out as being of primary importance, but, in general, the nuclear
changes are most useful. Highly anaplastic tumors may display many of these abnormalities
simultaneously. The cytologic diagnosis in such cases is relatively easy and very reliable. In general, at
least three or four criteria should be prominent in a high proportion of the cells in question before a
diagnosis of malignancy is made.
The more uniform the cells and nuclei, the more likely the tumor is benign. Most of the round
cell tumors are benign and look benign. An obvious exception to both of these rules is lymphoma. These
tumors are uniform, monomorphic, round, but malignant.

These features are diagnostically significant when present, but their absence does not exclude the
possibility of malignancy. Not all malignant tumors are cytologically anaplastic, and some malignant
tumors are “cytologically” benign e.g., canine thyroid carcinoma, “insulinoma” and canine anal sac
adenocarcinoma all may look benign but behave in a malignant fashion.

2
 The presence in a smear of both an inflammatory response and a dysplastic population of tissue
cells is a sign to proceed with caution. It brings into question the basic character of the lesion, and
complicates the assessment of potential malignancy. Granulomatous inflammation and even
pyogranulomatous inflammation versus neoplasia can still trick me!! They can look a lot like neoplasia!
This dilemma happens frequently in neoplasms of the urinary system because it is lined by transitional
epithelium, which can be quite “dysplastic” even with just inflammation. Histologic assessment will
probably be needed in these situations. With histology, the architectural features such as invasion into
surrounding tissue, lymphatics and/or blood vessels can be used to help identify malignancy.

CRITERIA OF MALIGNANCY BASED ON INDIVIDUAL CYTOMORPHOLOGY

Enlargement of the nucleus = karyomegaly = macrokaryosis - Find a neutrophil (or some
other cell you know) and compare size of neutrophil with nuclei in question. IF the nuclei are bigger
than the neutrophils (15 µ), there are some large nuclei present!
Increase in nucleocytoplasmic ratio - sometimes to 1:1. Large nuclei with little cytoplasm (e.g.
lymphoma).
Nuclear margin (membrane) irregularity - a feature of immaturity that is also a feature of
anaplasia. Multinucleation may occur, but this can also occur in inflammation by monocytes =
multinucleated giant cells.
Anisokaryosis - or anisonucleosis - variation in size and shape of nuclei.
Macronucleoli - Increase in size, number or variable shape to nucleoli. Nucleolus > 5 μm
diameter (canine RBCs are approximately 7 μm in diameter) is highly suggestive of malignant neoplasia.
Abnormal or frequent mitotic figures - Always “exciting” to find but many malignant lesions
may not contain appreciable mitotic figures and non-neoplastic lymphoid and hematopoietic tissues can
have frequent mitotic figures (high cell turnover). Histiocytoma is an example of a benign tumor
(spontaneously regress) with a high mitotic rate.
Cell enlargement - megalocytosis
Hyperchromasia - Increased cytoplasmic basophilia - hyperchromatic cells or cytoplasmic
hyperbasophilia implies an increased RNA content typical of rapid cell synthesis and growth = young
cells; hyperplastic or neoplastic. Cytoplasmic basophilia (hyperchromasia) is not a useful criteria for
neoplasia, as it is commonly observed in hyperplastic cells. It simply means the cytoplasm has an
increased amount of RNA because the cells are "young" and this is a feature of hyperplastic cells, as well
as neoplastic cells. A good example is the reticulocyte or polychromatophilic RBC (young immature
RBC) that is bluish in color because of the increased amount of RNA in its cytoplasm.

**Remember = The most important criterion of malignancy = VARIABILITY!

CYTOLOGIC CRITERIA FOR MALIGNANCY

1. Pleomorphic population of mononuclear cells with few or no inflammatory cells.
2. Large cells (try to find an inflammatory cell for a size reference).
3. Variation in size and shape of nuclei and cytoplasm. (KEY)
4. High nuclei-cytoplasmic ratio - large nuclei.
5. Nuclear abnormalities
- Multiple nuclei, varying numbers
- Large
- Variation in shape and size of nuclei
- Lobation, cleaving, molding, angulation
- Irregular clumping and dark chromatin
6. Nucleolar abnormalities - different numbers, sizes, and shapes per nucleus.

3
Hyperplasia vs. neoplasia can be difficult to assess cytologically. Each shares similar cytologic and
subcellular events (i.e., dark blue cytoplasm for both is due to increased RNA, large nuclei, chromatin
clumping, nucleoli, and mitotic figures are all features of young cells. The key is variability.
Malignant cells have many of the same features seen in hyperplastic cells, but these features are
heterogenous in malignant cells, e.g., variably sized and shaped cytoplasm, nuclei, and nucleoli.

