Theriogenology 86 (2016) 1399–1402

Contents lists available at ScienceDirect

Theriogenology
journal homepage: www.theriojournal.com

The effect of two levels of hemospermia on stallion fertility

C.E. Turner*, S.R. Walbornn, T.L. Blanchard, D.D. Varner, S.P. Brinsko,
K.A. LaCaze, S.R. Teague, C.C. Love
Department of Large Animal Clinical Sciences, College of Veterinary Medicine and Biomedical Sciences, Texas A&M University,
College Station, Texas, USA

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

Article history: Hemospermia can occur consistently or intermittently in stallion ejaculates and may cause
Received 26 February 2016 a reduction in the fertility of the affected ejaculate. It is unknown what amount of blood in
Received in revised form 29 April 2016 an ejaculate leads to subfertility. This study investigated the effect of higher and lower
Accepted 29 April 2016
levels of hemospermia (50% and 5%, respectively) on fertility using 24 reproductively
normal mares inseminated over three consecutive estrous cycles with fresh extended
Keywords:
semen. Mares inseminated with a 5% blood-contaminated ejaculate became pregnant at
Hemospermia
the same rate (75% per cycle; 18 of 24) as the mares inseminated with blood-free (control)
Stallion
Fertility semen (75% per cycle; 18 of 24). The ejaculates containing 50% blood were sterile (0% per
Blood cycle, 0 of 24). We concluded that it is the amount of blood, not the mere presence of
Sperm blood, in an ejaculate that impacts fertility.
Ó 2016 Elsevier Inc. All rights reserved.

1. Introduction ejaculate being sterile”; however, our anecdotal experience

Hemospermia, the presence of blood in semen, can
occur consistently or intermittently and has been described
as a factor associated with infertility or subfertility in
stallions [1]. It has been observed in many breeds and has
been reported to occur more commonly in stallions that are
used heavily for breeding [2,3]. The causes of hemospermia
in the stallion are varied and include penile and urethral
lacerations [4], urethritis [4], neoplasia [5], blocked
ampullae [6], epididymitis [7], inflammation of the acces-
sory sex glands [8], urethral rent [2], coital exanthema [7],
cysts of the uterus masculinus [9], and cutaneous habro-
nemiasis [3]. These conditions can result in variable
amounts of semen contamination with blood. Voss et al.
[10] reported that the cause of subfertility may be more
related to the presence of the erythrocytes than serum.
Voss and Wotowey [1] stated that “stallions that have overt
hemorrhage in each ejaculate are sterile, while the stallion
that bleeds occasionally is infertile with the affected
* Corresponding author. Tel.: (979) 845 3541; fax: (979) 458 4380.
E-mail address: ceturner@rocketmail.com (C.E. Turner).
suggests that it is the amount of blood (i.e., pink tinged to
overt hemorrhage) that impacts fertility. The aim of this
study was to evaluate the effects of two levels (low and
high) of blood contamination (5% and 50% v:v, respectively)
of stallion semen on fertility.
2. Materials and methods
2.1. Animals
The study was conducted in southeast Texas from
August to November of 2015. Animal use and procedures
were approved by the Texas A&M University Institutional
Animal Care and Use Committee (IACUC 2015–0026).
Twenty-four light breed mares (5–18 years) and one
light breed stallion (26 years), all in good body condition
[11], were included in the study. Mares were maintained on
pasture with ad libitum access to water and coastal
bermudagrass hay. The stallion, with known normal
fertility, was housed in a 3.25
3.88-m box stall with daily
hand walking and a diet of coastal bermudagrass/alfalfa
hay and concentrate with ad libitum access to water.

0093-691X/$ – see front matter Ó 2016 Elsevier Inc. All rights reserved.
http://dx.doi.org/10.1016/j.theriogenology.2016.04.084
1400 C.E. Turner et al. / Theriogenology 86 (2016) 1399–1402

