Clinical Pathology: Part ll 0195-5616/89 $0.00 + .

20

Cytology of the Canine

Reproductive System

Patrick]. Wright, BVSc, MVSc, PhD,*

and Bruce W. Parry, BVSc, PhDt

EVALUATION OF SEMEN

The examination of canine semen is a valuable aid in the assessment

of reproductive function in the dog. Findings within normal limits indicate
normal function. Findings outside normal limits can indicate disordered
function . The nature of the abnormalities detected, together with the
history and the results of clinical examination, can lead to a diagnosis,
appropriate treatment and management procedures, and to a prognosis.
The detection of abnormalities can explain infertility and further sampling
over time provides a means of monitoring changes in the condition. The
accurate prediction of fertility on the basis of the semen picture alone,
however, is an uncertain undertaking. There are few data correlating the
nature and level of semen abnormalities with fertility in the dog. Assertions
that relate semen picture to fertility tend to be based on generalizations
from poorly documented, anecdotal experiences and from the extrapolation
of findings from other species.
Collection of Semen
Semen can be collected from the dog by electroejaculation, 10 using an
artificial vagina, 8 • 25 or by manual stimulation. Electroejaculation is not done
routinely and requires general anesthesia. The artificial vagina can be a
simple rubber cone with glass collecting tube attached, or a water-jacketed
apparatus similar to that used for the collection of bull semen. The senior
author has found that use of the artificial vagina offers no advantages over
manual stimulation. Both procedures usually are successful in confident,

From the University of Melbourne School of Veterinary Science, Werribee, Victoria, Australia

*Senior Lecturer, Department of Veterinary Clinical Sciences

tDiplomate, American College of Veterinary Pathologists; Lecturer, Department of Veterinary
Paraclinical Sciences

Veterinary Clinics of North America: Small Animal Practice-Vol. 19, No.5, September 1989 851
852 PATRICK J. WRIGHT AND BRUCE W. PARRY

well-adjusted, sexually experienced dogs, even in the absence of a bitch on

heat. In timid, inexperienced, maladjusted animals, the procedures com-
monly, at least initially, are unsuccessful. Measures to improve the respon-
siveness of the latter dogs include the presence of a .bitch, preferably on
heat, and repeated visits, giving the dog opportunity to gain confidence. A
change in venue from a threatening examination room to a pleasant expanse
of grass may be helpful. The use of bitch pheromone, methyl p-hydroxy-
benzoate, 20 may assist.
Semen collection using an artificial vagina involves allowing the dog to
erect and ejaculate into the vagina. The initial stimulation to erection may
be a bitch or manual stimulation. All apparatus used in handling semen
must be warm, clean, dry, and free from detergents and chemical contam-
ination. Cold, detergents, and chemical contamination affect semen quality.
Success at the technique of manual stimulation requires some practice and
some empathy with the dog. The following description of the technique
assumes the veterinarian is right-handed.
The dog should be approached from the left side. He should be placed
in a nonthreatening situation, either on a table for toy breeds, or on the
floor. If on the floor, place the bitch in front of the dog, preferably with a
wall along their right side to limit movement. Gently, without alarming the
dog, using the fore-finger and thumb of the right hand take hold of the
penis just behind the bulbus glandis. The forefinger forms aU that apposes
the lateral and lower aspects of the penis. The thumb apposes the top of
the penis. Stimulation of the penis then involves gentle to firm pressure,
sometimes in pulses, and a small amount of massage of the rear of the
glans. The emphasis is on gentle pressure. When erection commences,
assist the penis to protrude through the preputial orifice by passing the
prepuce back over the erecting penis.
Semen is collected into a warm glass tulip flask held close to the
urethral orifice. Commonly (but not always) the ejaculatory behavior and
sequence is the same as if the dog were mating naturally. Initial partial
erection is followed by thrusting-seeking movements, which are followed
by a period of quiet and then attempts to turn as if to tie. During ejaculation,
maintain pressure on and behind the bulb and perhaps give manual pulses
in response to urethral contractions. The dog may be permitted to turn as
if tying.
Canine semen is ejaculated in three fractions. The first or pre-sperm
fraction, consisting of urethral gland fluid of 0.1 to 2 ml, is ejaculated over
30 to 60 seconds during the initial erection and seeking. The second or
sperm-rich fraction of 0.1 to 4 ml is ejaculated over 1 to 2 minutes, usually
during the period of thrusting. The post-sperm fraction, consisting of 1 to
25 ml of prostatic fluid, is ejaculated over 3 to 35 minutes. It is advisable
to collect all but the initial drops of the pre-sperm fraction because, in
some dogs, the sperm fraction comes promptly. Semen that should be
collected includes some of the pre-sperm fraction, the sperm fraction, and
some of the post-sperm fraction. On cessation of penile stimulation, the
erection usually will subside within 5 to 10 minutes.
Care must be taken to avoid pain or damage to the penis, especially
during thrusting. Placing the thumb of the left hand (holding the collecting
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 853

flask) onto the wrist of the right hand (holding the penis) permits both
hands and the penis to move in concert.
Laboratory Methods for Evaluation of Semen
The normal range of values for semen characters are presented in
Table 1. Some of the variation in dog semen characters relates to variation
in the size of dog. Testicle size, sperm production, and ejaculate volume
are correlated positively with the size of dog. 3
Color. Semen color varies from watery to milky, reflecting the concen-
tration of spermatozoa. Reliance on color can be an inaccurate estimate of
sperm concentration when cellular debris is present.
Motility. Motility is assessed by examining a droplet of semen micro-
scopically (at X 100 to X 400 magnification) on a warm glass slide on a warm
stage (37°C). A cover slip may be used. The percentage of spermatozoa
showing active forward movement is estimated. Dilution of concentrated
semen with warm isotonic saline may be necessary. Swirl patterns may be
seen in the undiluted sperm-rich fraction.
Concentration. The numbers of spermatozoa can be counted using a
hemocytometer. The semen is diluted 1:100 with isotonic saline or formol
saline. 24 Concentration is influenced by testicular size, time since last
ejaculation, and the volume of the post-sperm fraction that was collected.
Live Spermatozoa. The percentage of spermatozoa that is alive can be
estimated using a vital stain (for example, nigrosin-eosin). 24 Dead sperma-
tozoa take up more stain than those that are alive. Three to five drops of
semen are added to 0.5 ml warm buffered nigrosin-eosin solution, mixed
gently, allowed to stand for 3 minutes at 37°C. A drop then is placed on a
glass slide and a smear made by drawing the drop along using another glass
slide held at 45 degrees to the drop. An alternative method involving
mixing of semen and stain on the slide has been described. 26

Table I. Char-acteristics of Normal Dog Semen*

CHARACTER MEAN VALUE RANGE

Volume (ml) 10 1-25

Concentration (spermatozoa x Hf/ml) 125 20-540
Spermatozoa/ejaculate ( x 109) 1.25 0.3-1.9
Motility (%) 85 60-90
Alive(%) 90 > 50
Abnormal spermatozoa (%) <20 5-20
Abnormal heads (%) <10
Abnormal tails/mid-pieces
Proximal droplets (%) <5
Distal droplets (%) <10
Singly bent tails (%) <10
Doubly bent tails (%) <2
Coiled tails(%) <2
Tail-less heads (%) <10
*Adapted from Boucher JH, ' Foote RH, Kirk RW: Cornell Vet 48:67, 1958; Galloway
DB, Norman JR: Proc 9th Int Cong Anim Reprod AI 4:714, 1980; Hancock JL, Rowlands IW:
Vet Rec 61:771, 1949; Harrop AE: Vet Rec 67:494, 1955; Roberts SJ: Veterinary Obstetrics
and Genital Diseases. Ann Arbor, MI, Edwards Brothers, 1971; Taha MB, Noakes DE, Allen
WE: J Small Anim Pract 22:177, 1981
854 PATRICK J. WRIGHT AND BRUCE W. PARRY