4
 LUMP

Inflammatory Noninflammatory

Neutrophils Eosinophils Mixture Cyst Neoplasia

Abscess Parasitic F.B. Hematoma Granules
Bacteria Immune Fungal Seroma Mast
F.B. Fungal Parasite Abscess Melanoma
Neoplasia Epidermal Round cell
Mucocoele Mesenchymal
Cyst Spindle
Neoplasm Epithelial

NEOPLASIA

Round cell Connective Tissue epithelial-glandular

neoplasia (sarcoma) (carcinoma)
(adenoma)

Exfoliates: readily poorly readily (usually)

Cell groups: individual individual clumping (hopefully)
Cytoplasm: abundant spindle-shaped moderate → abundant
Nucleus: round oval to elongated round

Examples Examples Examples

Lymphoma Fibrosarcoma Sebaceous cell carcinoma
Mast cell tumor Fibroma Sebaceous cell adenoma
Histiocytoma Hemangiosarcoma Mammary adenocarcinoma
Transmissible venereal tumor Hemangioma Mammary adenoma
Plasma cell Liposarcoma Circumanal carcinoma
Basal cell Lipoma Circumanal adenoma
Myeloproliferative disease

Cytology of Lymph Nodes

Do not aspirate normal sized lymph nodes. Enlargement may be generalized in the case of
systemic diseases, or a localized enlargement when the lymph node is draining an area with the primary
problem. Cytologic or histologic evaluation is required to differentiate the most common causes of
“lymphadenomegaly” (big nodes) = lymphoid hyperplasia (“reactive” lymph node), lymphadenitis,
lymphoma and metastatic neoplasia.

5
 The critical question is: Does the animal have lymphoma? Lymphoma (or lymphosarcoma) in
dogs commonly presents as generalized (peripheral) lymph node enlargement, but can occur as regional
cluster or individual lymphadenomegaly. Differentials for lymphadenomegaly are: lymphoma, reactive
hyperplasia, lymphadenitis, or metastatic neoplasm (table) and cytology should identify which is
present.

NORMAL LYMPH NODE CYTOLOGY

Although you should not aspirate normal sized nodes there are two primary situations when you
might do this. One, is there a metastasis to a lymph node that is "draining" an area where a tumor is
located? Two, you are suspicious that an animal has lymphoma but cannot locate the tumor, and want to
determine if it may be in one or more lymph nodes that are not detectably enlarged (rarely to never is the
diagnosis lymphoma if the node is not enlarged).

Cytologic sampling of normal lymph nodes yields 70-90% small mature lymphocytes. These
cells are similar to the lymphocytes observed in peripheral blood: small (approximately 10-15 microns in
diameter) round cells with small amounts of basophilic cytoplasm and a single, round nucleus with
homogenous, dark purple chromatin. The nuclei are rounded, fill the majority of the cytoplasm, and,
occasionally, have a slight indentation on one pole.

Intermixed with the small lymphocytes are low numbers of larger mononuclear cells, consisting
of two populations: 1. prolymphocytes - These cells are 15-25 µm and have large, rounded to slightly
indented nuclei with open chromatin that stains less intensely than the small lymphocytes. There is
usually more cytoplasm than in small lymphocytes, and it is pale basophilic. 2. lymphoblasts – Few in
number; larger (20-35 µm in diameter) more abundant and more basophilic cytoplasm and, large nuclei
with open chromatin (light-violet) and 1-3 prominent nucleoli. It is okay to see immature cells in a
normal node, but they must be in the minority ( few!).

Plasma cells may be present, they are medium size, with moderate to abundant amounts of
deeply basophilic cytoplasm. They have a round, eccentric nucleus with coarsely clumped chromatin and
may have a pale zone next to their nuclei (Golgi apparatus used to package the immunoglobulins). In
tissues that are constantly exposed to antigenic stimulation, i.e., tonsil and mesenteric lymph nodes,
plasma cells will be numerous. A rare eosinophil and even mast cells can be seen in lymph nodes
draining the GI tract. Therefore, before over interpreting a specimen, know what normal is for that
particular tissue. Plasma cells are great cells to see, they are easy to recognize and the more you see the
more likely the lymph node has hyperplasia and not lymphoma!

There also will be low numbers of macrophages, even in a normal lymph node. This should be
expected since there is such close interaction between the lymphoid system and the
monocyte/macrophage system for the processing of antigens. Like plasma cells, macrophages are seen in
greater numbers in lymphoid tissue such as tonsil and mesenteric lymph nodes. Macrohages are 2-3X the
size of lymphoid cells and they are often confused with neoplastic cells.

Numerous red blood cells without erythrophagocytosis indicates trauma to blood vessels during
the aspiration technique. If erythrophagocytosis is present it indicates prior hemorrhage or the removal of
senescent rbcs or immune mediated extravascular hemolysis. If platelets are present, then a blood vessel
was entered, and intravascular blood was aspirated. You don’t see platelets in a hematoma or in an area
of hemorrhage; if platelets are present, then blood was aspirated from a vessel.

