Theriogenology 93 (2017) 33e39

Contents lists available at ScienceDirect

Theriogenology
journal homepage: www.theriojournal.com

The association between the serum concentration of canine prostate

specific esterase (CPSE) and the size of the canine prostate
B.S. Holst a, *, E. Holmroos a, L. Friling b, S. Hanås a, c, L.-M. Langborg d, M.A. Franko e,
K. Hansson a
a
Department of Clinical Sciences, Swedish University of Agricultural Sciences (SLU), Uppsala, Sweden
b
Diagnostic Imaging Clinic, University Animal Hospital, SLU, Uppsala, Sweden
c €sterås, Sweden
Evidensia Djurkliniken Va
d
ReDog Veterinary Clinic, Va€sterås, Sweden
e
Department of Energy and Technology, Applied Statistics and Mathematics, SLU, Uppsala, Sweden

a r t i c l e i n f o a b s t r a c t

Article history: Benign prostatic hyperplasia (BPH) is an age-related disorder in the intact male dog that is associated
Received 27 October 2016 with an increase in the prostatic size. Ultrasonography gives a reliable estimate of the prostatic size, but a
Received in revised form method for screening the prostate size using a serum sample has advantages, such as requiring less
11 January 2017
expensive equipment. The primary aim of the study was to study the association between the concen-
Accepted 18 January 2017
tration of the circulating biomarker canine prostate specific esterase (CPSE) and prostatic size. Seventy-
Available online 25 January 2017
nine dogs that were four years old or older were included in the study. Ultrasonography was used for
calculating the volume of the prostate. The calculated volume was divided by an estimate of the normal
Keywords:
Dogs
prostatic volume in dogs aged one to four years, to determine the relative prostatic size: the size of the
Benign prostatic hyperplasia prostate in relation to the normal size in dogs 1e4 years old (Srel). CPSE was analyzed from serum
Andrology samples. Multiple linear regression analysis was used for studying associations between variables. Pre-
CPSE diction intervals for the relative prostatic size based on CPSE concentrations were calculated, as were
receiver operating curves for CPSE concentrations predicting Srel. The concentration of CPSE was asso-
ciated with the relative size and contour of the prostate (P < 0.001). All dogs with clinical signs of BPH
had an Srel
2.5. A CPSE concentration of 200 ng/mL predicted Srel to 2.5 (95% P.I: 1.2e4.8). Based on ROC
analysis, the optimal discrimination threshold for CPSE concentration for Srel
2.5 was estimated as
90 ng/mL (95% confidence interval: 50e140), with a sensitivity of 85% and a specificity of 72%. Screening
for CPSE is of potential value in the aging intact male dogs. Although many dogs with an Srel
2.5 show
no clinical signs, the insidious nature of BPH supports further investigations of the prostate in these dogs,
corresponding to a CPSE concentration of approximately 90 ng/mL or higher.
© 2017 Elsevier Inc. All rights reserved.

1. Introduction hyperplastic prostates and the corresponding figure in the age

Dogs live long lives, and median longevities exceed 10 years for
several breeds [1]. With increased longevity, the quality of life at
older age becomes more important. BPH affects an increasing
proportion of intact male dogs with advancing age [2], but is often
clinically unnoticed [3e5]. In beagles 1e3 years old, 25% had
* Corresponding author.
E-mail addresses: Bodil.Strom-Holst@slu.se (B.S. Holst), elin.holmroos@hotmail.
com (E. Holmroos), Lisa.Friling@uds.slu.se (L. Friling), Sofia.Hanas@slu.se (S. Hanås),
liss-marie@redog.se (L.-M. Langborg), mikael.franko@slu.se (M.A. Franko), Kerstin.
Hansson@slu.se (K. Hansson).
group 5e10 years was 88% [6]. In another study, the prevalence
increased to almost 100% after six years of age [7]. The pathologic
features changes with age, from a glandular to a complex form [2].
The definition of BPH varies, and is based on histology, an increase
in prostatic volume, or an increased in volume together with clin-
ical signs [6,8e10]. When clinical signs are detected, they may
include signs related to the genitourinary tract (including
sanguineous urethral discharge and decreased fertility), the
gastrointestinal tract (e.g. constipation and dyschezia), lameness,
and systemic signs such as anorexia [3,4,9,11,12]. However, clinical
signs are not always present in dogs with histopathological changes
of BPH, and, when present, they may go unnoticed by the owners, at