2.2. Semen collection
Gel-free semen was collected using a Missouri model
artificial vagina (Nasco, Fort Atkinson, WI, USA) lightly
lubricated with sterile, nonspermicidal gel (Therio-Gel;
Agtech, Inc., KS, USA) and fitted with an inline nylon filter
(Animal Reproduction Systems, Chino, CA, USA). A mare
exhibiting behavioral estrus was used for sexual stimula-
tion, and a breeding phantom was used as a mount source.
Immediately before semen collection, the stallion’s erect
penis was cleansed with warm water and thoroughly dried.
2.3. Semen processing
Immediately after semen collection, the gel filter was
removed from the collection bottle, and the gel-free semen
was weighed for volume and placed in an incubator (37  C).
Sperm concentration was determined using the Nucleo-
Counter SP-100 (ChemoMetec A/S, Allerød, Denmark) as
previously described [12]. Prewarmed (37  C) extender
(INRA 96; IMV Technologies, L’Aigle, France) supplemented
with ticarcillin/clavulanate (1 mg/mL; Timentin; Glax-
oSmithKline, Research Triangle Park, NC, USA; INRA-T) was
added to an aliquot of neat semen to obtain a sperm
concentration of 30
106/mL. This aliquot was used to
examine sperm motility (IVOS version 12.2 L; Hamilton
Thorne Biosciences, Beverly, MA, USA) and viability [13] of
each ejaculate to verify that semen quality was represen-
tative of what the stallion typically ejaculated (% total
motility 41, average 59%; % viability 39, average 50%) for
use in the experiment.
For the fertility trial, a portion of each ejaculate was
extended 1:1 with INRA-T in a 50-mL conical tube and was
subjected to cushioned centrifugation using MaxiFreeze
cushion fluid (1 mL; IMV technologies, L’Aigle, France) as
reported previously [14]. After centrifugation, the super-
natant was immediately aspirated, and the majority of the
cushion fluid was removed. The remaining sperm pellet
was resuspended to 25
106 sperm/mL using each of the
following diluents: (1) INRA-T (T0; control); (2) INRA-T
containing 5% whole blood (T5); or (3) INRA-T containing
50% whole blood (T50). The final inseminate volume
(10 mL) contained 250
106 sperm with 0%, 5%, or 50%
(v:v) blood. All blood for the experiment was obtained from
the project stallion by jugular venipuncture in red-top
tubes. Blood was collected immediately before mixing
with semen and within 5 minutes before insemination to
avoid clotting.
NJ, USA; 250 mg intramuscularly). Once mares developed a
follicle with an average diameter 35 mm and were
deemed in estrus because of the presence of uterine
edema, a relaxed cervix, and/or behavioral signs of estrus,
they were administered deslorelin (SucroMate; Thorn
BioScience LLC, KY, USA; 1.8 mg intramuscularly) and
inseminated once 24 hours later. Ovulation was confirmed
by ultrasound and palpation at 24-hour intervals. All
mares ovulated by 48 hours after insemination. Mares that
accumulated fluid after breeding were examined daily
until no detectable fluid remained. Intrauterine fluid was
generally not detectable by 2 days after insemination. No
mares were treated for intrauterine fluid. Pregnancy
diagnosis was determined at 14 days after ovulation. All
mares were administered cloprostenol (250 mg intramus-
cularly) on Day 14 after ovulation regardless of pregnancy
status. Embryonic vesicles present were not reduced
manually in an effort to minimize manipulation of the
reproductive tract. Mares were inseminated on the next
estrus for each of the remaining treatments using the
same protocol.
2.5. Statistical analysis
Data were analyzed using a general linear model fitted
with the method of linear least squares (PROC ORTHOREG;
SAS Institute, Cary, NC, USA) to determine the effects of
mare, treatment, cycle number (i.e., 1, 2, or 3), and carry-
over effect due to treatment. Carryover effect describes
whether a previous treatment (i.e., T0, T5, T50) altered
pregnancy outcome on the next cycle. The Fisher exact test
(Statistix 9, Tallahassee, FL, USA) was used to evaluate the
effect of previous pregnancy (positive or negative) on
pregnancy rate during the next cycle. Differences were
considered significant at P < 0.05.
3. Results
Pregnancy rate was not different between groups T0 and
T5, and pregnancy rates in these groups were higher than
that of group T50 (Table 1). No effect of mare on pregnancy
outcome was detected. The carryover effect (i.e., the effect
of the semen treatment during the previous cycle on
Table 1
Per-cycle pregnancy rate by treatment (T0, T5, T50) for cycles 1, 2, and 3
[number of mares pregnant/number of mares bred (percent pregnant
per-cycle)].