6
CYTOLOGY OF REACTIVE LYMPHOID HYPERPLASIA

This is a common and nonspecific finding in lymphoid tissue. Marked lymph node enlargement
can be seen with this problem, especially if the primary problem is a dermatitis. There are numerous
causes of lymphoid hyperplasia because this is a defensive reaction by the body's immune system, i.e.,
dermatitis, septicemia, regional inflammation, metastatic tumor, etc., all stimulate lymphoid hyperplasia.
Cytologic evaluation is very helpful in differentiating between neoplasia and lymphoid hyperplasia.
Rarely is the inciting agent present in the reactive node.

In the majority of the cases, even with extreme reactive lymphoid hyperplasia, the small
lymphocyte is still the predominant cell type seen. The key to the diagnosis is a greater proportion of
plasma cells. Numerous plasma cells lead you away from lymphoma. Macrophages, as well as
prolymphocytes and lymphoblasts, will also increase. Relative numbers of the various cell types are quite
variable, but generally the small lymphocyte is still greater than 70% of the total cellularity. Other cells
that you may identify in a reactive node, and that are not compatible with a diagnosis of lymphoma are:
mast cells, eosinophils, a few neutrophils and "Mott" cells - these are plasma cells filled with packets of
immunoglobulin that are visible as eosinophilic globules (Russel bodies). A heterogeneous cell
population indicates that lymphoma is not present and that reactive hyperplasia is present. You still
may need to determine why the lymphoid system is hyperplastic, but at least lympoma is ruled out.

Immature lymphoid cells will be increased because there is hyperplasia of lymphocytes and
therefore cell division. These large, immature cells will catch your attention and may cause you to
consider lymphoma. However, estimate the mature lymphoid cells compared to these large immature
cells and you’ll see that the mature cells still greatly outnumber immature cells. Don’t consider
lymphoma until the immature cells are greater than 50% of the mature cells. Fortunately, in most cases of
lymphoma, the immature cells will be 80-100% of the cell population. Prolymphocytes and lymphoblasts
should be present in a node with reactive hyperplasia.

CYTOLOGY OF LYMPHADENITIS

Inflammatory disease involving the lymphoid tissue is classified based on the type of
lymphadenitis (suppurative, eosinophilic, granulomatous, etc.) Do not include lymphocytes when trying
to classify the type of inflammation (i.e., "lymphocytic lymphadenitis" is not in my dictionary).

If eosinophils are numerous (>10% of total nucleated cells), then it is an eosinophilic

lymphadenitis; if neutrophils predominate, then it is a purulent lymphadenitis, etc. When these cells are
seen, search for an etiologic agent. Admixed with these inflammatory cells, there will also be numerous
plasma cells and immature lymphocytes. It is characteristic to have some degree of reactive hyperplasia
with the inflammation. In some purulent nodes there may be 100% neutrophils in the aspirate (abscess of
lymph node).

Etiology

eosinophils : parasitic, immune, dermatitis, some tumors – (Produce IL3 and IL5 and stimulate
eosinophilopoiesis)

neutrophils : if degenerate=septic : Staphylococcus, Streptococcus, etc.

if nondegenerate=nonseptic: necrosis, tumor, viral, post antibiotic therapy of a
bacterial infection

7
granulomatous : fungal : blastomycosis, other yeast and hyphal organisms, foreign material
mycobacteriosis – some cases may be 100% macrophages- they are large and filled
with organisisms

CYTOLOGY OF PRIMARY LYMPHOID NEOPLASIA

Lymphoma is the number one differential diagnosis for peripheral lymphadenopathy

(enlargement) without a known cause. In the majority of these cases, cytologic evaluation can give a
specific diagnosis before histopathologic evaluation is even begun. The primary thing to remember is that
in all of the non-neoplastic situations noted above, the cytology has presented with a mixture of various
cell types, and the small lymphocyte was the most common cell seen. With lymphoma, unless in the very
early stages of development (which is rarely seen in clinical settings in which lymph nodes are enlarged),
there is replacement of the lymph node by a monomorphic population of large immature lymphocytes.
The small lymphocyte is now an uncommon cell and lymphoblasts are the number one cell population.
These cells are bigger than neutrophils, have small to moderate amounts of blue cytoplasm, and a large
nucleus with fine chromatin and prominent nucleoli. Plasma cells and other cells are not present or are
rare. The key to the diagnosis is replacement of small lymphocytes by large lymphoblasts. If you need a
reference point for size, find a neutrophil (~15μ) or a red blood cell (7μ) for comparison. At this point,
you are probably done, but if you want to further classify the lymphoma, you can try yourself or you can
send slides to a reference laboratory: lymphoblastic - most common, large cell, light violet
pink/amphophilic chromatin, nucleoli common; prolymphocytic - next most common, medium size,
clumpy chromatin - but not dark like a small lymphocyte, nucleoli uncommon; histiocytic - 3rd on your
list, largest cell type, most abundant cytoplasm, open chromatin = very light, indentations in nuclei,
nucleoli not obvious; lymphocytic - rare, small cell, dense chromatin, nucleoli not visible. Try not to
make a diagnosis of lymphocytic lymphoma because this type of lymphoma is rare in veterinary
medicine. It is probablt just a "normal" lymph node with many small lymphocytes. If you are convinced
there is lymphocytic lymphoma then send slides from at least two lymph nodes to a reference laboratory
for their interpretation. These forms do exist, of course it will be in “cats” and to add to the diagnostic
difficulty it will be in mesenteric lymph nodes. There are other classifications based on cytology,
histology, and distribution. The key is to diagnose lymphoma. T and B lymphoid neoplasms cannot be
identified via routine cytology or histology. This requires the use of flow cytometry or
immunochemistry. However, the single best predictor of response to therapy and survival is B vs T cell
lymphoma. B-cells respond better to chemotherapy and have a longer survival than do T-cell neoplasms
in dogs. In cats the criterion to predict survival and/or response to treatment are not well established.
There is a suggestion that lymphocytic lymphomas in cats respond better to some forms of therapy.