http://dx.doi.org/10.1016/j.theriogenology.2017.01.032
0093-691X/© 2017 Elsevier Inc. All rights reserved.
34 B.S. Holst et al. / Theriogenology 93 (2017) 33e39
least initially.
The etiology and pathophysiology of BPH is not fully clarified,
but the condition is androgen dependent, and dihydrotestosterone
appears to play a major role [2,13,14]. BPH can be induced experi-
mentally in dogs by a combined treatment with androgens and
estrogens [15]. The histopathological changes in experimentally
induced disease and the naturally occurring disease are identical
[6]. The androgen receptor (AR) is expressed in epithelial cells of
both normal and hyperplastic tissue [16], and the expression of the
AR and the estrogen receptor (ER) a decreases when the hormonal
influence is withdrawn [17].
Diagnosing BPH is challenging. Hyperplasia is a histologic term,
and when BPH is defined as “a large prostate with histologic veri-
fication of hyperplasia and no other histologic lesion” [9], or
“obvious increase in the amount of secretory epithelium” [6], this
necessitates a biopsy for histology, that can be considered the gold
standard to diagnose BPH in the living animal. However, this is not a
routine procedure in clinical practice. Others describe BPH merely
as an age-related increase in prostate volume and weight [10]. In
the clinic, the diagnosis is usually based on clinical signs in com-
bination with an enlarged prostate. Determining the size is thus an
important step in BPH investigations, but it should be kept in mind
that dogs may have prostates of varying sizes without clinical signs
[18]. The size of the prostate may be approximated through
transrectal digital examination. The method has low sensitivity [5];
often only the caudal part of the prostate is reached. The procedure
may be particularly difficult in both small and large breeds, but it
allows investigation of e.g. asymmetry and consistency, and is
valuable for detecting prostatic pain. Ultrasonography is often the
method of choice for determining prostatic size [19,20]. The
method has the additional advantage of showing the internal pa-
renchyma, allowing visualization of e.g. mineralization and cyst-
like lesions, and evaluation of echogenicity and echotexture
[18,21,22]. A drawback is the need for advanced equipment and
trained personnel [23].
In addition to measures of prostatic size, circulating biomarkers
may be of benefit for examination of prostatic status. Canine
prostate specific esterase (CPSE) is an arginine esterase and the
major secretory product of the canine prostate [24]. CPSE is regu-
lated by testosterone and the activity of CPSE decreases when
testosterone concentrations decline [25]. Serum CPSE activity has
been described to be significantly higher in dogs with BPH than in
normal dogs [26,27], but a difference in concentrations between
dogs with BPH, bacterial prostatitis or prostatic carcinoma has not
been shown [26]. CPSE is thus a potentially valuable biomarker, and
its use has been suggested in the diagnostic workup of dogs with
clinical signs of BPH and for screening of geriatric dogs [28]. Still,
information about the clinical use of CPSE is scarce.
The primary aim of the present study was to study the associ-
ation between the concentration of the circulating biomarker CPSE
and prostate volume. Secondary aims were to study associations
between CPSE concentration and dog age and weight, ultrasono-
graphic appearance of the prostate, and clinical signs of BPH. In
addition, the associations between the prostate volume and dog