Abnormal Mid-pieces and Tails. The percentage of spermatozoa with

abnormal mid-pieces and/or tails can be estimated by studying a suspension
of spermatozoa in buffered formol saline under phase contrast microscopy
(at X 400 magnification). 24 A few drops of semen are added to 0.5 ml warm
buffered formol saline and mixed gently. This is a permanent preparation.
A similar assessment can be made using the nigrosin-eosin stained smear,
but the results are less reliable.
The frequency of the following categories is recorded over a total of
400 spermatozoa-normal with respect to mid-piece and tail; and those
with proximal cytoplasmic droplets, distal cytoplasmic droplets, tail-less
heads, singly bent tails, doubly bent tails, and coiled tails.
Abnormal Heads. The percentage of spermatozoa with abnormal heads
can be assessed by examining stained smears of semen. Smears are made
of drops of fresh semen by drawing out the droplet using another slide held
at 45 degrees to the drop. A variety of stains can be used, including carbol-
fuchsin-eosin, 17 Wright's, and Giemsa.
The frequency of the following categories is recorded over a total of
400 spermatozoa-normal with respect to head morphology; heads that are
narrow, narrow-at-base, pear-shaped, round, large, small, broad, or unde-
veloped; and those with abaxial tail attachment, or with the tail around the
head. Acrosomal abnormalities can be assessed in this preparation, but
better results are obtained using other stains. 7
Cellular Elements Other than Spermatozoa. Other cells can be iden-
tified in hematoxylin-eosin 17 or Romanovsky-stained smears. These elements
include bacteria, red blood cells, white blood cells, epithelial cells from the
tract, and epididymal cells. Occasional white blood cells are normal in dog
semen.
Abnormal Spermatozoa and Semen Pictures
A number of sperm defects and factors resulting in abnormal semen
pictures have been described in detail.
Age of Dog. The semen from immature dogs (around puberty) has
lower concentration of spermatozoa, total sperm numbers, volume, and
motility, and a higher frequency of abnormal spermatozoa (particularly
spermatozoa with cytoplasmic droplets) than the semen from mature dogs. 50
In our clinical experience, a common cause of infertility in old dogs (9
years or more of age) is senile testicular atrophy. In this condition, there
are no, or very few, spermatozoa in the ejaculate.
Size of Dog and Frequency of Collection. Daily sperm production is
related to the size of the testicles, which, in turn, is related to the size of
the dog. Large dogs therefore produce more spermatozoa than small dogs.
Spermatozoa for ejaculation are stored in the tail of the epididymis and the
ductus deferens. Daily ejaculation will result in these reserves being 25
per cent below those of a sexually rested dog, and reduced sperm concen-
tration in ejaculates has been reported with daily3 or twice-daily8 · 52 collec-
tion.
Early ejaculates after sexual rest may have reduced motility, reduced
per cent live spermatozoa, and an increased amount of cellular debris.
Azoospermia Related to Spermatogenic Arrest. This condition has been
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 855

described in a total of over 170 dogs of a wide variety of breeds. 16• 22· 27 • 28 · 39
The dogs had been fertile and were presented at 2 to 5 years of age with a
history of recent infertility. Their libido was normal and in about one third
of cases the testicles were reduced in size. Testicle histology revealed
seminiferous tubules lined only with Sertoli cells and spermatogonia. The
cause is unknown. Recovery was reported in two cases. 16
Oligospermia and Macrophages. This was described in a 2.5-year-old
bull terrier with a history of infertility. 2 The testicles were small. The
semen picture showed oligospermia (20 X 106 spermatozoa per ml), 30 per
cent motility, 88 per cent abnormal spermatozoa, and large numbers of
round macrophages. The etiology was unknown.
Knobbed Spermatozoa. Infertility associated with a high proportion of
spermatozoa showing a "knobbed" appearance because of an acrosome
defect has been noted. 41 Sperm motility and concentration were normal.
Mid-piece Abnormalities. A number of reports of infertility associated
with mid-piece abnormalities are summarized in Table 2.
Testicular Degeneration and Epididymal Damage. The semen picture
associated epididymitis and testicular degeneration caused by Brucella canis
infection has been described well. 18 Initial abnormalities included a reduc-
tion in per cent motility,' increased numbers of spermatozoa with retained
perinuclear sheaths, deformed acrosomes, swollen mid-pieces, protoplasmic
droplets, tail-less heads and bent tails, inflammatory cells, and spermatozoa
that were clumped and attached to inflammatory cells. At 60 to 100 weeks
after infection there were few inflammatory cells, motility was 60 to 70 per
cent, and there were 50 to 70 per cent abnormal spermatozoa (mainly with
coiled tails and detached heads), and head-to-head agglutination of sper-
matozoa was observed. Some dogs had unilateral testicular atrophy, and
removal of the other testicle resulted in azoospermia. (Additional features
of this disease are discussed later in this article.)
lmmotile Cilia Syndrome. The immotile cilia syndrome in humans is

Table 2. Abnormalities of Canine Semen Characterized by Spermatozoa

with Defective Mid-pieces
PREVIOUSLY
BREED AGE (YEARS) FERTILE DESCRIPTION

Greyhound• 5 ? Abnormal spermatozoa (41%). High

incidence of spermatozoa with double
tails or thickened mid-pieces.
Maltese Poodle 32 2.5 No Motility good (75%). Abnormal
spermatozoa (92%); most with
swelling, kinking, or disintegration at
the neck.
Llasa Apso37 9 Yes Motility 70%. Most spermatozoa (87%)
with proximal droplets and mid-piece
ultrastructural defects.
Nine dogs40 * 3-9 Yes Spermatozoa with mid-piece swelling
close to head (4 to 96%). Two
recovered fertility.
*Setter, Labrador Retriever, Staffordshire Bull Terrier, Cavalier King Charles Spaniel
856 PATRICK J. WRIGHT AND BRUCE W. PARRY

associated with structural defects in the cilia of ciliated cells throughout the
body and in the tails of spermatozoa, resulting in immotile spermatozoa.
The syndrome is associated with respiratory disorders because of the failure
of normal ciliary activity to clear the respiratory passages of mucus, inhaled
particles, and cellular debris. A similar syndrome has been described in
dogs. 1• 15 The limited reports of reproductive function in these animals
suggest an association with immotile spermatozoa.
Interpretation of Findings
There are no good comprehensive studies reporting values for semen
characters in normal dogs or values compatible with good fertility, nor are
there comprehensive studies relating levels of semen abnormalities to
infertility. This reflects the few studies reported and the lack of a standard-
ized method for the examination of semen and the reporting of semen
findings .
The mean percentage of abnormal spermatozoa in normal dog semen
is reported to be 5 to 20 per cent, but the types of abnormality are not
recorded (see Table 1). Higher frequencies of abnormal spermatozoa and
types of abnormality have been documented but the fertility of the dogs
was not stated. 56 Somewhat guarded suggestions indicate that infertility will
result if total head and mid-piece abnormalities are over 40 per cent, or if
acrosomal abnormalities and proximal cytoplasmic droplets are over 20 per
cent. 3o, 42
Possible causes of semen abnormalities are presented in the following
sections.30 • 41 • 42
Color. Abnormal color can result from the presence of urine (yellow),
fresh blood (red), hemolysed blood (brown), pus (greenish), or lack of
spermatozoa (clear). The presence of blood may indicate prostatitis, hema-
turia, or hemorrhage caused by trauma associated with semen collection.
Low Concentration of Spermatozoa. Low sperm concentration can
reflect failure of sperm production, sperm transport, or normal ejaculation.
Normal dogs sometimes fail to ejaculate a sperm-rich sample. Further
collection attempts therefore are indicated when the first is unsuccessful.
Causes of decreased sperm production include testicular degeneration,
spermatogenic arrest, and testicular neoplasia. Testicular degeneration may
result from testicular hyperthermia (for example, high ambient temperature,
fever, or scrotal dermatitis), localized or systemic infection, cryptorchidism,
endocrinopathies, stress or old age, testicular hypoplasia (which can occur
in intersex states), 11 a congenital abnormality, and hypopituitarism.
Causes of defective sperm transport include segmental aplasia of the
duct system, blockage of the duct system (for example, with infection or a
spermatocoele), inadequate ejaculation, or retrograde ejaculation into the
bladder. Collection of excess post-sperm fraction, overuse of the dog for
stud work, and immaturity also may be causes of low sperm concentration.
Poor Motility. Poor motility reflects defective or damaged spermatozoa,
or low temperature at examination. Defects can arise in the testicle. Damage
also can occur in a hostile epididymal environment or after collection. Some
morphologically abnormal spermatozoa, with cytoplasmic droplets or mid-
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 857

