Pyometra in a pluriparous Standardbred mare.

Keywords: pyometra, equine, treatment.

Primary author: Dr Rob Lofstedt.  Dept of Health Management, Atlantic Veterinary College.  Additional authors:  Dr Anna  Potter (primary clinician & corresponding author: apotter@upei.ca ) and Dr Martha Mellish, both of the section of Theriogenology, Dept of Health Management, Atlantic Veterinary College & Dr Shannon Martinson of the Department of Pathology and Microbiology, Atlantic Veterinary College. Editors: Drs Rob Lofstedt (lofstedt@upei.ca) and Allan Gunn (algunn@csu.edu.au).

A 21 year-old nulliparous Standardbred mare in a teaching herd was euthanized because of respiratory pathology, lameness and a diagnosis of pyometra. Understandably, this mare had experienced numerous per vagina examinations, uterine cultures, uterine biopsies and at least one hysteroscopy.

Early in April 2019, an enlarged, fluid-filled uterus was palpable. On transrectal ultrasonography, this fluid was partially echogenic and suspected to be pus (see figure 1). Due to its size and dependency, it was not possible to delineate the uterus. It was also not possible to palpate the ovaries. Trans-abdominal ultrasonography revealed that the uterus was highly distended and at its cranial aspect, lay adjacent to the xiphoid process. Nevertheless (and typical for mares with pyometra) the mare was bright, alert and responsive and had a normal hemogram.

The mare’s ovaries containing no luteal tissue; not an unusual finding in equine pyometra. Pyometra in mares does not necessarily develop in a progesterone dominated milieu i.e. unlike the situation in cattle, cervical closure and myometrial quiescence due to progesterone is not required for the development of pyometra. In mares it is more likely that retention of  pus within the uterus is a function of deficient myometrial activity than cervical pathology (See LeBlanc et al 1994 and Troedsson 1999), especially in older animals. Occasionally however, cervical pathology is implicated.

Figure 1. Note the accumulation of moderately echogenic pus within the uterine lumen and follicles within the ovaries. This image also serves as an excellent example of  two common ultrasonographic artifacts; reverberation and enhancement through transmission (ETT). Image Copyright: Dr Martha Mellish. mmellish@upei.ca  Image size: 862 x 947 px.

During per-vagina examination, the mare’s cervix was found to be closed tightly. Cervical patency was only established after approximately 20 minutes of digital manipulation and only then, did pus drip from the vulva lips as shown in figure 2. This suggested that cervical pathology was indeed implicated in retention of pus within the uterus, perhaps in conjunction with poor myometrial tone as discussed earlier. As seen in figure 2, a large diameter stomach tube was eventually passed through the cervix to drain the uterus. Cytology and culture were performed prior to drainage of the pus.

Figure 2. The main image shows how pus dripped from the mare’s vulva after gradual dilation of the cervical canal. The inset shows how pus was then drained from the uterus. Image Copyright: Dr Martha Mellish. mmellish@upei.ca  Image size: 1513 x 1024 px

Several days after initial drainage, the uterus was  flushed repeatedly with saline (see figure 3). Then oxytocin was administered to facilitate the expulsion of any remaining fluid. In addition, 1000 mg of prostaglandin E-1 (misoprostol) in methylcellulose was applied to the cervix and cervical canal to facilitate dilation for drainage and further treatment.

Figure 3. Serial saline flushes showing the increase in clarity with each successive flush. Copyright: Dr Anna Potter apotter@upei.ca Image size 2000 x 1215

Pending culture and sensitivity results, the uterus was lavaged daily with lactated ringers or saline solution.  Two days after the initial drainage, culture results revealed Pseudomonas aeruginosa, sensitive to gentamicin, amikacin and enrofloxacin.  Treatment was initiated with 1g of gentamicin buffered with 10ml of 8.4% sodium bicarbonate infused into the uterus after uterine lavage, every 24 hours for five days.  On the fourth day of antibiotic treatment, an additional treatment of 100ml of 90% DMSO was added into the second to last flush for three days. These treatments were followed by twice daily oxytocin treatment at appropriate intervals.

Two weeks after initial dilation of the cervix and uterine flushing, ultrasonography revealed a uterus devoid of any free fluid, multiple small follicles in both ovaries and a corpus luteum. Uterine culture at that time revealed Citrobacter koseri and Streptococcus zooepidemicus. However, approximately a month days later, the uterus had re-filled with pus and uterine culture again revealed growth of Streptococcus zooepidemicus. Six days later i.e. approximately 48 days from the initial dilation of the cervix, approximately 2.5 liters of pus was drained and treatment with saline flushes, DMSO and oxytocin treatment were re-started. On this occasion, treatment lasted for four days; the antibiotic being penicillin, not gentamicin. On day four, another uterine culture revealed growth of Citrobacter koseri.