age and weight, ultrasonographic appearance of the prostate and
clinical signs of BPH were investigated.
2. Material and methods
2.1. Study design
Owners of male dogs were invited to participate with their dogs
in the study. Written information was distributed via e-mail to
veterinary students and university employees, and also via internet
(using a facebook account of the university), and directly to dog
owners visiting the participating hospitals and clinics. Dogs from
two animal hospitals and two animal clinics were included. Breed,
age, and weight were recorded. Anamnestic information on clinical
signs of possible prostatic disease were recorded (related to the
gastrointestinal tract, genitourinary tract, locomotor system or
general signs). Information on fertility or previous matings were
not recorded. The prostate was examined by ultrasonography using
a standardized protocol (Supplementary file 1), and a venous blood
sample was collected for analysis of CPSE.
2.2. Dogs
Inclusion criteria were intact male dogs aged
four years,
weighing
five kg. Exclusion criterion was treatment with anti-
androgens within the preceding six months. In total, 79 dogs
were included; 70 purebred dogs of 40 different breeds and 9
mixed breed dogs. Their mean age was 7.5 years (SD 2.6), range
4e13 years, and mean weight 24.5 kg (SD 12.2), range 5.6e48.6 kg.
Seventy-three of the dogs had no clinical signs of prostatic
disease and six dogs had clinical signs including urination problems
and urine dripping, hematuria, and problems with defecation or
ribbon-like stools. Four of the dogs with clinical signs initially
participated as healthy volunteers, and the clinical signs were
found when recording case information.
2.3. Ultrasonography
The veterinarians performing ultrasonographic examinations
were experienced ultrasonographers and had adequate equip-
ment.1 Transducer type and frequency were optimized for the in-
dividual dog to cover the entire prostate for measurements and to
get as good detail resolution as possible for the parenchymal
evaluation.
Length, width, and height of the prostate were measured three
times on different still images (Fig. 1) and the mean value was used
in further calculations. The coefficient of variation (CV) was calcu-
lated by dividing the standard deviation from the three measure-
ments by the mean value. The prostatic volume (V) for the
individual dog was calculated according to Kamolpatana et al. as
V ¼ (1/2.6) * (Length  Width * Height) þ 1.8 cm3 [20]. In addition
prostatic parenchyma (homogenous or heterogenous), contour
(smooth or uneven); shape (symmetrical or asymmetrical), and
presence of cysts (none, 1e5 or > 5) were recorded.
2.4. Analysis of CPSE
Venous blood samples were collected and serum was stored
at 70  C until CPSE was analyzed using a commercial assay,
Odelis® CPSE (Virbac). Intra assay coefficient of variation (CV) was
<5% and interassay CV was <7%, according to the manufacturer.
Samples with concentrations
200 ng/mL were diluted 1:100 and
1:1000. According to the manufacturer, a threshold of 61 ng/mL can
be used for diagnosing BPH.
2.5. Statistical analysis
The normal prostate volume (Vnorm) was calculated according to
Sannamwong and coworkers [29]. The formula includes weight of
the dogs, and is based on healthy dogs aged 1.5e4 years:
Vnorm ¼ 0.33  BW (kg) þ 3.28. The calculated volume (V) was
1
GE Medical LOGIQ 9 Ultrasound Imaging System, General Electric Medical
Systems, MyLab™ Alpha, Esaote SpA, Philips IE33 Ultrasound, Philips Medical
Systems.
 B.S. Holst et al. / Theriogenology 93 (2017) 33e39 35