piece defects, for example, can have normal motility. A finding of good
motility therefore does not necessarily indicate good quality semen.
The causes of reduced per cent motile spermatozoa include sexual
immaturity of the dog, collection of a sample after sexual rest, testicular
degeneration, tail or mid-piece defects, immotile cilia syndrome, epididy-
mitis, prostatitis, contamination of the sample with urine, cold shock, or
chemicals.
Poor Per Cent Alive. The causes of reduced per cent alive spermatozoa
include sexual immaturity of the dog, first ejaculate after sexual rest,
testicular degeneration, epididymitis, prostatitis, contamination of the sam-
ple with urine, cold shock, or chemicals.
Head Abnormalities. Head and acrosome abnormalities originate in
the testicle and can reflect testicular degeneration, testicular hypoplasia,
knobbed sperm defect, and sexual immaturity.
Head-to-head agglutination or clumping of spermatozoa suggests the
presence of anti-sperm antibodies. This may be observed in Brucella canis
infection. 18
Tail and Mid-piece Abnormalities. Tail and mid-piece abnormalities
may arise in the testicle, in the epididymis, or after ejaculation. Osmotic
disturbances in the semen can lead to water passing through the sperm
membrane, causing bent tails. 57 Protoplasmic droplets are normal in sper-
matozoa leaving the testicle, and are shed during epididymal passage.
Persistence of droplets reflects either testicular or epididymal malfunction.
The causes of increased per cent of spermatozoa with proximal or distal
droplets include sexual immaturity, old age, testicular degeneration, dis-
turbances of epididymal function , and specific sperm defects (see Table 2).
The causes of increased per cent of tail-less heads include testicular
degeneration, inflammation of the duct system, and high temperatures
affecting spermatozoa in the epididymis.
Increased per cent of spermatozoa with singly bent tails can be caused
by testicular degeneration, epididymal disorders, cold shock, and water
contamination of the semen.
Increased per cent of spermatozoa with doubly bent tails can be caused
by a testicular disorder or an epididymal disorder.
Other Cells. The presence of red blood cells, neutrophil leukocytes,
macrophages, and rounded epithelial cells can reflect inflammatory process
in the tract-for example, epididymitis and prostatitis. Blood in the semen
also may be caused by trauma during semen collection.

ORCHITIS AND EPIDIDYMITIS

Clinical abnormalities usually are localized to the testis and/or epidid-

ymis , and include heat, pain, and swelling. Cytology of the sperm-rich
fraction of an ejaculate sample may confirm the diagnosis. Semen normally
contains very few inflammatory cells; in orchitis and epididymitis, such
cells usually are present in increased numbers, mainly as neutrophils.
Bacteria may be observed and have greatest significance when present in
large numbers or when phagocytosed by inflammatory cells. If low numbers
858 PATRICK J. WRIGHT AND BRUCE W . PARRY

of nonphagocytosed bacteria are noted, they may be the result of contami-

nation by the normal distal urethral and preputial microflora. Such micro-
organisms include Acinetobacter, Corynebacterium, Flavobacterium, He-
mophilus, Klebsiella, Moraxella, Staphylococcus, Streptococcus, and
Escherichia coli.6
If an ejaculate sample cannot be obtained, a fine-needle aspiration
biopsy could be taken from the affected area. In a normal dog, such a
specimen should be sterile and without inflammatory cells.
Microbiology is indicated whenever orchitis and/or epididymitis are
suspected. B. canis should be considered as the possible causative organism
in cases with epididymitis, and appropriate diagnostic steps should be taken
(see later discussion).

TESTICULAR TUMORS

The three most common testicular neoplasms are interstitial cell

(Leydig) tumors, Sertoli cell tumors , and seminomas.49 There is little
veterinary literature on the cytological appearance of these tumors. Because
the recommended treatment for both descended and undescended testicular
tumors is castration, definitive diagnosis of the neoplasm usually is by
histopathology. Fine-needle aspiration biopsy, however, could be useful to
distinguish testicular enlargement caused by inflammation from that caused
by neoplasia.

EVALUATION OF PROSTATIC FLUID

Clinical signs associated with prostatic disease frequently are related

to prostatic enlargement and include tenesmus, constipation, stranguria, or
dysuria. In cases of prostatitis, dogs may be depressed, anorectic, and
pyrexic.
Sample Collection
Cytology of prostatic fluid is a useful diagnostic tool in most cases with
an enlarged prostate. Specimens can be obtained by ejaculation or by
prostatic massage/wash. Fine-needle aspiration biopsy may be worthwhile
in some cases.
Ejaculate Sample. Methods for collection of ejaculate samples of semen
are outlined earlier in this article. Prostatic fluid comprises the last fraction
of the ejaculate and should be collected separate from the preceding sperm-
rich fraction.
Prostatic Massage/Wash. A satisfactory method is as follows . The dog
first is encouraged to urinate to empty its bladder, then a urinary catheter
is passed aseptically along the urethra into the bladder. The bladder is
lavaged several times with sterile physiological saline and the last 10 to 20
ml of lavage fluid is kept as a premassage specimen. While the prostate is
palpated via the rectum , the catheter is withdrawn until its tip is distal to
the gland. The prostate then is massaged, via the rectum and/or across the
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 859