The mare was re-examined about three and a half months after initial presentation and summer rest on pasture. At that time, three liters of pus was drained from her uterus, followed by saline lavage.

In early November, 2019, re-examination revealed that a large amount of pus had accumulated in the uterus again. The mare was then euthanized and submitted for post mortem exam. Her pus-filled uterus is shown in the inset of Figure 4.

Figure 4. The pus-filled uterus of the mare seen during postmortem examination. The appearance of pus shown here is typical for equine pyometra. Image Copyright: Dr Shannon Martinson. smartinson@upei.ca  Image size: 1500 x 911 px.

Figure 5. It is probable that the cervical lesion seen here was implicated in the development of pyometra, together with general cervical cicatrization and myometrial compromise. The cause of the lesion was unknown and could not have been caused by foaling as this was a nulliparous mare. Image Copyright: Dr Shannon Martinson. smartinson@upei.ca Image size: 2002 x 1232 px.

Bearing in mind the difficulty of cervical dilation in this case and the absence of spontaneous drainage of pus before and after treatment, cervical pathology (seen in figure 5; probably fibrosis after cervical damage) was probably important in the development of pyometra in this case.

Editor’s comments: In light of the typical recurrence and usually dismal prognosis of equine pyometra, some may be critical of the handling of this case with regard to repeated attempts at physical drainage, antibiotic and anti inflammatory treatment. Consider however, that in recent years, the use of cervical wedge resection and intra-cervical stents have provided favorable outcomes in cases of equine pyometra. This suggests that even if one presumes the presence of myometrial inadequacy, attempts to dilate the cervix both physically and hormonally (using PGE analogs) may hold merit and should be considered in some cases. Of course, wedge resection and cervical stents should be considered as well.

The bacteriology in this case also deserves comment. The significance of any of the bacteria isolated is open to question. For example, a literature search of Citrobacter koseri is devoid of examples of infertility in mares so this can be presumed to be a contaminant.  Also, Streptococcus zooepidemicus is not only a common cause of endometritis in mares, it is a common commensal and contaminant of uterine cultures as well. Also, Pseudomonas aeroginosa is commonly isolated from soil samples and it therefore likely to be present on mares at pasture. The absence or presence of bacteria in single or serial cultures in this mare is also open to discussion; bacterial cultures vary in success according to sampling or culture methods. Essentially therefore, the bacterium or bacteria responsible for pyometra in this case remains a question.

Selected references:

Aguilar, J. et al. 2006. Importance of using guarded techniques for the preparation of endometrial cytology smears in mares. Theriogenology. 66:423-430

Arnold, C.E. et al. 2015 Cervical wedge resection for treatment of pyometra secondary to transluminal cervical adhesions in six mares. J. Am Vet Med Assoc. 246: 13540-1357

Blanchard, T. et al 1981. Comparison between two techniques for endometrial swab culture and between biopsy and culture in barren mares. Theriogenology 16: 541-552

Ismaïl, R. et al 2013. Methods for recovering microorganisms from solid surfaces used in the food industry: A review of the literature. Int. J. Environ. Res. Public Health. 10:6169-6183

Katila, T, 2016 Evaluation of diagnostic methods in equine endometritis. Reproductive Biol. 16:189-196

Krohn, J. et al 2019 Use of a cervical stent for long‐term treatment of pyometra in
the mare: A report of three cases  Reprod Dom Anim. 54:1155–1159.

LeBlanc M.M. et al. 1994 Scintigraphic measurement of uterine clearance in normal mares and mares with recurrent endometritis.
Equine Vet.J. 26: 109-133

Pasolini M.V. et al. 2015 Endometritis and infertility in the mare – The challenge in equine breeding industry–A review. Open access peer-reviewed chapter.

Rötting A.K. et al 2004 Total and partial ovariohysterectomy in seven mares. British Equine Vet. J. 36:29-33

Troedsson. M.H.T. 1999 Uterine clearance and resistance to persistent endometritis in the mare. Theriogenology. 52:461-471

Endoscopy and ultrasonography of pyometra

Keywords: pyometra, mare, uterus

This image shows the appearance of pyometra in the uterus of a mare during postmortem examination. In this case the condition was supported by a progestogenic environment; two corpora lutea (arrows) being visible on the right ovary. If she had not been euthanized, this mare would have been treated with prostaglandins and physical drainage.