Fig. 1. a. Longitudinal section of the prostate showing a measure of the length (red line). b. Transverse section of the prostate showing the width (red line) and height (blue line).
(For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

compared to Vnorm to get a relative value of the prostatic size
(Srel ¼ V/Vnorm). Dogs with an Srel > 1 thus had a larger prostate
than would be normal for dogs 1e4 years old. The dogs were
categorized into two groups: dogs with clinical signs that may be
attributable to prostatic disease (e.g. problems with urination or
defecation, urethral sanguineous discharge, urinary incontinence)
and dogs with no clinical signs of prostatic disease.
The relative prostate size, Srel, and the concentration of CPSE
were transformed using the natural logarithm to achieve better
normality because they were found to be right-skewed. Values of
CPSE under the detection limit of 25 ng/mL were set to half this
value.
A linear regression model was fit with logCPSE as response, age,
weight, and logSrel as continuous explanatory variables and cate-
gory (clinical signs of BPH or not), contour of the prostate (smooth/
uneven), shape of the prostate (symmetrical/asymmetrical), pa-
renchyma (homogenous or heterogeneous), and presence of pros-
tatic cysts (none, 1e5, more than 5) as categorical explanatory
36 B.S. Holst et al. / Theriogenology 93 (2017) 33e39
variables. The model was then simplified by sequential elimination
of nonsignificant variables.
A linear regression model was also fit with logSrel as response,
age, weight, and logCPSE as continuous explanatory variables and
category (clinical signs of BPH or not), outline of the prostate
(smooth or not), shape of the prostate (symmetrical/asymmetrical),
parenchyma (homogenous/heterogeneous) and presence of pros-
tatic cysts (none, 1e5, >5) as categorical explanatory variables. This
model was also simplified by sequential elimination of nonsignifi-
cant variables.
Receiver operating characteristic (ROC) analysis was performed
to determine which threshold CPSE concentration that maximizes
sensitivity and specificity for detecting Srel
2.5. The rational for
choosing Srel
2.5 was that the dogs with signs related to prostatic
disease all had Srel
2.5. Since a high sensitivity was considered
more important than high specificity, the optimality criterion was
chosen as the maximum of 2  sensitivity þ specificity in order to
put double weight on high sensitivity.
A prediction analysis of Srel with respect to CPSE was performed.
The statistical package R version 3.2.3 was used for all calculations.
2.6. Ethical considerations
The study was approved by the local animal ethical committee,
permit number C89/14. Participating dog owners gave their
informed written consent.
3. Results
3.1. The prostatic size was associated with CPSE concentration and
with number of cysts
The median volume of the prostate was 22 cm3 (IQR
13e32 cm3). The median Srel was 1.9 (IQR 1.4e2.8). Seventy-five
dogs in the present study had an Srel > 1. The median CV for
prostatic length, width and height was <5% (IQR 1.1e4.6%). The
maximum CV achieved was <15%.
The six dogs with clinical signs indicative of prostatic disease
had Srel
2.5 (Fig. 1). In total, 26 of the dogs (33%) had a prostate
with an Srel
2.5. Log concentration of CPSE (P < 0.001) and
presence of more than five cysts (P < 0.001) were variables
significantly associated with logSrel (Table 1).
3.2. Ultrasonographic appearance of the prostate
The contour of the prostate was smooth in 91% of the dogs (72/
79). The shape of the gland was symmetrical in 87% (69/79). The
parenchyma was homogenous in 51% of the dogs. Prostatic cysts
were detected in 44 dogs; 30% (24/79) had 1 - 5 cysts and in 25%
(20/79) > 5 cysts were detected. Most cysts were small, but in two
dogs their diameter were between 1 and 2 cm.
Five of the six dogs with clinical signs of BPH had a smooth
contour, a symmetrical gland, and a heterogeneous parenchyma.
All of them had prostatic cysts; two had one to five cysts, and four
had more than five cysts.
Table 1
Association of logSrel with logCPSE concentration and number of cysts (n ¼ 79).
Estimate Standard error
Intercept 0.34 0.12
logCPSE 0.20 0.028
Number of cysts (1e5) 0.12 0.084
Number of cysts (>5) 0.41 0.096
Srel: Relative size of the volume, measured volume/estimated normal volume; CPSE:
Canine prostate specific esterase; R2 ¼ 61%.
3.3. The concentration of CPSE was associated with size and
contour of the prostate
The median concentration of CPSE was 75 ng/mL (IQR
28e164 ng/mL). Fifty-seven per cent of the dogs had a concentra-
tion
61 ng/mL. The log concentration of CPSE was positively
associated with logSrel and with contour, R2 ¼ 55% (Table 2).
Dogs with clinical BPH had a median CPSE concentration of
513 ng/mL, IQR 281e2137, range 54e4510, and dogs without clin-
ical signs had a median concentration of 67 ng/mL, IQR 28e138,
range 0e1444.
Dogs with an Srel < 2.5 had a median CPSE concentration of
47 ng/mL, IQR 25e99, range 0e497, and dogs with a median Srel of