abdominal wall, for 1 to 2 minutes. Then, while the urethral orifice is

gently occluded around the catheter, about 20 to 30 ml of sterile physiolog-
ical saline is infused through the catheter. Fluid is then aspirated, especially
from the area of the prostate, while slowly sliding the catheter past the
prostate and into the bladder. Most of the fluid recovered actually will
come from within the bladder. Cytology is then performed on pre- and
postmassage washes. Sediment smears usually are easier to read than direct
smears, except in some cases of prostatitis.
Fine-needle aspiration biopsy is not recommended for collecting sam-
ples from dogs with suspected prostatitis or prostatic abscessation because
of the risk of secondary peritonitis. Prostatic massage probably is the
method of choice. Ejaculate samples may be hard to collect in these cases
because prostatic pain may inhibit normal sexual behavior.
Fine-needle Aspiration Biopsy. The biopsy technique is similar to that
described elsewhere (see "Lymph Node Cytology" by Mills or "Cytology
of Cutaneous Lesions" by Cowell and Tyler in the July issue), except that
a 21-gauge, 4 to 6 em spinal needle (with stylet) may be the preferred
biopsy needle. 6 Perirectal and transabdominal approaches have been de-
scribed, depending on the prostate's location in the abdomen. 6 Ultrasono-
graphic guidance of the biopsy needle is useful but not essential. Palpation
of the gland via the rectum usually is sufficient for accurate needle insertion
and aspiration. This technique is valuable for diagnosis of prostatic tumors,
but is not recommended when prostatitis or prostatic cysts are suspected.
Cytology
Disorders of the prostate may be categorized into four main groups,
as follows. Cytology of premassage samples usually is fairly unremarkable
in benign prostatic hyperplasia, prostatic neoplasia, and prostatic cysts. Red
blood cells are not uncommon in cytologies of postmassage specimens,
because of pre-existing hemorrhage or following minor trauma at collection.
Benign Prostatic Hyperplasia. In this condition, the prostatic epithelial
cells are normal morphologically (Fig. I). They therefore are quite uniform
in size and appearance, with a moderate nucleus:cytoplasm ratio and a
round to slightly oval nucleus, which has a "roughened" chromatin pattern
and, frequently, a single small round nucleolus. Their cytoplasm is fairly
basophilic and, usually, somewhat vacuolated. There are very few, if any,
inflammatory cells present in cases of prostatic hyperplasia.
Prostatic Neoplasia. In adenocarcinoma of the prostate (Fig. 2), the
prostatic epithelial cells usually exhibit mild to marked anisokaryosis and
anisocytosis. There also is mild to marked variation in the nucleus:cytoplasm
ratio, which often is very high. Cells have a round to oval nucleus, usually
with multiple nucleoli of varying size and shape. Their cytoplasm frequently
is very basophilic and vacuolated. In some cases, the nucleus may be
distorted because cells are compressing each other in the tumor mass.
Normal and, possibly, abnormal mitotic activity also may be noted. Con-
current mild inflammation and chronic hemorrhage, evidenced by erythro-
phagocytic and/or hemosiderin-laden macrophages, may be seen.
Prostatitis. Acute and chronic prostatitis and prostatic abscessation are
similar cytologically. The premassage sample may reveal an inflammatory
860 PATRICK J. WRIGHT AND BRUCE W. PARRY

Figure 1. Cluster of morphologically normal prostatic epithelial cells. Note uniform

appearance of cells and moderate nucleus:cytoplasm ratio. (May-Grunwald-Giemsa stain;
X400)

reaction because cystitis may accompany prostatitis. The postmassage fluid

usually reveals a much greater inflammatory reaction.
Inflammatory cells mainly are neutrophils with good to markedly
karyolytic cell morphology (Fig. 3). Both free and/or phagocytosed bacteria
often are observed and may be in large numbers. Leukophagocytosis,
erythrophagocytosis, and hemosiderin-laden macrophages also may be
noted. Prostatic epithelial cells usually have normal morphology; in chronic
prostatitis and prostatic abscessation, however, they may exhibit squamous
metaplasia.
Prostatic Cyst. In this condition, cytology preparations frequently are
quite unremarkable. Prostatic epithelial cells may have normal morphology
or may exhibit squamous metaplasia. There are few inflammatory cells
present, thus distinguishing a cyst from an abscess.
Ancillary Tests. Other clinical pathology tests can assist with the
evaluation of these cases. An inflammatory leukogram frequently is present
in acute prostatitis and prostatic abscessation as well as in chronic prostatitis.
An inflammatory leukogram may be present in prostatic adenocarcinomas,
especially when the tumor has areas of necrosis and/or secondary infection.
A concurrent hyperproteinemia (hyperglobulinemia) and a mild nonregen-
erative anemia of chronic (inflammatory) disease also may be noted in any
of the latter conditions. In contrast, hematology usually is within reference
limits in cases of benign hyperplasia and those with prostatic cysts.
Urinalysis may be useful in the initial evaluation of prostatic diseases.
In cases of prostatitis and prostatic abscessation, hematuria, pyuria, pro-
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 861

Figure 2. Cluster of prostatic cells from a dog with an adenocarcinoma. Note moderate
to marked variation in nuclear and cellular size and variable (often high) nucleus:cytoplasm
ratio. (May-Grunwald-Giemsa stain; X 400)

Figure 3. Septic prostatitis. Inflammatory cells exhibit mild to marked karolysis. Massive
numbers of bacteria are also present. (May-Grunwald-Giemsa; X 400)
862 PATRICK J. WRIGHT AND BRUCE W. PARRY

teinuria, and bacteriuria often are present. Hematuria and proteinuria also
may occur with prostatic neoplasia. Collection of pre- and postmassage
samples will help decide whether these findings are associated with con-
current disease of the upper urinary tract.

CANINE VAGINAL CYTOLOGY

The evaluation of vaginal cytology in the bitch is a well-studied, 19• 45-48

well-reviewed, 5 • 12• 35 • 36 • 36 • 44 and well-used36• 48 aid for the management of
canine reproduction. Changes in the vaginal epithelium, stimulated by
reproductive hormones-particularly estrogens, are reflected in the cells
found in vaginal smears. Examination of vaginal smears therefore can assist
in determination of the stage of the cycle and in the diagnosis of abnormal-
ities of the estrous cycle. Vaginal cytology also can assist in the diagnosis
of reproductive disorders by the detection of cells reflecting pathological
processes-leukocytes, red blood cells, and tumor cells, for example.
Sample Collection
The method of sample collection should be quick, easy, of minimal
discomfort to the bitch, and give a good preparation of vaginal cells for
examination. Samples should be taken from the caudal vagina. Samples
from the clitoral fossa or vestibule areas are not satisfactory because these
areas contain keratinized cells and smears from them do not represent
hormone-related stages of the cycle clearly. 43 Impression smears obtained
by everting the lips of the vulva therefore are not satisfactory.
Smears can be collected using a spatula, glass rod, or cotton-wool
swab-stick. The cotton-wool swab-stick may be moistened with saline and
can be inserted unguarded or using a speculum. The senior author uses a
sterile glass tube (10 em X 9 mm) to avoid contamination with cells from
the vestibule. Samples also can be collected by inserting a pipette containing
a small volume of saline and taking up cells by expulsion then aspiration of
the fluid .
Introduction of a sampling device into the vagina involves parting the
lips of the vulva and inserting the device, initially in an upward direction
at 45 degrees to avoid the clitoral fossa and urethral opening, then
horizontally deeper into the vagina. The device then should be rotated to
wipe the vaginal mucosa to pick up cells from the vaginal wall. The device
then is removed from the bitch and gently rolled over a glass slide so the
cells fold off in an interpretable format, not as an uninterpretable porridge.
Staining Methods
Staining methods are of two main types, each with different advantages
and disadvantages. One group of methods identifies keratin precursors; the
other group does not. The early staining procedures that identified keratin
precursors in canine vaginal smears were a modification of Schorr's tri-
chrome stain. 45 Other workers have simplified this procedure further, by
reducing the number of steps and the time taken. 33• 44 • 53 Keratin precursors
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 863
are contained in the cells characteristic of estrus. Cells containing keratin
precursors stain orange-red, permitting easy interpretation of the smear.
The methods that stain these cells differentially are more time-
consuming and require more care with fixation of cells and with the quality
of the staining preparations than other methods. Reliance on color of cells
for the interpretation of smears may lead to errors if the smear does not
stain properly, as can happen if the cells are not fixed when wet. 33 In this
instance, all cells stain orange-red. It therefore is important, if using these
techniques, to be careful with preparation methods and stain quality and
to assess morphology as well as color.
Using staining methods that do not stain keratin precursors differen-
tially requires the cells to be identified by morphological features alone.
These staining procedures generally are simpler, quicker, and more robust.
Such stains include Wright's, Giemsa, 5 • 35• 38 and new methylene blue. 35
The Cells
Estrogen acts on the vaginal epithelium, causing proliferation, differ-
entiation, and exfoliation of cells. 45 Differentiation involves an increase in
size and change in shape of the cells, the appearance of keratin precursors
in the cells, and degeneration (pyknosis) of the nucleus. The degree of
change in the vaginal epithelium and the vaginal smear reflects the amount
of circulating estrogens produced by developing ovarian follicles , mainly
during proestrus. The changes are maximal at estrus.
Although the changes induced in vaginal cells are continuous, the cells
on a smear can be classified into five main types. These classifications are
on the basis of cell size and shape, and nuclear morphology. The classes of
cells can be identified irrespective of the staining procedure used. Using
Romanovsky-type stains, the cell's nucleus is deeply basophilic. The cyto-
plasm is more palely basophilic, with the degree of basophilia being greatest
in the less mature (less cornified) cells. Using Shorr's trichrome staining
methods, the cell nucleus is red. The cytoplasm is green when keratin
precursors are absent and becomes progressively more orange-red as keratin
precursors increase in conce ntration.
The categories of cells described in the following sections are recog-
nizable. The cytoplasmic staining characteristics refer to smears prepared
with modified Shorr's trichrome stain. 45 Bacteria may be seen at any stage
of the cycle. Some epithelial cells from darkly pigmented bitches may
contain melanin. 35
Basal Cells. These rarely are seen in smears. They are small epithelial
cells attached to the basement membrane of the vagina.
Parabasal Cells. These are round or oval cells with a vesicular nucleus.
They have the highest nucleus:cytoplasm ratio of exfoliated cells. Their
cytoplasm stains green and nuclei, dark red. Occasional parabasal cells
contain cytoplasmic vacuoles and then are called "foam cells." Some
parabasal cells may contain leukocytes and then are called "metestrum
cells."
SmaU Intermediate Cells. These are larger and more variable in size
than parabasal cells. They have vesicular nuclei and cell borders that usually
are rounded. Their cytoplasm stains green.
864 PATRICK J. WRIGHT AND BRUCE W . PARRY