“Lymphoglandular” bodies are pinched off pieces of cytoplasm that appear as anuclear,
basophilic structures scattered in between cells. They are more prominent with lymphoma but can be
seen with lymphoid hyperplasia in reactive lymph nodes. Sometimes, lymphoma cells can look
plasmacytoid. These types have more abundant basophilic cytoplasm, an eccentric round nucleus and a
prominent Golgi. This pattern is especially easy to see in body cavity fluids that contain exfoliated
lymphoma cells.

Another type of lymphoma is the large granular cell lymphoma (LGL). The cells and nuclei are
round, they look lymphoid but the cytoplasm contains a few to many eosinophilic granules. The granules
are usually small and inconspicuous but occasionally can be prominent. When granules are observed one
of the differentials is a mast cell neoplasm. LGL have less cytoplasm, fewer granules, smaller size and no
or few eosinophils as compared to mast cell neoplasms. LGL stain positively with PTAH, and negative
with toluidine blue, mast cell tumors stain just the opposite. LGL’s are more common in cats than dogs,

8
they often involve mesenteric lymph nodes and they are not associated with FeLv. I am unaware of
definitive techniques, or if they even exist, to distinguish LGL from globular leukocyte cell tumors.

As always, correlate the laboratory data with the patient, i.e. age, hypercalcemia, weight loss,
nonregenerative anemia, mediastinal mass, leukemia is actually rare, and lymphopenia is common. In
veterinary medicine, the majority of cases of lymphoma are lymphoblastic.

SUMMARY - Lymphoma is easily recognized when the neoplastic process is fully developed. Key to
the diagnosis is to find 80-100% of the lymphoid cells immature and "blastic". If there are numerous
mature lymphocytes and clearly visible plasma cells don't diagnose lymphoma. The more plasma cells
visible and especially if neutrophils or eosinophils are seen, then the diagnosis is not lymphoma. What
makes lymphoma so easy to diagnose is that the neoplastic population is generally the only cell present in
the preparations. Don't diagnose a mature lymphocytic lymphoma because they are very rare in
veterinary medicine and you're most likely seeing mature non-neoplastic lymphocytes. Don't diagnose
lymphoma in the nose. Nasal lymphoma is rare, and it does not exist as the only place lymphoma could
be found. However, nasal tumors (poorly differentiated sarcomas and carcinomas) can look lymphoid
because of the round shape to the nuclei, little visible cytoplasm and prominent nuclear chromatin. This
is a characteristic pattern for nasal carcinomas that are undifferentiated and they are not lymphomas .
Cutaneous lymphoma can be difficult to differentiate from histiocytoma.

Location - Lymph nodes, skin, liver, spleen, GI, eyes, CNS, reproductive, heart, anyplace. Usually by
the time the tumor is causing problems it can be located on both sides of the diaphragm, cats tend to have
more visceral involvement (thoracic and abdominal) and dogs tend to have more peripheral lymph node
involvement. The concept of nervous, ocular, gastrointestinal, cutaneous, etc. forms of lymphoma is
slightly misleading. In “all” these cases the lymphoma is also located in lymph nodes, someplace!! The
primary clinical problem may be located in one of these systems, but if the diagnosis is lymphoma then
neoplastic cells are present in lymph nodes and neoplastic cells can be found elsewhere in the body.

CYTOLOGY OF SECONDARY NEOPLASIA = METASTATIC TUMOR

Cytology is as accurate as histology in identifying this problem. The discovery of a metastatic

tumor in a lymph node is entirely dependent on features out of your control, i.e., how much of the lymph
node has tumor, is the needle in an affected area? The way to combat these odds is to take multiple
aspirates from the suspect lymph node. Should you aspirate the right area, you will see cytologic
abnormalities compatible with a reactive lymph node, plus variable numbers of the tumor cells. The
tumor cells are usually huge, compared to small lymphocytes (3-10 x their size) and may occur as singles
or clusters. Some tumors have nuclei larger than leukocytes. You are just as likely to find a carcinoma as
a sarcoma in a lymph node; neither has preferential pattern of metastases in lymphatics vs. blood vessels.
If the suspected tumor cells are in clusters, rafts, acini, spheres etc. it is probably a carcinoma. Anytime
you diagnose a metastatic tumor consider granulomatous inflammation as a differential diagnosis (and
vice versa). Macrophages can be large and easily misinterpreted as neoplastic cells. Search for
phagocytosis in the large cells = large vacuoles, abundant cytoplasm and an admixture of other
inflammatory cells – all of which favor granulomatous inflammation rather than metastatic tumor; and
correlate cytologic abnormalities with clinical suspicion.