2.4. Fertility study Treatment Cycle number Total

1 2 3
Each mare was inseminated on three consecutive T0 7/9 (78) 6/7 (86) 5/8 (63) 18/24 (75)c
estrous cycles such that each mare randomly received T5 3/7 (43) 10/10 (100) 5/7 (71) 18/24 (75)c
each treatment. All mares were examined by transrectal T50 0/8(0) 0/7 (0) 0/9 (0) 0/24 (0)d
palpation and B-mode ultrasonography (Edge; FUJIFILM Total 10/24 (42)a 16/24 (67)b 10/24 (42)a 36/72 (50)
SonoSite, Inc) of the genital tract for staging of the estrous Data derived from 24 mares bred on three consecutive estrous periods
cycle. These procedures were repeated as necessary for with three different treatments (T0, T5, T50) for a total of 72 cycles.
breeding management and pregnancy diagnosis for each 1 ¼ first estrous cycle; 2 ¼ second estrous cycle; 3 ¼ third estrous cycle;
T0 ¼ control, semen with 0% blood; T5 ¼ semen with 5% blood v:v;
mare for the duration study. All mares with active luteal T50 ¼ semen with 50% blood v:v.
tissue (based on palpation and ultrasound findings) were ab
Within rows, values with different superscripts differ (P < 0.05).
cd
administered cloprostenol (Estrumate; Merck & Co., Inc., Within columns, values with different superscripts differ (P < 0.05).
 C.E. Turner et al. / Theriogenology 86 (2016) 1399–1402 1401

pregnancy outcome during the next cycle) did not affect

pregnancy outcome.
Mares inseminated on cycle 2 had a higher pregnancy
rate than mares inseminated on cycles 1 or 3 (Table 1).
There was no effect of pregnancy status during the
previous cycle on pregnancy outcome during the next
cycle.