Superficial Intermediate Cells. These are larger than the small inter-
mediate cells, but with a similar nucleus. They have a polygonal shape,
with angulated and folded cell borders. Their cytoplasm may stain wholly
or partially orange-red, indicating the presence of keratin precursors.
Superficial Cells. These are the largest epithelial cells in the smear.
They are polygonal, with angulated and folded borders. The nucleus is
pyknotic or absent. Their cytoplasm stains orange-red. Some superficial
cells may contain cytoplasmic granules, 5 • 35 the significance of which is
unclear.
Indices
Various indices have been derived to describe the cell pictures seen
in vaginal smears. These include the eosinophilic index (EI), superficial cell
index (SCI), and kariopyknotic index (KPI). 38• 47 The EI and SCI are the
most useful.
The EI relates to smears stained differentially for keratin. The EI can
be expressed as the proportion (per cent) of epithelial cells that are
keratinized (orange-red cytoplasm). The EI increases from near zero at the
start of proestrus to approach 100 per cent at estrus. During diestrus and
anestrus, the El is low (less than 10 per cent). On the first day of diestrus,
the EI declines markedly (by at least 10 to 20 per cent), with the sudden
appearance of small intermediate and parabasal cells. 29
The SCI is the number of cells from the superficial layers (superficial
and superficial intermediate cells) as a percentage of the total number of
epithelial cells. The SCI parallels the EI and increases from near zero at
the start of proestrus to approach 100 per cent at estrus and remains at this
level until the onset of diestrus.
The KPI is the number of anuclear cells and cells with pyknotic nuclei
as a percentage of the total number of superficial cells. The KPI is high
during estrus. Peak levels and their timing vary widely among bitches,
however.
Changes in Vaginal Cytology During the Estrous Cycle
Proestrus. Early in proestrus, the main cells present are parabasal
cells and small intermediate cells. As proestrus progresses, the proportion
of these cells declines and the numbers oflarge intermediate and superficial
cells increases . The EI and SCI therefore increase progressively during
proestrus. Red blood cells usually are present and commonly decline toward
the onset of estrus. Some bitches have little proestrous bleeding ("colorless
heats") and, therefore, fewer red blood cells on the smear. Some neutrophils
may be present early in proestrus.
Estrus. In estrus, superficial cells predominate (Fig. 4). The proportion
that are anuclear varies widely among bitches. The EI and SCI usually are
over 90 per cent and commonly reach 100 per cent. The smear commonly
has a "clean look," uncluttered by debris. The number of red blood cells
varies; in smears from some bitches, only few may be seen, but in other
bitches, there may be a considerable bloody discharge throughout estrus
and many red blood cells on the smear. On the last day of estrus, the cells
may appear in clumps or in sheets, and cell outlines are less sharp. 29
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 865

Figure 4. Superficial cells in a vaginal smear from a bitch in estrus. (Schorr's trichrome
stain; x 400)

Diestrus. The onset of diestrus can be defined on the basis of behavior

(day of first refusal to accept mating) or on the basis of changes in the
vaginal smear. The onset of diestrus determined on the basis of behavior
usually follows the onset determined cytologically. Cytologically, the onset
of diestrus (day 1 = Dl) is characterized by the sudden appearance of
significant numbers of cells from deeper layers of the vaginal mucosa,
namely intermediate and parabasal cells (Fig. 5). 29 The EI and SCI fall
abruptly. A marked increase in the number of leukocytes commonly occurs
around this time. It may not be possible to distinguish between the first
few days of diestrus and proestrus on the basis of a smear alone.
Subsequent vaginal cytological changes during diestrus and anestrus
are less dramatic, and may reflect circulating plasma progesterone concen-
trations. 14 Plasma progesterone concentrations are higher during the first
part of diestrus than during the second part and are undetectable during
anestrus. Vaginal smears in early diestrus (to D30) are characterized mostly
by parabasal and small intermediate cells, with a few large intermediate
and superficial cells. Many white blood cells usually are present, with
greatest numbers over a few days just at the end of estrus. Vaginal smears
late in diestrus are characterized by cuboidal and columnar cells, and by
cytolysis, identified by the presence of nuclear shadows with no cytoplasm.
Few white blood cells typically are present. 14
Anestrus. In anestrus, the number of cells from deeper layers of the
vaginal mucosa predominate, namely parabasal and some cuboidal and
columnar cells. White blood cells are quite uncommon. 14
866 PATRICK J. WRIGHT AND BRUCE W. PARRY

Figure 5. Predominantly parabasal and intermediate cells in a vaginal smear from a bitch
at day 2 of diestrus. (Schorr's trichrome stain; X 400)

Use of Vaginal Smears

Vaginal smears can be used to assist in the identification of the stage
of the estrous cycle and the time for mating, to estimate the time of
ovulation retrospectively, and to estimate the time of whelping prospec-
tively. Vaginal smears do not permit the accurate timing or prediction of
the plasma luteinizing hormone surge or of ovulation. 12• 31 Vaginal smears
also can be a guide to appropriate treatments and can assist in the diagnosis
of reproductive disorders.
Time of Ovulation and of Whelping. Vaginal smears can indicate
whether the bitch is in proestrus or estrus. From a proestrus smear, it is
not possible to predict accurately when the bitch will be in estrus. Serial
smears each 2 to 3 days are required to detect estrus. Once in estrus, it is
not possible to predict the time of ovulation. It is, however, possible to
determine when estrus has finished, by the characteristic changes described
for Dl of diestrus29 (see earlier discussion). Studies that related whelping
dates to D 1 indicated that ovulation occurred around 6 days before the
onset of diestrus (D - 6), and that fertilization occurred 3 days before the
onset of diestrus (D- 3). The day of whelping was around 057. The
estimation of the time of whelping was much more accurate when based
on the onset of diestrus than when it was based on the time of mating.
Because no direct measure me nts were made relating the occurrence of
ovulation to the onset of diestrus, it is unclear whether the variation in the
time of whelping, whe n related to Dl, reflects variation in the duration of
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 867