2.5 had a median CPSE concentration of 275 ng/mL, IQR 105e726,
range 0e4510 (Fig. 2).
The prediction analysis showed that a CPSE concentration of
200 ng/mL predicted Srel to 2.5 (95% C.I: 2.2e2.7, Fig. 3).
The ROC analysis showed that a threshold of CPSE for estimating
Srel
2.5 with a sensitivity of 85% and a specificity of 72%
was ¼ 90 ng/mL (95% confidence interval: 50e140, Fig. 4).
4. Discussion
Diagnosing BPH is challenging, and it can be assumed that
nearly all intact male dogs will develop histopathologic BPH if they
live long enough. The diagnosis of clinical BPH is usually based on
typical clinical signs together with prostatic enlargement. However,
the size of the prostate varies widely among dogs without clinical
signs [18], which was also seen in the present study. Reliable esti-
mates of prostatic volume can be achieved with ultrasonography,
but this requires both costly equipment and trained personnel.
Clinical signs may initially go unnoticed by the owner. Additional
diagnostic tools are therefore desirable for an early detection of the
disease.
In the present study, the concentration of the circulating
biomarker CPSE was significantly associated with the relative size
of the prostate (Srel), and CPSE is thus a marker for prostatic size.
Other factors must be important for CPSE concentrations, because
there were dogs with large prostates that had low concentrations of
CPSE and vice versa. These factors are unknown, but as the secre-
tion is androgen dependent [24], androgen concentrations may
play a role. Another possible cause for varying concentrations of
CPSE in dogs with BPH would be variations in the histopathological
changes of the prostate: glandular or complex, and the degree of
fibrosis. Prostate enlargement due to other causes than BPH, such
as prostatitis and prostatic neoplasia, may also affect CPSE con-
centrations, but these conditions were less likely to be present in
the present study, based on ultrasonographic appearance and case
histories. No histopathological examinations were carried out,
which is a study limitation, and would be interesting to include in
further studies.
There is no clear cut measure to define the normal size of the
prostate. In the present study, we used a formula that was devel-
oped based on dogs aged 1e4 years, considered free of prostatic
disease based on clinical signs and ultrasonography, and that
Table 2
Association of logCPSE concentration with the logSrel and contour of the prostate.
P-value
Estimate Standard error P-value
0.006
<0.001 Intercept 3.98 0.42 <0.001
0.14 logSrel 1.93 0.23 <0.001
<0.001 Contour 0.94 0.38 0.016
CPSE: Canine prostate specific esterase; Srel: Relative size of the prostate, measured
volume/estimated normal volume, R2 ¼ 55%.
 B.S. Holst et al. / Theriogenology 93 (2017) 33e39 37

Fig. 2. Serum concentration of canine prostate specific esterase (CPSE) in relation to the relative size of the prostate (Srel). Srel is calculated as measured volume of prostate (V)
calculated as described by Kamolpatana et al. [20] divided by the normal volume (Vnorm), as calculated according to Sannamwong et al. [29]. Blue squares depict dogs with clinical
signs attributable to prostatic disease. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)

Fig. 3. Prediction of the relative size of the prostate (Srel) based on concentrations of canine prostate specific esterase (CPSE). The line describes the average Srel, predicted by
different concentrations of CPSE, and the dotted lines represent 95% prediction intervals (PI).

includes weight but not age [29]. Using this formula, 75 of the 79
dogs in the present study had an Srel > 1, which is to be expected in
this older population. Another formula has been described by Ruel
et al. [19], based on dogs from 9 months to 14 years. According to
this formula, only three of the dogs in the present study had an Srel
> 1, and only one of these had clinical signs of BPH (data not
shown). It is stated in that article that possibly some dogs with BPH
were included in the material [19]. Atalan and coworkers have
developed a formula [30], also based on dogs with a wide age range
(1e14 years). Ultrasonography is a method for estimating the size of
the prostate that has been validated [23,31]. The experience of the
observer is related to the accuracy of the measurements [23]. In the
present study, all the observers were experienced ultrasonogra-
phers with equipment well suited for the purpose, securing reliable
measures with the achieved low CVs.
The manufacturer recommends a cut-off concentration of CPSE
of 61 ng/mL for diagnosing BPH, based on a study showing 97.1%
sensitivity and 92.7% specificity using this cut-off [32]. However, in
that study the mean age of the 34 dogs with BPH was 9.25 years
whereas 45 out of 55 control dogs (82%) were less than two years
old, an age when an enlarged prostate is rare. The low age of the
control dogs likely contributes to the low number of dogs without
BPH that have a CPSE
61 ng/mL, and thus to the high specificity
even using this lower cut-off. In the present study, all dogs were 4
years or older. At this age the prostate is larger, and 57% of the dogs
had a CPSE concentration
61 ng/mL, most with no clinical signs of
BPH. The cut-off of 61 ng/mL for diagnosing BPH thus seems sen-
sitive, but rather unspecific in dogs 4 years or older. The develop-
ment of BPH is gradual and it can even using histopathology be
difficult to determine when the physiologic development stops and
38 B.S. Holst et al. / Theriogenology 93 (2017) 33e39
Fig. 4. ROC (receiver operating characteristic) analysis of canine prostate specific
esterase (CPSE) concentration for Srel
2.5.
the pathologic processes start [33,34]. The six dogs in the present
study with clinical signs indicative of prostatic disease had Srel