9
 A draining lymph node is sometimes aspirated to determine if a tumor has metastasized. If the
primary tumor is a melanoma or a mast cell neoplasm then the preparation must contain numerous cells
with granules and preferably the cell will be in groups or clusters. Melanophages and melanocytes can be
found in normal lymph nodes or especially in nodes draining areas of inflammation (dermatitis). So,
simply finding pigmented laden cells is insufficient (but highly suspicious) to confirm a melanoma.
Furthermore, hemosiderin (green, golden brown, yellow, blue) must be differentiated from melanin. Mast
cells are often in hyperplastic lymph nodes, especially if the node is draining a dermatitis (allergic or
parasitic) or is a gastrointestinal lymph node. The latter tend to have numerous plasma cells and other
inflammatory cells (mast cells included) due to the tissue they are responding to. If mast cells are
numerous or in clusters it will increase your confidence of a metastatic pattern.

SUMMARY: HYPERPLASIA VS. LYMPHOMA

“Normal” Hyperplasia Lymphoma

(reactive)
Small Lympocytes >90% >50% <50%
Large lymphocytes <10% <50% >50%
Plasma cells Few Many Few to none
Other cells Few Increased Rare to none
Macrophages
Mast cells
Eosinophils

Common causes of nondiagnostic slides.

Samples of nondiagnostic quality may be your fault or they may be inherent to the tissue. Unlike your
children, your dog or your technician someone might actually be responsible for a screwup! However, the
most likely candidate is you, and that is good news, because it means you can fix it. You simply need to
know what the problem is and what to do about it. The rest of the talk is divided into possible
explanations for nondiagnostic samples.

Sample is of low cellularity. This may be because the target tissue was missed, sampling technique is
bad, inadequate search of the slide for cellular elements, or the tissue is inherently difficult to get
nucleated cells from. The latter situation is true for mesenchymal tumors. Mesenchymal tumors such as a
fibroma or fibrosarcoma may be of low cellularity or the matrix produced by the tumor cells surrounds
the cells and makes it difficult to impossible to aspirate cells. The more benign the spindle cell tumor the
fewer nucleated cells are present and the greater the matrix and therefore the fewer cells you will see in
your preparation. Hemangiosarcomas poise a slightly different problem, they are notorious for producing
a large amount of blood and very few or no tumor cells. This is because the majority of the tumor is
blood-filled space. So in these two instances no one is to blame and your clinical diagnosis may explain
the sample of low cellularity. Tissues or “diagnoses” that are difficult to sample and therefore get
diagnostic material from include: hemangiosarcoma, splenic lesions, nasal tumors, mycotic rhinitis, bone
tumors and some mesenchymal tumors. Approaches to each of these will be covered in our group of talks
but here are a few tips for each one. Hemangiosarcoma- ultrasound guided entry into splenic mass, don’t
worry if sample is just blood, correlate with CBC and signalment. For example, if the CBC contains

10
acanthocytes, shistocytes and inappropriate numbers of nucleated RBC’s for the degree of regenerative
anemia then these findings may be more suggestive for HSA than is the result of aspirational cytology. If
splenectomy is performed then try to find some area of the mass that is white, yellow, tan any color other
than red. Tumor cells are white, not red. If the mass is red and bloody and it looks like a hematoma, then
it is a hematoma, sample elsewhere. Nasal lesions- perform a “rotor rooter” job! You have to get a
sample from deep within the lesion, sampling the nasal discharge is easy but it is usually unrewarding.
Get as many pieces as you can, roll or touch some of these onto a glass slide for cytology, consider
submitting one for fungal culture and put the rest in formalin. The more pieces you have the more likely
to get a positive diagnosis. Bone- examine the radiographs and enter a lytic area avoiding osteo-
productive areas, penetrate the cortex and enter the medullary cavity. Osteosarcomas arise intramedullary;
get at least three pieces or six aspirates. Mesenchymal tumors- if aspirating expect low cellularity
especially if the tumor is benign, use re-directional aspiration, consider pincushion technique, even if
nothing is visible in the syringe or the hub of the needle disconnect the syringe and add 2 cc of air and
squirt the contents onto a slide. If the entire tumor is removed then scrape a freshly cut and blotted
surface of the tumor with a glass slide or scalpel blade, place that material on another slide and make a
smear.