gestation or variation in the time relationship between D 1 and ovulation or

fertilization. Day 1 of diestrus (and expected whelping date) can be
determined from a series of smears made by the owner and presented for
staining and examination when the bitch clearly has finished her season.
Time for Mating. Vaginal cytology can indicate the period of estrus
over which mating should occur. The number of matings necessary for
fertility can be based on estimates of the longevity of spermatozoa in the
uterus (4 to 6 days), 13 experience with experimental dogs (one mating is
sufficient, provided it is not very early or very late in estrus), 29 or concern
that semen quality or conditions in the female tract may not be optimal for
fertility and a wish to optimize chances of pregnancy (each 2 to 3 days
during estrus until 01 of diestrus).
Indications for Treatment for Misalliance. Treatment for misalliance
with estradiol should be avoided if possible, because of the potential side
effects, including bone marrow suppression, resulting in thrombocytopenia;
prolongation of estrus; and cystic endometrial hyperplasia. In a bitch
suspected of having had a recent inappropriate mating, a vaginal smear
indicating proestrus or diestrus would suggest that any mating would not
be fertile and that there was no need for treatment. Vaginal cytology also
can assist in the determination of whether a mating has occurred. The
presence of spermatozoa is diagnostic of mating. The absence of spermatozoa
does not necessarily indicate that mating has not occurred. 36
Guide to Appropriate Progestagen Treatment. Long-term progestagen
treatment, if considered necessary, should commence during anestrus, to
lessen the risk of development of cystic endometrial hyperplasia. Unfortu-
nately, vaginal cytology will not distinguish between diestrus and anestrus.
The latter requires measurement of plasma progesterone concentration.
The progestagen dose rate to postpone estrus (used in a bitch during
anestrus) is lower than the dose rate to suppress estrus (used in a bitch at
the start of proestrus). Vaginal cytology can distinguish between anestrus
and the start of proestrus, indicating the appropriate dose regimen of
progestagen to be used.
Diagnosis of Disorders of the Estrous Cycle. Vaginal smears, by
reflecting underlying endocrine and physiological phenomena, can assist in
the diagnosis of endocrine disorders of the cycle. 41 • 48 Short proestrus, not
followed by estrus, can occur in pubertal bitches commencing cycling, and
in bitches with endoctine disorders such as hypothyroidism. Prolonged
proestrus, not followed by estrus, can be caused by cystic follicles, granulosa
cell tumor, or administered estrogens. Prolonged estrus can be caused by
failure of ovulation, cystic follicles, ovarian tumor, or estrogen treatment.
Vaginal cytology also can assist in the diagnosis of behavioral abnor-
malities. Bitches with vaginitis or endometritis may be attractive to dogs
and may have a red vulval discharge. Examination of the vaginal smear will
distinguish between such bitches and estrous bitches. Examination of
smears also will determine whether failure to mate is caused by the bitch
not being in estrus or some other cause-for example, psychological,
anatomical, sociological.
Assessment of Purulent Vulval Discharges. Acute metritis, open pyo-
metra, and vaginitis all can produce a purulent vulval discharge. The
868 PATRICK J. WRIGHT AND BRUCE W. PARRY

differential diagnosis of these disorders is based mainly on history, clinical

signs, and clinical examination. Radiology and hematology also may be
helpful, particularly in the diagnosis of pyometra. An enlarged uterus may
be visualized on radiography. Hematology usually reveals a relative poly-
cythemia, caused by dehydration; an inflammatory leukogram; and hyper-
proteinemia, associated with dehydration and inflammation. These hema-
tologic findings also are often observed in acute metritis.
Cytology of purulent vulval discharges will reveal inflammatory cells,
predominantly neutrophils, in numbers much greater than expected at the
stage of the estrous cycle indicated by concurrent observation of epithelial
cell types on the same smear. Cell morphology can be quite variable, from
normal to marked karyolytic degeneration. In cases of acute metritis and
pyometra, endometrial cells may be observed and the smears often have a
mucoid background, with moderate numbers of red blood cells also present.
Bacteria frequently are observed in the smears from these cases. It
must be remembered, however, that the vagina normally has a resident
microflora, so bacteria usually are observed in smears from bitches during
anestrus, proestrus, estrus, and diestrus. Bacteria are considered to be
important cytologically (and clinically) when they are observed in conjunc-
tion with an inflammatory reaction and (especially) when they are seen to
have been phagocytosed by neutrophils and/or macrophages.
Microbiological culture and sensitivity tests are worthwhile in many of
these cases. Results must be interpreted with the aforementioned caveat.
The following (incomplete list of) organisms have been isolated from the
normal canine vagina: species of Bacillus, Corynebacterium, Moraxella,
Neisseria, Pasteurella, Proteus, Staphylococcus, Streptococcus, and E.
coli. 34 Several of these organisms also have been isolated as potential
pathogens in cases of acute metritis, pyometra, and vaginitis.
B. canis causes abortions, typically within the last 2 weeks of gestation.
Affected bitches subsequently have a brown to greenish-grey vulval dis-
charge for 1 to 6 weeks. It is discussed more fully below.
Assessment of Bloody Vulval Discharges. Bloody discharges from the
vulva are normal during proestrus and sometimes estrus, as outlined earlier,
and can occur for up to 6 weeks after whelping. If the amount of blood loss
is considered to be excessive, a coagulopathy may be present. Evaluation
of the latter is discussed in the July issue of this clinic (see "Laboratory
Evaluation of Hemorrhagic Coagulopathies in Small Animal Practice" by
Parry). Bleeding from the vulva also may occur following trauma to the
reproductive tract; from friable vaginal tumor masses, especially polyps and
transmissible venereal tumors; in conjunction with the purulent discharge
of acute metritis and pyometra (already discussed), and in a condition
referred to as subinvolution of placental sites.
The normal uterus involutes within 6 weeks of parturition. Subinvo-
lution of placental sites is an uncommon disorder that usually results from
delayed (or, rarely, failed) involution of the sites of placentation.54 The
discharge usually is serosanguineous in nature, but may vary from essentially
serous to frankly bloody. Its volume may vary from scanty to copious. In
rare cases, usually those with a complicated whelping, uterine perforation
and peritonitis also may be present. Cytology of the discharge reveals
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 869
mainly erythrocytes and possibly some vacuolated syncytial trophoblast-like
cells. The discharge is not purulent unless a secondary metritis has
developed. Parabasal epithelial cells and neutrophils may be present if the
vaginal wall was swabbed during specimen collection.
If significant blood loss has occurred, the bitch may have a regenerative
anemia. Persistence of such blood loss could result in iron-deficiency
anemia.
From a clinical viewpoint, the discharge associated with subinvolution
of placental sites is not to be confused with lochia. Lochia is the discharge
that is present while placental sites are separating. It occurs at the time of
parturition (or abortion) and for up to 3 weeks thereafter. 36 Cytologically,
the discharges in these two situations may be quite similar. In lochia,
however, cells containing green "bile pigment" called uteroverdin may be
observed readily. 36 Large amounts of this pigment will result in the lochia
being grossly green in color. The cytological similarity of these two situations
is not surprising, since subinvolution of placental sites is more or less a
pathological extension of the physiological process that produces lochia.
Neoplasia. Tumors of the uterus are uncommon in the bitch55 and
usually would be diagnosed by histopathology. Leiomyomas are encoun-
tered most frequently . Tumors involving the vagina or vulva are more
common than those of the uterus. Fibromas, fibropapillomas (polyps),
leiomyomas, leiomyosarcomas, metastatic liposarcoma, metastatic lympho-
sarcomas, metastatic mammary adenocarcinomas, squamous cell carcino-
mas, transmissible venereal tumor, and transitional cell carcinomas of the
vagina or vulva have been documented. 36• 43 • 55 Cytology may assist in the
diagnosis of these neoplasms. Methods of specimen collection include
vaginal swabs and fine-needle aspiration biopsy. The cytological character-
istics of several of these types of tumors are described elsewhere (see
"Cytology of Cutaneous Lesions" by Cowell and Tyler in the July issue).
Transmissible venereal tumor is discussed later in this article.