4. Discussion

In our practice, stallions often present for hemospermia

without a concurrent history of subfertility. Previously,
others have stated that the presence of any amount of blood
in stallion semen results in sterility [1]; however, an
experimental study revealed a per-cycle pregnant rate of
7.7% (1 of 13 cycles) when mares were inseminated with
raw semen containing 20% whole blood [10]. In the present
study, per-cycle pregnancy rate was high (75%) in mares
inseminated with ejaculates containing 0% or 5% whole
blood. None of the mares inseminated with a high (50% v:v)
blood concentration in the semen became pregnant. These
findings support the premise that it is the amount of blood
in the ejaculate, not just the mere presence of blood, which
affects the fertility of the semen sample.
It is not clear why large amounts of whole blood in the
semen may decrease fertility. Others have suggested that
red blood cells may exert a direct effect on fertility because
inseminates containing 20% serum did not result in lower Fig. 1. Visual evaluation of blood-contaminated semen. This figure shows
pregnancy rates [10]. Possibly, a large amount of whole the color of 0% (T0), 5% (T5), and 50% v:v (T50) blood to raw semen.
blood may simply form a clot in the uterus and hinder
intrauterine sperm transport. In our study, T50 treatments
similar to the pregnancy rate (13 of 15, 87%) of mares during
had to be inseminated within 5 minutes after the addition
cycles that followed infertile treatment (T50).
of blood to prevent extensive clot formation in the insem-
Concerns that pregnancy may have a detrimental effect
ination syringe and pipette. The T5 treatment did not result
on the ability of the mare to become pregnant during the
in clot formation before insemination. This finding could be
next cycle were unsubstantiated. Prior pregnancy status
attributed to the greater proportion of extender as well as
had no effect on pregnancy outcome on the next cycle. It
the potential anticoagulant effect of seminal plasma
has been reported that repeated breeding and subsequent
components such as citric acid [15]. If clot formation is the
uterine flushing for embryo (8-day-old) retrieval do not
primary reason for a reduction in fertility in the T50
impair embryo recovery in the later cycles [16]. The results
treatment, then the addition of anticoagulants to blood
of our study indicate that the presence of a 14-day-old
contaminated semen may preclude the adverse effects of
embryonic vesicle at the time of prostaglandin adminis-
high blood content in an ejaculate on fertility.
tration also does not affect fertility on the following cycle.
The effect of treatment on fertility of the next estrous
The amount of blood contamination in an ejaculate may
cycle (i.e., carryover effect) was measured. Mares in group
be difficult to estimate based on the gross color change. Pink-
T50 tended to become pregnant on the next cycle at a
tinged ejaculates contain less than 1% (v:v) blood, whereas
higher rate (87%) than groups T0 (38%) and T5 (35%). The
ejaculates containing 5% blood (v:v) or 50% blood (v:v) were
reason for this may be that mares bred during cycles 1 or 2
both red in appearance (Fig. 1). This makes it clinically diffi-
with either the T0 or T5 treatments would have a higher
cult to visually estimate the amount of blood contamination
likelihood of being inseminated with the T50 treatment
in an ejaculate in the range that could adversely affect
(which resulted in no pregnancies) on the next estrous
fertility. The results of this study show that it is the amount of
cycle. Thus, these mares would have a higher probability of
blood in an ejaculate, not the mere presence of blood, which
being not pregnant. Conversely, cycles that followed a cycle
reduces fertility when mares are bred with fresh extended
with a T50 treatment would have a higher probability of
semen. Although 5% blood contamination did not decrease
pregnancy because they would have been inseminated with
fertility, the minimum level of blood contamination required
either the T0 or T5 (equally fertile) treatments. One could
for reduced fertility remains unknown.
argue that a T50 carryover effect improved the likelihood of
pregnancy (i.e., imparted a therapeutic effect). This is un-
likely because the pregnancy rate (12 of 16, 71%) of cycles Acknowledgments
during which mares were bred with fertile treatments (T0
or T5) that had followed a cycle where the same mares were This study was funded by the Legends Premier Stallion
previously bred with a fertile treatment (T0 or T5) was Season Auction, Texas A&M University. The authors thank
1402 C.E. Turner et al. / Theriogenology 86 (2016) 1399–1402
Taylor Garcia for assistance with stallion and mare
management and handling during the study.
References
[1] Voss JL, Wotowey JL. Hemospermia. Proc 18th Annu Am Assoc
Equine Pract 1972;18:103–12.
[2] Sullins KE, Bertone JJ, Voss JL, Pederson SJ. Treatment of hemo-
spermia in stallions: a discussion of 18 cases. Comp Contin Educ
Pract Vet 1988;10:1395–403.
[3] McKinnon AO, Voss JL, Trotter GW, Picket BW, Shideler RK, Squires EL.
Hemospermia of the stallion. Equine Prac 1988;10:17–23.
[4] Voss JL, Pickett BW. Diagnosis and treatment of haemospermia in
the stallion. J Reprod Fertil Suppl 1975;23:151–4.
[5] Bedford SJ, McDonnell SM, Tulleners E, King D, Habecker P. Squa-
mous cell carcinoma of the urethral process in a horse with
hemospermia and self-mutilation behavior. J Am Vet Med Assoc
2000;216:551–3.
[6] Love CC, Riera FL, Oristaglio RM, Kenney RM. Sperm occluded
(plugged) ampullae in the stallion. Proc Annu Mtng Soc Ther-
iogenology 1992:117–25.
[7] Turner RM. Urospermia and hemospermia. In: Mckinnon AO,
Pycock JF, Samper JC, editors. Current therapy in equine reproduc-
tion. St. Louis: Saunders; 2007. p. 258–65.
[8] Sojka JE, Carter GK. Hemospermia and seminal vesicle enlargement
in a stallion. Comp Cont Educ Pract Vet 1985;7:587–90.
[9] Pozor MA, Macpherson ML, Troedsson MH, Klein C, Diaw M,
Buergelt C, et al. Midline cysts of colliculus seminalis causing ejac-
ulatory problems in stallions. J Equine Vet Sci 2011;31:722–31.
[10] Voss JL, Pickett BW, Shideler RK. The effect of hemospermia on
fertility in horses. Proc 8th Int Congr Anim Reprod Artif Insemin
1976;4:1093.
[11] Henneke DR, Potter GD, Kreider JL, Yeates BF. A scoring
system for comparing body condition in horses. Equine Vet J
1983;15:371.
[12] Comerford KL, Love CC, Brinsko SP, Edmond AJ, Waite JA, Teague SR,
et al. Validation of a commercially available fluorescence-based
instrument to evaluate stallion spermatozoal concentration. Proc
54th Annu Conv Amer Assoc Equine Pract 2008:367–8.
[13] Foster ML, Love CC, Varner DD, Brinsko SP, Hinrichs K, Teague S,
et al. Comparison of methods for assessing integrity of equine
sperm membranes. Theriogenology 2011;76:334–41.
[14] Bliss SB, Voge JL, Hayden SS, Teague SR, Brinsko SP, Love CC, et al.
The impact of cushioned centrifugation protocols on semen quality
of stallions. Theriogenology 2012;77:1232–9.
[15] Mann T. Biochemistry of stallion semen. J Reprod Fertil Suppl 1975;
23:47–52.
[16] Imel KJ, Squires EL, Elsden RP. Embryo recovery and effect
of repeated uterine flushing of mares. Theriogenology 1980;
13:97.