Lymph nodes-it is relatively easy to miss a lymph node that is not enlarged. Only aspirate
enlarged lymph nodes. Normal sized lymph nodes are normal so don’t try to sample them, even if you’re
looking for an occult lymphoma. Don’t do it. Squirt the material onto the slide and if it looks like fat, it
is fat. Stain one slide quickly and if no nucleated cells are seen then fat was aspirated. Repeat the
procedure. The popliteal lymph node is easy to miss and fat is aspirated instead. When aspirating the
submandibular lymph node be aware that the mandibular salivary gland is larger. I often get biopsies and
fine needle aspiration cytology for evaluation of the submandibular lymph node and instead what is
sampled is normal salivary gland. Examine a stained slide shortly after you do the procedure and make
sure there are numerous lymphoid cells present. If there are large cells that look like macrophages or
epithelial cells then the salivary gland was aspirated.
Diffuse versus focal lesion in an organ- if there is a nodule and or a mass in an organ then you
have to get a needle in that mass, that usually means an ultrasound guided technique. Blind aspirates into
organs such as the liver only work when the lesion is diffuse, which can be the case for myeloproliferative
diseases, many lymphomas, vacuolar hepatopathy, severe inflammation, hepatic lipidosis and cholestasis.
Lesions such as tumor nodules in the liver, spleen, urinary bladder and other internal locations require
ultrasound guided sampling. For example transitional cell carcinoma of the urinary bladder in dogs:
approximately 30% of transitional cell carcinomas can be diagnosed from urine cytologic examination,
77% from prostatic or urethra washes and 90% from percutaneous fine needle aspirational cytology of
ultrasound guided sampling. Relying on neoplastic cells that exfoliate into the urine, and that can then be
retrieved in cytologic preparations is insufficient. Similar examples but with less specific enumeration
will be given for tumors located in other tissues.

“Lousy quality” samples can have many causes and here are some. There are no to few intact
cells and there is nuclear debris strewn across the slide. Usually this is from rough cytologic preparation.
If the material is crushed or the slide is pushed too firmly then cells can be lysed and their contents
dragged across the slide. This creates numerous purple strands that look like fungal hyphae or a bowl of
yarn. Cells within an abscess or septic environment may be so degenerate that they are fragile and it is
easy to induce this artifact. A kinder and more gentler approach is the best first step to try. If the sample
is allowed to sit at room temperature for 12 plus hours the cells may start to degenerate and therefore are
prone to this artifact. If the sample was from a fluid, especially if the fluid is septic then the longer it sits
at room temperature the more fragile the cells become.

Cells stain too lightly - If the sample is of very high cellularity, or is too thick then the stain will
not penetrate and all the cells will have a washed out appearance with no cellular detail. Sometimes nuclei

11
can be seen. These slides simply need to be re-stained. If the stain is a diff Quik type stain then only use
the last jar, the blue stain. Be careful when you do this as the present cells are ’primed’ and will readily
accept more stain. Try just dipping the slide two to four times, rinsing off the stain and re-examining.
Repeat this procedure until the slide is stained adequately. If you dip the slide numerous times the cells
will absorb too much of the blue stain and then all the cells are dark blue and cytologic features cannot be
seen.
Failure to blot excessive fluid and blood from the surface of the lesion will dilute out the
nucleated cell population. In fact there may only be blood and fluid to examine. Blot the surface until
fluid and blood no longer absorb into the paper tissue. If you continue to blot and the specimen starts to
stick to the surface of the paper then the surface is too dry. You need to prepare a new cut surface and
start the procedure over.
XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX

Certain microenvironments can induce abnormal cytologic features. Cells exposed to saline
(washes), low protein fluids (CSF) and urine often have a lacy chromatin pattern, many nuclei will be
devoid of cytoplasm and nucleoli may be especially prominent. The microenvironment is such that cells
lyse or are damaged and naked nuclei are usually present. When this happens chromatin gets quite lacy
and nucleoli become obvious. These are not features of the neoplasm but are features of the
microenvironment. Poor technique can produce slides that look like road kill’s, ink blot studies, scratches
and skid marks on a racetrack. Usually these are from trying to get an excessive amount of material onto
the slide. If a little bit of material is good and a few cells are good then a whole bunch of material must
be better. However, there is a point of diminishing returns. If you get an excessive amount of fluid, blood,
tissue and cells then the material is so thick nothing can be seen. Your goal is to prepare a thin film with
excellent cellular detail. This is best accomplished by not putting an excessive amount material onto the
slide. Usually in all slides, even those that are darkly stained there will be a portion that is thin enough
that you can see individual cells. Scratch marks through the smear are usually due to attempts to dry the
slide with a piece of paper, tissue or inadvertently thinking you are wiping fluid off of the bottom of the
slide. Use slides with a frosted glass edge, label them clearly and always put the cellular material on that
surface of the slide. These type of slides are easy to label and you will “always” know which side of the
slide the cells are on.
Field of view is hazy at 40X– you may be looking at the wrong side; try flipping the slide over
and be sure you’re examining the surface of the slide that has cells. If you try to focus from the bottom of
the slide to the top you will be able to see cells at low magnification but at high magnification the field
will be hazy and you not be able to get the cells in focus. If you have the correct surface facing the
objective and the field is still hazy then try placing a dry cover slip on top of the slide and examine with
the 40X. Do not immediately go to oil. High dry objectives (40X) require a cover slip on the surface of
the slide. If you do not put on a cover slip the field will be hazy and you will probably think the objective
is dirty or has oil on it. You have to try this technique to believe it and then you will be a believer.
Rarely to never do you need oil immersion to diagnose a tumor. Oil immersion is for “dotology 101” and
is used to see micro-organisms. If a dry cover slip does not produce a clear field of vision then the
objective may be dirty. If this is the case then clean the objective with an alcohol swab and a Q-tip. The
green colored slide cleaning fluids will never clean the objective adequately; however, most of the time
simply placing a cover slip on the slide will correct this problem. If you insist on using oil immersion then
try to position your objectives such that the high dry objective is not adjacent to the oil immersion
objective. When these two objectives are kept side-by-side on the nose piece then the high dry objective
will invariably be dragged through the oil.
Stain precipitate – if the stains are not changed regularly precipitate will accumulate in the fluids
and appear on the slides. Stain precipitate is blue to purple, it forms individual dots or clumps of material
that are usually in a different plane of focus from the rest of the slide. Most of the diff quick type fluids
will grow microorganisms (fungi and bacteria) overtime. This is another reason to change the fluids or
you will increase your diagnoses of mycotic problems. If you diagnose septic or mycotic disease then