SPECIFIC DISEASES

Brucella canis Infection

The following discussion is based on a recent review of canine brucel-
losis by Green and George. 21 Dogs and bitches infected by B. canis seldom
are very ill. Affected dogs usually are presented because of testicular
enlargement and scrotal dermatitis. The latter results from persistent licking
of the scrotum, with secondary bacterial infection (not B. canis) and scrotal
thickening. The testicular enlargement results from an epididymitis, mainly
involving the epididymal tail, caused by B. canis. Inflammation exposes
(foreign) spermatozoal antigens to the host immune system, with production
of anti-sperm antibodies. Infertility and interference with spermatogenesis
result. Actual orchitis rarely occurs.
Affected bitches usually are presented following the abortion of dead
fetuses . Embryonic and fetal death can occur at any stage, but often is at
45 to 59 days of gestation. Any puppies born alive usually die as neonates.
870 PATRICK J. WRIGHT AND BRUCE W. PARRY

Infection with B. canis does not affect the estral cycle, but bitches may be
presented because of infertility related to early embryonic death.
Bacteremia and/or immune complexes may result in other disorders,
including anterior uveitis, discospondylitis, meningoencephalitis, protein-
losing glomerulopathy, and prostatitis.
B. canis is an obligate intracellular gram-negative coccobacillus. It is
shed in seminal and prostatic fluids, urine, and vaginal discharges following
abortion. The organism can invade any mucosal surface, but especially
those of the oral cavity, conjunctiva, and vagina. Following a bacteremia,
it can localize in various tissues of the reticuloendothelial system, including
lymph nodes, spleen, liver, and bone marrow, and in the epididymis,
prostate, or uterus. The organism cannot multiply in the non-pregnant
uterus. In the pregnant bitch, bacterial proliferation occurs and in utero
transmission will cause embryonic, fetal, and/or neonatal death.
A complete blood count may reveal an inflammatory leukogram and
hyperproteinemia (hyperglobulinemia), possibly with a mild nonregenera-
tive anemia. Evaluation of semen from affected males shows a progression
of spermatozoal abnormalities. Beginning about 5 weeks after infection,
they include spermatozoal immaturity, acrosomal deformation, swelling of
mid-pieces, and retention of protoplasmic droplets. By 15 weeks after
infection, they include tail-less heads, bent tails, and head-to-head agglu-
tination, caused by anti-sperm antibodies. The latter could occur in any
disease that disrupts the blood/tubule barrier, but is quite common on B.
canis epididymitis. Inflammatory cells, principally neutrophils, are numer-
ous at this stage and spermatophagic macrophages are present. By 20 weeks
after infection, less than 10 per cent of spermatozoa are normal morpholog-
ically. Aspermia eventually occurs and the inflammatory reaction subsides.
Microbiology is indicated in suspected cases. Suitable specimens
include semen and urine of dogs and vaginal discharges, uterus, placenta,
and milk of affected bitches. Culture of blood, lymph nodes, and bone
marrow also can be attempted. Diagnosis of B. canis infection, however,
usually is by serology. Various tests are available and have been reviewed. 21
Antibodies to B. canis cross-react with B. ovis. This forms the basis of a
rapid slide agglutination test that often is used for diagnosis of B. canis
(Canine brucellosis diagnostic kit, Pitman-Moore, Mundelein, IL 60060).
Transmissible Venereal Tumor
The following presentation is based on the review by Calvert. 9 Trans-
missible venereal tumor (TVT) is a contagious allograft that is transmitted
(transplanted) by licking, biting, and coitus. It can affect bitches and dogs.
The tumor cells are immunogenic. Spontaneous regression, without recur-
rence, occurs in about 9 of 10 cases, usually within 6 months. The remainder
of cases tend to have a protracted clinical course. Local tissue invasion is
common in the latter cases, but the tumor is metastatic only occasionally,
and, then, generally to regional lymph nodes.
Tumors can graft onto genital and extragenital sites. Genital TVT
usually occur in the vestibule of the vagina in the bitch and anywhere on
the mucosa of the penis or prepuce in the dog. At these sites, TVT
frequently are not visible externally. The tumors start as hyperemic papules
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 871
and grow to cauliflower-like masses of variable size, which are very friable
and bleed easily. Extragenital TVT can arise on the skin of the head, neck,
limbs, or trunk and on the mucosa of the oral and nasal cavities. At these
sites, TVT tend to be less pedunculated than those of the genital tract.
Impression smears or fine-needle aspiration biopsies usually are diag-
nostic. The cytology is that of an undifferentiated round cell tumor. Cells
tend to be present in clusters, with little connective tissue stroma. Individ-
ual cells are large and discrete, with mild to moderate anisokaryosis and
anisocytosis. They have a moderate nucleus:cytoplasm ratio. The nucleus
usually is round to slightly oval, with a finely granular chromatin pattern
and a large, prominent, round nucleolus of fairly uniform size. Mitotic
activity often is apparent readily. The cytoplasm frequently is vacuolated
and somewhat basophilic (Fig. 6).

SUMMARY

The methods for semen collection, its laboratory examination, and the
interpretation of findings are presented in this article. The lack of compre-

Figure 6. Transmissible venereal tumor cells. Note characteristic "round cell" appearance.
Nucleus has a coarse chromatin pattern with a prominent nucleolus. Cytoplasm is highly
vacuolated. (Wright's stain; X 1000) (From Cowell RL, Tyler RD: Cytology of cutaneous
lesions. Vet Clin North Am [Small Anim Pract]l9:789, 1989; with permission.)
872 PATRICK J. WRIGHT AND BRUCE W. PARRY

hensive data for normal dogs and the lack of data associating actual
percentages of spermatozoa with specific abnormalities with fertility or
infertility are highlighted. Consequently, there is a need for standardization
and completeness of semen examination procedures, especially in studies
destined for publication. Collection and analysis of prostatic samples then
is discussed, and the distinguishing cytological features of benign prostatic
hyperplasia, prostatic adenocarcinoma, prostatitis (including prostatic ab-
scessation), and prostatic cysts are presented.
This is followed by an assessment of the clinical usefulness of vaginal
cytology, particularly to assist in the management of normal canine repro-
duction and in the diagnosis of reproductive disorders. The ways in which
vaginal smears can facilitate the diagnosis of the stage of the estrous cycle
and the diagnosis of abnormalities of the cycle and other disorders of
reproduction are presented. Further consideration is given to its use to
estimate the time of ovulation retrospectively and estimate the time of
whelping prospectively.
Finally, two specific diseases that can affect dogs and bitches are
reviewed, namely, canine brucellosis and transmissible venereal tumor.