Image size: 1800 x 1052px

Pyometra may be characterized by a purulent discharge but often it is only discovered on routine ultrasonography.  The pus in cases of pyometra is often highly echogenic due to its cellular content. However, as seen in the inset ultrasonographic images shown here, it can be remarkably non-echogenic. In such cases, endoscopy will ascertain the diagnosis (seen inset image at upper right).

The general characteristics of pyometra

Keywords: pyometra, mare, equine

A mare with pyometra discharging pus from her vulva. Note the copious amounts of pus on the floor (arrow). Pyometra is easily appreciated by transrectal ultrasonography (inset).

Image size: 1800 x 1321px

Pyometra in mares is unlike the condition in cows. It is not usually a postpartum phenomenon and does not invariably block estrous cycles as it does in cows. Pyometra is often an incidental finding in mares unless there is a purulent discharge as occurred here. It is usually due to infection with Streptococcus. zooepidemicus.

Affected mares (and cows) are not usually systemically ill like bitches and queens; not even the hemogram is significantly affected. Contrary to what is occasionally stated, the cervix does not have to be abnormal for mares to develop pyometra.

Like cows, mares are treated with prostaglandin if a corpus luteum is present, but unlike cows, the cornerstone of treatment involves physical drainage of the uterus, not prostaglandin treatment on its own.

During treatment, the cervix is dilated manually and a sterile stomach tube or stallion catheter inserted. Saline is admitted to the uterus to start a siphon and the pus is then drained. Volumes of pus are variable but can be great. This bucket in this image was one of two drained from this mare. The uterus is flushed with saline after it has been drained and about 10 million IU of penicillin (sodium or potassium) in ~100 ml of saline are left in the uterus. Systemic antibiotics are not usually required.

Unfortunately the condition usually re-occurs; some mares requiring physical drainage every two or three months.  The prognosis for complete resolution of pyometra is usually dismal.

Hysterectomy is not a common form of treatment because of its surgical difficulty in mares.

Although there is little evidence to support such a statement, it is thought that failure to drain pus on a regular basis may predispose mares to uterine rupture.

Pyometra and the use of intrauterine spheres .

Keywords: sphere, marble, uterus, pyometra, luteal

A 14-year-old Quarter Horse mare was presented for a chronic vaginal discharge of several months duration. Although there was a history of breeding, details of the breeding dates were unknown. Therefore a pregnancy diagnosis performed by the rDVM. This diagnosis was positive for pregnancy but was presumably not done with the assistance of ultrasonography (see below). A chronic discharge began after the pregnancy diagnosis and the rDVM and (presumably after a negative pregnancy diagnosis), the uterus was flushed with saline and intrauterine antibiotics were administered. The attending veterinarian reported  that there was a mummified fetus could be palpated within the uterus. Oxytocin was ineffective in expelling this object so the case was referred to the AVC.

Transrectal ultrasonography revealed a uterus flilled with moderately echogenic fluid. This is shown below. Although pus need not be echogenic, the appearance of this uterus was consistent with most cases of pyometra in mares. The image at left, shows a cross section of a uterine horn and the image at right, the body, adjacent to the cervix (arrow).

Image size: 902 x 480 px

The pyometra was supported by a corpus luteum in the left ovary:

Image size: 640 x 480 px

The highly echogenic object seen in this video was present close to the midline. Both observers (I wish to acknowledge the assistance of my colleague Dr B. Crane on this case) were initially startled at the extremely symmetrical and highly echogenic appearance of the calvarium of the supposed "mummified fetus". Our impression was that only the head of the mummified fetus remained! However it soon became obvious that this was some other sphere-like object; possibly a marble. The reader should recall that there was no history of any insertion of a marble in this animal.

Play the video here using the built in controller or click on the link in the image. The FULL SCREEN setting is not recommended because the video has been compressed for mobility.

Prior to extraction of the object, it was noted that a pre-existig tear was present in the cervix at the seven o'clock location. The depth of the tear was approximately 2 cm caudal to cranial. 

The mare was tranquilized with detomidine and analgesia was supplied with butorphanol. Over a 1 1/2 hour period, the cervix was gradually dilated using digital expansion. Then, using transrectal manipulation, the object was moved towards the cervix. A black glass sphere 3.5 cm in diameter (a "marble") was delivered through the cervix

Image size: 3325 x 4720 px

Following extraction of the marble, several liters of purulent material were drained from the uterus. The uterus was then flushed serially with a total of 15 L of normal saline. Ten million international units of sodium penicillin G. was instilled into the uterus and because of the presence of a corpus luteum in the right ovary, prostaglandin was administered to cause luteolysis.