2.5. The ROC analysis showed that the CPSE cut-off concentration
for Srel
2.5 was 90 ng/mL (85% sensitivity and 72% specificity). In
the present study, a CPSE concentration above approximately
90 ng/mL was thus indicative of a prostate that is 2.5 times
enlarged, indicating further investigations and possibly treatment,
especially when clinical signs are present. However, because clin-
ical signs are not related only to prostatic size, 90 ng/mL should not
be regarded as a precise cut-off value. Several different breeds were
included in the present study. Their different weights were taken
into account when estimating the degree of prostate enlargement.
There are significant breed variations in prostatic size irrespective
of the weight of the dog, and notably the Scottish terrier has
significantly larger prostates than other breeds [34]. No Scottish
terrier was included in the present study. However, a larger number
of dogs and more dogs with clinical signs would be beneficial in
future studies of the diagnostic value of CPSE. Canine prostate
specific esterase is similar to human prostate specific antigen (PSA);
both are serine proteases [35]. Interestingly, serum PSA is a strong
predictor of future prostatic growth in humans [36]. The value of
CPSE for predicting future prostatic growth in dogs is not known.
From a clinical perspective this would be highly valuable infor-
mation. If CPSE concentrations can predict future growth, this will
further help in deciding whether to treat the dog or not.
In the present study, Srel was not associated with age. This may
be related to the fact that all dogs, according to the inclusion
criteria, were four years or older, and that increased growth after
this age differs between individuals. The prostate increases in size
up to four years in beagles, and the increase thereafter is at a lower
rate until at least 10 years [2,7]. In dogs of older age, the relative
weight of the prostate may start to decrease [34]. Srel was associ-
ated with the number of cysts in the present study. Cyst formation
has been described as a feature that may be prominent in prostatic
hyperplasia [6,34].
In contrast to what has previously been described [27], the
concentration of CPSE was not associated to dog age, although
multivariable analysis was not performed in the previous study.
Selection of different study populations (e.g. different age
distributions) may also explain this finding, because, as discussed
above, in the present study, neither was the relative prostate size
associated with age. The fact that prostate size and contour were
the only variables that were associated with CPSE concentrations
emphasizes the value of CPSE in diagnosing an enlarged prostate.
In retrospective studies on male dogs attending veterinary
hospitals, the incidence of suspicion of prostatic disease is low:
0.6e2.5% [3,9]. The most common prostatic disorders are prosta-
titis, BPH, prostatic adenocarcinoma and prostatic cysts [3,4,9]. In
dogs without signs of prostatic disease but presented for fertility
disorders, prostatic disorders were diagnosed more commonly
than in the general population, suggesting that prostatic disease
may be more prevalent than is usually diagnosed [4]. In one study
on beagle dogs, all males aged 6 years or older had histological
characteristics of complex benign prostatic hyperplasia [37]. The
varying definitions used together with insidious clinical signs make
it difficult to compare the prevalence of BPH in different studies.
This is similar to the situation in the human population, in which
the described prevalence of BPH varies depending on which defi-
nition is used [38]. The fact that the owners of most of the dogs with
clinical signs of BPH considered their dogs to be clinically healthy
stresses that the initial signs of BPH are indeed insidious and not
always judged to necessitate a veterinary visit. It is the re-
sponsibility of the practicing veterinarian to take actions needed to
minimize discomfort caused by BPH in the aging intact male dog.