REFERENCES

l. Afzelius BA, Carlsten J, Karlsson S: Clinical, pathological, and ultrastructural features of

situs inversus and immotile cilia syndrome in a dog. J Am Vet Med Assoc 184:560,
1984
2. Allen WE: Macrophage cells in the semen of a dog with oligospermia. Vet Rec 109:310,
1981
3. Amann RP: Reproductive physiology and endocrinology of the dog. In Morrow DA (ed):
Current Therapy in Theriogenology. 2. Philadelphia, WB Saunders, 1986, p 532
4. Aughey E, Renton JP: The ultrastructure of abnormal spermatozoa in a stud dog. J Reprod
Fertil 25:303, 1971
5. Barrett RP: Exfoliative vaginal cytology of the dog using Wright's stain. Vet Med Small
Anim Clin 71:1236, 1976
6. Barsanti JA: Genitourinary tract infections. In Greene CE (ed): Clinical Microbiology and
Infectious Diseases of the Dog and Cat. Philadelphia, WB Saunders, 1984, p 269
7. Bongso TA: Comparative silver staining patterns of water buffalo, goat, and pig sperma-
tozoa. Arch Androlll:13, 1983
8. Boucher JH, Foote RH, Kirk RW: The evaluation of semen quality in the dog and the
effects of frequency of ejaculation upon semen quality, libido, and depletion of sperm
reserves. Cornell Vet 48:67, 1958
9. Calvert CA: Canine viral and transmissible neoplasias. In Greene CE (ed): Clinical
Microbiology and Infectious Diseases of the Dog and Cat. Philadelphia, WB Saunders,
1984, p 461
10. Christensen GC, Dougherty TW: A simplified apparatus for obtaining semen from dogs
by electrical stimulation. J Am Vet Med Assoc 127:50, 1955
11. Clough E, Pyle RL, Hare WCD, et al: An XXY sex chromosome constitution in a dog
with testicular hypoplasia and congenital heart disease. Cytogenet Cell Genet 9:71,
1970
12. Concannon PW, DiGregorio G: Canine vaginal smears. In Burke TM (ed): Small Animal
Reproduction and Infertility. Philadelphia, Lea and Febiger, 1986, p 96
13. Doak RL, Hall A, Dale HE: Longevity of spermatozoa in the reproductive tract of the
bitch. J Reprod Fertil 13:51, 1967
14. Dore MA: The role of the vaginal smear in the detection of metoestrus and anoestrus in
the bitch. J Small Anim Pract 19:561, 1978
CYTOLOGY OF THE CANINE REPRODUCTIVE SYSTEM 873
15. Edwards DF, Patton CS, Bemis DA, et al: Immotile cilia syndrome in three dogs from a
litter. J Am Vet Med Assoc 183:667, 1983
16. Evans J, Renton JP: A case of azoospermia in a previously fertile dog with subsequent
recovery. Vet Rec 92:198, 1973
17. Galloway DB, Norman JR: Laboratory examination of semen as a diagnostic aid. Proc 9th
Int Cong Anim Reprod AI 4:714, 1980
18. George LW, Duncan JR, Carmichael LE: Semen examination in dogs with canine
brucellosis. Am J Vet Res 40:1589, 1979
19. Gier HT: Estrous cycle in the bitch. Vet Scope 5:2, 1960
20. Goodwin M, Gooding KM , Regnier F: Sex pheromone in the dog. Science 203:559, 1979
21. Greene CE, George LW: Canine brucellosis. In Greene CE (ed): Clinical Microbiology
and Infectious Diseases of the Dog and Cat. Philadelphia, WB Saunders, 1984, p 646
22. Hadley JC: Spermatogenic arrest with azoospermia in two Welsh springer spaniels. J
Small Anim Pract 13:135, 1972
23. Hancock JL, Rowlands IW: The physiology of reproduction in the dog. Vet Rec 61:771,
1949
24. Hancock JR: The morphology of boar spermatozoa. J R Micr Soc 76:84, 1957
25. Harrop AE: A new type of canine artificial vagina. Br Vet J 110:194, 1954
26. Harrop AE: Some observations on canine semen. Vet Rec 67:494, 1955
27. Harrop AE: An aspect of infertility in the dog. Vet Rec 72:362, 1960
28. Harrop AE: The infertile male dog. J Small Anim Pract 7:723, 1966
29. Holst, PA, Phemister RD: Onset of diestrus in the beagle bitch: Definition and signifi-
cance. Am J Vet Res 35:401, 1974
30. Larsen RE: Evaluation of fertility problems in the male dog. Vet Clin North Am 7:735,
1977
31. Linde C, Karlsson I: The correlation between the cytology of the vaginal smear and the
time of ovulation in the bitch. J Small Anim Pract 25:77, 1984
32. Oettle EE, Soley JT: Infertility in a Maltese poodle as a result of a sperm mid-piece
defect. J S Afr Vet Assoc 56:103, 1985
33. Oettle EE, Weldhagen A: A modified Schorr's stain: A practical rapid stain for canine
vaginal cytology. J S Afr Vet Assoc 53:267, 1982
34. Olson PNS, Mather EC: Canine vaginal and uterine bacterial flora. J Am Vet Med Assoc
172:708, 1976
35. Olson PN, Thrall MA, Wykes PM, et al: Vaginal cytology. Part 1. A useful tool for staging
the canine estrous cycle. Comp Cont Ed Pract Vet 6:288, 1984
36. Olson PN, Thrall MA, Wykes PM, et al: Vaginal cytology. Part 2. Its use in diagnosing
canine reproductive disorders. Comp Cont Ed Pract Vet 6:385, 1984
37. Plummer JM, Watson PF, Allen WE: A spermatozoal mid-piece abnormality associated
with infertility in a llasa apso dog. J Small Anim Pract 28:743, 1987
38. Post K: Canine vaginal cytology during the estrous cycle. Can Vet J 26:101, 1985
39. Renton JP, Aughey E: Certain interesting aspects of infertility in two Shetland collies.
Vet Rec 88:408, 1971 ·
40. Renton JP, Harvey MJA, Harker S: A spermatozoal abnormality in dogs related to
infertility. Vet Rec 118:429, 1986
41. Roberts SJ: Veterinary Obstetrics and Genital Diseases. Ann Arbor, Ml, Edwards
Brothers, 1971
42. Rosenthal R: infertility in the male dog. Comp Cont Ed Pract Vet 5:983, 1983
43. Roszel JF: Genital cytology of the bitch. Vet Scope 19:2, 1975
44. Roszel JF: Normal canine vaginal cytology. Vet Clin North Am 7:667, 1977
45. Schutte AP: Canine vaginal cytology. I. Technique and cytological morphology. J Small
Anim Pract 8:301, 1967
46. Schutte AP: Canine vaginal cytology. II. Cyclic changes. J Small Anim Pract 8:307, 1967
47. Schutte AP: Canine vaginal cytology. III. Compilation and evaluation of cellular indices.
J Small Anim Pract 8:313, 1967
48. Simmons J: The vaginal smear and its practical application. Vet Med Small Anim Clin
65:369, 1970
49. Susaneck SJ, Withrow SJ: Tumors of the canine male reproductive tract. In Morrow DA
(ed): Current Therapy in Theriogenology. 2. Philadelphia, WB Saunders, 1986, p 561
50. Taha MB, Noakes DE, Allen WE: Some aspects of reproductive function in the male
beagle at puberty. J Small Anim Pract 22:663, 1981
874 PATRICK J. WRIGHT AND BRUCE W. PARRY

51. Taha MB, Noakes DE, Allen WE : The effect of season of the year on the characteristics
and composition of dog semen. J Small Anim Pract 22:177, 1981
52. Taha MB, Noakes DE, Allen WE: The effect of the frequency of ejaculation on the
seminal characteristics and libido in the beagle dog. J Small Anim Pract 24:309, 1983
53. Thomas TN: Staining techniques for vaginal cytology evaluation. Proc Ann Meet Soc
Therio, 1987, p 262
54. Wheeler SL: Subinvolution of placental sites in the bitch. In Morrow DA (ed): Current
Therapy in Theriogenology. 2. Philadelphia, WB Saunders, 1986, p 513
55. Withrow SJ, Susaneck SJ: Tumors of the canine female reproductive tract. In Morrow
DA (ed): Current Therapy in Theriogenology. 2. Philadelphia, WB Saunders, 1986, p
521
56. Wong WT, Dhaliwal GK: Observations on semen quality of dogs in the tropics. Vet Rec
116:313, 1985
57. Wright PJ: Serum spermagglutinins and semen quality in the bull. Aust Vet J56:10, 1980

Department of Veterinary Clinical Sciences

School of Veterinary Science
University of Melbourne
Werribee, Victoria
Australia 3030