12
there has to be inflammation present. If there is no inflammation and you see fungi or bacteria, especially
if they are extracellular then they are likely to be contaminants. If you want to search for fungal elements
look in the areas that form blue blobs or areas that are too thick to see individual cells. Many times fungal
hyphae from mammals do not stain well but you can see unstained parallel tracks within the thick blobs.
However, hyphae from host species with nucleated red blood cells do stain well and are fairly easy to see
in cytologic and histologic preparations. Fungal infections tend to produce a “mixed” cellular infiltrate
characterized by neutrophils, lymphocytes, macrophages, multinucleated giant cells, plasma cells and
even eosinophils.
If the slide is of good quality such that cells are easily visualized and the cells are intact then
usually a diagnosis is straightforward. A good-quality preparation, free of artifacts is usually easy to
interpret. Any of your veterinary technicians should be able to do this. Keep in mind they can already
aspirate blood, make films and examine them. Someone that can aspirate and make a high-quality
preparation is worth a raise.

Applied cytology: Pearls of wisdom/ Tips from the top

The following are tips for cytologic collection, preparation and interpretation. They cover a variety of
systems and topics. Some are stand-alone and the message can be derived from the written text; others
will be better explained by visual images.

Almost any preparation can be interpreted if the entire slide is examined, even what I refer to as “road-kill
prepartions. Despite this make high quality good preparations with few artifacts and this will greatly help
interpretation. Occasionally I will get a slide that I just cannot interpret and these are usually due to lysis
of all the cells present. I am not sure why this happens, sometimes it is due to excessive trauma during
preparation, or exposure of the slides to formalin or inherent features in the material being examined.

Use Diff-quick type stains and dip a slide in each solution 12 times (I don’t care what the directions say).
Rinse at the end and dry. If the preparation is too thick and cells are stained in adequately then dip the
slide in solution number three (the blue fluid) two more times. Repeat this until the slide is stained
adequately, rinse with water after staining. Be careful not to dip too many times when re-staining because
the material is “primed” and if dipped too many times it will absorb a large amount of the stain and
obscure all details.

I rarely need oil immersion for most cytologic assessments. I do use oil immersion when performing
“Dotolgy 101”, the examination for bacteria, hemobartonella etc. Most individuals are frustrated by high
dry magnification because the field of view is cloudy. Sometimes this is due to oil on the high dry
objective but most of the time this can be corrected by simply placing a dry cover slip on top of the slide.
No oil is needed, the cover slip may move slightly but simply place a cover slip on the glass slide and find
a field of view with low magnification and then proceed to 40 X magnification. 90% of the time the
image will now be clear. This is because those objectives were manufactured in such a way that a cover
slip is required for adequate visualization. You do not need to buy a new microscope, simply buy some
cover slips. Occassionally the high dry objective is dirty, and if so use alcohol swabs with a Q-tip to
clean the objective. Do not soak the objective in alcohol and do not use green colored lense cleaning
fluids as they will not remove the oil.

Sampling a lesion adequately is critical and especially in lymph nodes, bone, nose and suspected
hemangiosarcomas. The only lymph node that ever needs to be aspirated is one that is enlarged.
Aspiration of normal size lymph nodes rarely yields a helpful diagnosis and often these attempts miss the
lymph node and only fat is aspirated. Sometimes a parotid salivary gland is confused with the lymph
node and salivary tissue is aspirated. If fat is aspirated it is readily apparent by clear droplets on the glass
slide and after staining no or very few nucleated cells are seen.