5. Conclusion
Circulating concentrations of CPSE were associated to relative
size of the prostate and analysis of CPSE may be valuable in the
screening of geriatric intact male dogs and in the work up of BPH. A
higher cutoff than generally recommended by the manufacturer for
the diagnosis of BPH is advisable for determining an increase in
prostatic size of 2.5 or more in dogs
four years.
Acknowledgments
The skilled technical assistance by Annlouise Jansson is greatly
€
acknowledged. The authors thank Dr. Sofie Odling at Evidensia
€ msholm, Sweden, for contribution of
Specialist Animal Hospital Stro
cases. Virbac kindly supported the study with Odelis CPSE™ ELISA
kits. The study was financially supported by the Department of
Clinical Sciences, Swedish University of Agricultural Sciences,
Uppsala, Sweden.
Appendix A. Supplementary data
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.theriogenology.2017.01.032.
References
[1] O'Neill DG, Church DB, McGreevy PD, Thomson PC, Brodbelt DC. Longevity and
mortality of owned dogs in England. Vet J 2013;198:638e43.
[2] Berry SJ, Strandberg JD, Saunders WJ, Coffey DS. Development of canine
benign prostatic hyperplasia with age. Prostate 1986;9:363e73.
[3] Palmieri C, Lean FZ, Akter SH, Romussi S, Grieco V. A retrospective analysis of
111 canine prostatic samples: histopathological findings and classification.
Res Vet Sci 2014;97:568e73.
[4] Polisca A, Troisi A, Fontaine E, Menchetti L, Fontbonne A. A retrospective study
of canine prostatic diseases from 2002 to 2009 at the Alfort Veterinary College
in France. Theriogenology 2016;85:835e40.
[5] Mukaratirwa S, Chitura T. Canine subclinical prostatic disease: histological
prevalence and validity of digital rectal examination as a screening test. J S Afr
Vet Assoc 2007;78:66e8.
[6] DeKlerk DP, Coffey DS, Ewing LL, McDermott IR, Reiner WG, Robinson CH,
et al. Comparison of spontaneous and experimentally induced canine prostatic
hyperplasia. J Clin Investig 1979;64:842e9.
[7] Berry SJ, Coffey DS, Ewing LL. Effects of aging on prostate growth in beagles.
 B.S. Holst et al. / Theriogenology 93 (2017) 33e39
Am J Physiol 1986;250:1039e46.
[8] Johnston SD, Kamolpatana K, Root-Kustritz MV, Johnston GR. Prostatic dis-
orders in the dog. Anim Reprod Sci 2000;60e61:405e15.
[9] Krawiec DR, Heflin D. Study of prostatic disease in dogs: 177 cases (1981-
1986). J Am Vet Med Assoc 1992;200:1119e22.
[10] Leroy BE, Northrup N. Prostate cancer in dogs: comparative and clinical as-
pects. Vet J 2009;180:149e62.
[11] Zambelli D, Cunto M, Gentilini F. Validation of a model to develop a symptom
index for benign prostatic hyperplasia in dogs. Reprod Domest Anim
2012;47(Suppl 6):229e31.
[12] Read RA, Bryden S. Urethral bleeding as a presenting sign of benign prostatic
hyperplasia in the dog: a retrospective study (1979-1993). J Am Anim Hosp
Assoc 1995;31:261e7.
[13] Andriole G, Bruchovsky N, Chung LW, Matsumoto AM, Rittmaster R,
Roehrborn C, et al. Dihydrotestosterone and the prostate: the scientific
rationale for 5alpha-reductase inhibitors in the treatment of benign prostatic
hyperplasia. J Urol 2004;172:1399e403.
[14] Cohen SM, Werrmann JG, Rasmusson GH, Tanaka WK, Malatesta PF,
Prahalada S, et al. Comparison of the effects of new specific azasteroid in-
hibitors of steroid 5 alpha-reductase on canine hyperplastic prostate: sup-
pression of prostatic DHT correlated with prostate regression. Prostate
1995;26:55e71.
[15] Walsh PC, Wilson JD. The induction of prostatic hypertrophy in the dog with
androstanediol. J Clin Investig 1976;57:1093e7.
[16] Gallardo F, Mogas T, Baro T, Rabanal R, Morote J, Abal M, et al. Expression of
androgen, oestrogen alpha and beta, and progesterone receptors in the canine
prostate: differences between normal, inflamed, hyperplastic and neoplastic
glands. J Comp Pathol 2007;136:1e8.
[17] Gallardo F, Lloreta J, Garcia F, Moll X, Baro T, Gonzalez LA, et al. Immuno-
localization of androgen receptors, estrogen alpha receptors, and estrogen