The mare was discharged to the care of the referring veterinarian with instructions to re-evaluate her uterus in approximately 2 weeks and again several months later to exclude the possibility of chronic pyometra.

Notes: The use of marbles and other spherical structures has been investigated over the last 10 to 15 years in horses for the purpose of suppressing estrus in mares. The motivation behind doing so is to suppress estrus that is thought to compromise athletic performance or behavior while showing.  In some cases it has be used to prevent unwanted conception. As noted by Dr A. Tibary, a similar technique has been used by Bedouin camel herders, using date pits instead of marbles.

Interestingly, the correlation between estrous behavior and athletic ability has not been well studied and opinions are subjective at best.  This also applies to perceived links between the estrous cycle and problem behavior.  Serum progesterone samples taken during “problem” periods of behavior have led this author to believe that many such problems are not related to ovarian activity at all. That is not to say that they cannot be. Indeed the author has had first-hand and unequivocal experience of this on stud farms. However, when problems have been related to ovarian activity for lack of a better explanation, the diagnosis must be approached with circumspect. This is true even when the owner reports that performance or behavior has improved after treatment because owners seek to believe that the expense or discomfort for the mare was justified.  

Most important is that fact that it is often impossible to know if a mare is in estrus without force teasing (nose-to-nose; nose-to-tail) or ultrasonograpy even if her behavior on casual observation suggests otherwise. Indeed, it has been well established that ovariectomized mares and those with inactive ovaries can show overt signs of estrus. Also,  many cases of presumed estrus-associated behavior are reported in the winter when most mares are not cycling. 

  

When intrauterine spheres were used, some studies have shown that mares did not respond at all but in most studies, 40 to 75% of mares showed prolonged period of diestrus.  This is because the corpus luteum fails to undergo luteolysis. Failure of luteolysis is not a consequence of exhaustive release of prostaglandin but instead it appears to be because prostaglandin secretion is suppressed.   The reason for this is currently unknown (2013) and is especially perplexing because most of these spheres remain in the center of the uterus, close to the body, not migrating like conceptuses.

When a marble causes failure of luteolysis, the corpus luteum is retained for a variable period, but usually between 60 and 90 days; perhaps as long as 150 days in some cases. Obviously field-use of marbles is not usually accompanied by progesterone assays or routine teasing so an accurate estimate of the range of luteal retention is not available.

In this case, a 35 mm glass marble was present but silver, copper or glass or water filled plastic spheres have been used as well.  Smaller diameter marbles have also been used but published evidence (Nie, G.J. et al 2003) suggests that larger spheres are more effectively retained.

Most pertinent to this case is the fairly consistent finding in all studies that intrauterine “marbles” do not usually appear to cause endometritis or pyometra. In a few cases, only slight fluid accumulation and mild endometritis was present. Furthermore, in one study, treated mares had normal conception rates after the spheres were removed.  Therefore, it is probably not justified to imply that this marble was the cause of pyometra in this case. Indeed, the cervical tear in this case (and the age of the mare) probably played a greater role in the genesis of pyometra than did the intrauterine marble.

The use of marbles to suppress estrus in mares remains a contentious issue.

The ultrasound image of the marble lends itself to an interesting discussion on artifacts. Some of them are shown here. The pronounced acoustic interface is probably due to a gas bubble. Note the over-estimation of the size of the marble in the ultrasound image.

Image size: 637 x 450 px. With thanks to Dr Paul Rist (ACVR) pending changes

References:

1.  Hedberg, Y. 2006 Oestrus in the mare with emphasis on deviant behaviour and adrenal gland function Alm Acta Universitatis Agriculturae Sueciae ISSN 1652-6880  ISBN 91-576-7250-4
2. Turnbull,  E.B.  et al. 2010 The effect of intra-uterine devices on the reproductive physiology and behaviour of pony mares. The Vet J. 186: 39–46
3. Nie, G.J. 2001. Use of a Glass Ball to suppress behavioral estrus in mares  Proc. AAEP 47: 246 -248
4. Nie, G.J. 2003. Use of an intra-uterine glass ball protocol to extend luteal function in mares. J. Equine. Vet. Sci. 23: 266-273
5. Rivera del Alamoa, M.M. et al, 2008. Mechanisms behind intrauterine device-induced luteal persistence in mares.  Anim. Reprod. Sci.  107: 94–106