13
 NOSE: Both mycotic rhinitis and nasal neoplasms start deep in the nose, therefore that is where
samples need to be taken from. These cases are often difficult to diagnose, so plan on using ancillary lab
tests. Any “persistent” nasal lesion should be submitted for culture, histopathology, and cytology! Both
mycotic rhinitis and nasal neoplasms start deep in the nose, and samples need to be taken from this deep
location. Submit several cytologic preparations and ask for them to be stained for fungi: PAS or GMS
stains. Fungi are difficult to see in routinely stained preparations from mammals. Fungal hyphae are
usually unstained or light blue with conventional stains. However, hyphae stain well when they are from
birds, reptiles or snakes. The key to an accurate diagnosis of a nasal lesion is adequate sampling: get
deep in the nasal cavity and take multiple pieces. Nasal lymphoma is infrequent in dogs, (occurs in cats)
and it does not exist as the only place lymphoma could be found. However, nasal tumors (poorly
differentiated sarcomas and carcinomas) can look lymphoid because of the round shape to the nuclei, little
visible cytoplasm and prominent nuclear chromatin. This is a characteristic pattern for nasal carcinomas
that are undifferentiated and they are not lymphomas

BONE: The most important reason to aspirate a bone lesion is to determine if the lesion is
neoplastic or infectious. This distinction is easy if the true lesion is aspirated, however, a common
problem in bone tumors is that the majority of the lesion may be non-neoplastic. It consists of a large
amount of reactive woven bone (seen on radiographs as a “sunburst” pattern, or Codman’s triangle), as
well as areas of necrosis, inflammation, and fibrosis. Several tips to help get a good sample for either
aspirational cytology or histology are: 1) take at least three samples from three different areas of the mass;
2) be sure to enter the medullary cavity with the needle; 3) most osteosarcomas arise in the medullary
cavity of metaphyseal bone, therefore take samples from the metaphysis, possibly the epiphysis but not
the diaphysis. If an osteosarcoma (OSA) is in the diaphysis they are often due to a prior fracture, metal
fixation and/or poor healing. If the clinical data indicates there is a lytic and proliferative lesion in the
metaphyseal region of the appendicular skeleton in a large breed dog of middle-age or older, then there is
a 95%+ chance the lesion is an osteosarcoma. Sometimes, the biggest problem is just finding the tumor
cells because so much of the lesion is not neoplastic. If the sample is predominated by mononuclear cells
with some of these features and it was aspirated from a bony lesion, there are really only three likely
options - plasma-cell neoplasia, lymphoma, or osteosarcoma (three bad tumors).

SPLEEN, hemangiosarcoma: Clinicopathologic changes associated with HSA are fragmentation anemia,
acanthocytes, schistocytes, increased nucleated red cells and Howell-Jolly bodies, thrombocytopenia,
neutrophilia and DIC. Often the majority of the mass is non-neoplastic tissue such as hematoma or
necrotic parenchyma and therefore it is difficult to aspirate tumor cells from these lesions. Even in
histologic preparations the majority of the mass is usually hematoma, necrosis and not the tumor.
Ultrasound guided sampling is critical as is redirecting the needle and aspirating from several regions of
the tumor. These tumors yield few neoplastic cells, do not expect to see numerous sarcomatous cells.
When found the neoplastic cells are characterized by large plump spindle-shaped cells with large round to
oval nuclei, and 1-2 prominent, irregularly shaped nucleoli. Anisocytosis, anisokaryosis, pleomorphism
and bizarre cellular atypia can be marked. Cells can vary from unremarkable endothelial cells to
anaplastic undifferentiated “round cells.” Hemorrhage, erythrophagia, hemosiderin pigment, necrotic
debris and neutrophils are frequently present in the background. Cellularity is usually low.
Hemangioma, a benign tumor of vascular endothelial cells, can not be diagnosed via cytology.

LYMPH NODES. Plasma cells = reactive lymph node; get their identity down!!!!! Most lymphomas are
lymphoblastic and easy to diagnose. Diagnose all of these in your practice and send out all the slides.

Lymphoma is easily recognized when the neoplastic process is fully developed. Key to the diagnosis is to
find 80-100% of the lymphoid cells immature and "blastic". If there are numerous mature lymphocytes

14
and clearly visible plasma cells don't diagnose lymphoma. The more plasma cells visible and especially
if neutrophils or eosinophils are seen, then the diagnosis is not lymphoma. “Lymphoglandular” bodies
are pinched off pieces of cytoplasm that appear as anuclear, basophilic structures scattered in between
cells. They are more prominent with lymphoma but can be seen with lymphoid hyperplasia in reactive
lymph nodes.

Don't diagnose a mature lymphocytic lymphoma because they are very rare in veterinary medicine and
you're most likely seeing mature non-neoplastic lymphocytes. Don't diagnose lymphoma in the nose.
Cutaneous lymphoma can be difficult to differentiate from histiocytoma.

15