beta receptors in experimentally induced canine prostatic hyperplasia.
J Androl 2009;30:240e7.
[18] Russo M, Vignoli M, England GC. B-mode and contrast-enhanced ultrasono-
graphic findings in canine prostatic disorders. Reprod Domest Anim
2012;47(Suppl 6):238e42.
[19] Ruel Y, Barthez PY, Mailles A, Begon D. Ultrasonographic evaluation of the
prostate in healthy intact dogs. Vet Radiol Ultrasound 1998;39:212e6.
[20] Kamolpatana K, Johnston GR, Johnston SD. Determination of canine prostatic
volume using transabdominal ultrasonography. Vet Radiol Ultrasound
2000;41:73e7.
[21] Bradbury CA, Westropp JL, Pollard RE. Relationship between prostatomegaly,
prostatic mineralization, and cytologic diagnosis. Vet Radiol Ultrasound
2009;50:167e71.
[22] Moxon R, Bright L, Pritchard B, Bowen IM, de Souza MB, da Silva LD, et al.
Digital image analysis of testicular and prostatic ultrasonographic echogencity
and heterogeneity in dogs and the relation to semen quality. Anim Reprod Sci
2015;160:112e9.
[23] Leroy C, Conchou F, Layssol-Lamour C, Deviers A, Sautet J, Concordet D, et al.
39
Normal canine prostate gland: repeatability, reproducibility, observer-
dependent variability of ultrasonographic measurements of the prostate in
healthy intact beagles. Anat Histol Embryol 2013;42:355e61.
[24] Chapdelaine P, Dube JY, Frenette G, Tremblay RR. Identification of arginine
esterase as the major androgen-dependent protein secreted by dog prostate
and preliminary molecular characterization in seminal plasma. J Androl
1984;5:206e10.
[25] Juniewicz PE, Barbolt TA, Egy MA, Frenette G, Dube JY, Tremblay RR. Effects of
androgen and antiandrogen treatment on canine prostatic arginine esterase.
Prostate 1990;17:101e11.
[26] Bell FW, Klausner JS, Hayden DW, Lund EM, Liebenstein BB, Feeney DA, et al.
Evaluation of serum and seminal plasma markers in the diagnosis of canine
prostatic disorders. J Vet Intern Med 1995;9:149e53.
[27] Wolf K, Kayacelebi H, Urhausen C, Piechotta M, Mischke R, Kramer S, et al.
Testicular steroids, prolactin, relaxin and prostate gland markers in peripheral
blood and seminal plasma of normal dogs and dogs with prostatic hyper-
plasia. Reprod Domest Anim 2012;47(Suppl 6):243e6.
[28] Levy X, Nizanski W, von Heimendahl A, Mimouni P. Diagnosis of common
prostatic conditions in dogs: an update. Reprod Domest Anim 2014;49(Suppl
2):50e7.
[29] Sannamwong N, Saengklub N, Sriphuttathachot P, Ponglowhapan S. Formula-
derived prostate volume determination of normal healthy intact dogs in
comparison to dogs with clinical BPH. In: 7th international symposium on
canine and feline reproduction; 2012. p. 226. July 26-29.
[30] Atalan G, Holt PE, Barr FJ. Ultrasonographic estimation of prostate size in
normal dogs and relationship to bodyweight and age. J Small Anim Pract
1999;40:119e22.
[31] Atalan G, Holt PE, Barr FJ, Brown PJ. Ultrasonographic estimation of prostatic
size in canine cadavers. Res Vet Sci 1999;67:7e15.
[32] vy X, Mimouni P. Hyperplasie be
Le nigne de la prostate: actualite
s. Le Point
Vete
rinaire 2009:39e43.
[33] Schlotthauer CF, Bollman JL. The prostate gland of the dog. Cornell Veteri-
narian 1936:26.
[34] O'Shea JD. Studies on the canine prostate gland. I. Factors influencing its size
and weight. J Comp Pathol 1962;72:321e31.
[35] Dube JY, Lazure C, Tremblay RR. Dog prostate arginine esterase is related to
human prostate specific antigen. Clin Investig Med Med Clin Exp 1986;9:
51e4.
[36] Roehrborn CG, McConnell J, Bonilla J, Rosenblatt S, Hudson PB, Malek GH, et al.
Serum prostate specific antigen is a strong predictor of future prostate growth
in men with benign prostatic hyperplasia. PROSCAR long-term efficacy and
safety study. J Urol 2000;163:13e20.
[37] Lowseth LA, Gerlach RF, Gillett NA, Muggenburg BA. Age-related changes in
the prostate and testes of the beagle dog. Vet Pathol 1990;27:347e53.
[38] Bosch JL, Hop WC, Kirkels WJ, Schroder FH. Natural history of benign prostatic
hyperplasia: appropriate case definition and estimation of its prevalence in
the community. Urology 1995;46:34e40.