Endometritis update (Proceedings)


Endometritis is the most common infectious pathologic condition affecting the uterus in brood mares. Pathologic endometritis is an abnormal extension of a physiologic process. In the pathologic form excessive inflammation of the endometrium occurs and then persists.

Context and significance

Endometritis is the most common infectious pathologic condition affecting the uterus in brood mares. Pathologic endometritis is an abnormal extension of a physiologic process. In the pathologic form excessive inflammation of the endometrium occurs and then persists. In an industry that often lacks a focus on profitability, it is a major cause of lower pregnancy rates and poor reproductive efficiency. Mares that are prone to repeated pathologic episodes of endometritis are difficult to get in foal and are called subfertile. Hughes and Loy (1969) emphasized the economic and opportunity costs of missed conception and pregnancy. In this review the predisposing factors that culminate in endometritis, the pathogenesis of endometritis, the principles of treatment and therapy for this condition will be discussed.

Predisposing factors

When advising clients on the attributes of high quality brood mares the age of the mare and her perineal conformation are some of the most important characteristics. This is because the long term reproductive health of the mare depends on her ability to maintain her reproductive tract with minimal contamination from genital pathogens. While other factors such as the mare's: genetics, breed, pelvic conformation, immunologic constitution, reproductive history, and internal physical features such as the vestibulovaginal sphincter, cervix, ovaries and uterus, all play a role, perineal conformation is commonly what is at fault, and contributes to reduced fertility particularly as the mare ages.

Perineal conformation

Perineal conformation is very important for reproductive health. There should be a focus on the shape of the perineum and tilt of the vulva of brood mares. The vulva should be vertical in its orientation and in line with the anus. The mare should have 1/3rds of the vulva above the pelvic floor and 2/3rds below. In defective conformation the top of the vulva is recessed inward and forms a slope down towards the anus. Opening the lips of the vulva should create a situation where air rushes into the vagina. This is called pneumovagina or "wind sucking." Passage of a vaginal speculum should be met with resistance. The physical barriers, such as the vulvar seal and the position of the vulva relative to the pelvic floor, prevent manure and thus contamination from entering the reproductive tract. If these barriers are suboptimal the mare will have increased contamination of the reproductive tract with genital flora and will be predisposed to infections such as bacterial endometritis. For example pneumovagina is a common predisposing factor for endometritis, and is usually due to a conformational defect such as: abnormal angulation of the vulva (so called tipped perineal conformation) or a defective vulvar shape or seal. In some mares one half of the vulva will override the other half. Age: Older mares are over represented in the population of mares with pathologic endometritis.Age of the mare is important factor because fertility begins to decline in mares after the age of 12. Age related deterioration in fertility in mares is often linked to undesirable changes in the mare's anatomic barriers to infection, such as a sloping perineal conformation, the cumulative toll of inflammatory and degenerative processes in the uterus, the aging of the immune system and a loss of competent oocytes. The immune system of the mare ages in a process that has been referred to as "inflammaging." While veterinarians are able to do many things in terms of reproductive management we are unable to make mares younger.


Breeds such as the Thoroughbred, where the focus is more on athletic than reproductive performance, and breeds with a narrow genetic basis, such as the: Tennessee Walker, Clydesdale, Shire and Friesian are more prone to subfertility in both mares and stallions.

Impaired physical barriers

In addition mares that have cervical injuries or that sustained injuries to the vulva, tumours of the reproductive tract, endocrine disturbances, abnormal physiological responses (such as a long cervix that does not relax while in estrus), or a dependent uterine location are prone to excess genital contamination due to impaired physical barriers. Recently a laparoscopic procedure, uteropexy, was reported to help repair a pendulous uterus (Brink and Schumacher 2010).

Pathogenesis of endometritis

Endometritis is simply inflammation of the endometrium (mucosa and submucosa of the uterus), and in its most common form it is a brief physiologic event. The term is used to describe both a physiogic and pathologic process. The main means of studying endometritis scientifically is the use of an experimental uterine infection model, which utilizes the most common genital bacterial isolate, Streptococcus equi zooepidemicus (Strep). Streptococcus equi zooepidemicus is a Gram positive organism and is commonly found on the mare's skin. Experimentally fertile mares resolved the intrauterine Strep challenge quickly, while subfertile mares were noted to develop excess amounts of free intrauterine fluid, and persistent inflammation 72 hours post inoculation. It has become customary to term the mares that could clear low level Strep infections "Resistant" those that could not clear the Strep challenge as "Susceptible" (Hughes, 1969).

Pioneering work was done by investigators such as Asbury, LeBlanc, Hughes, Troedsson and Liu. Studies showed that Susceptible mares have an underlying defect in uterine clearance due to poor myometrial contractility (Rigby et al. 2001). The myometrium is further disadvantaged by having to contract against gravity in mares with a dependent (abdominal rather than pelvic) uterine location. The myometrial contractions were reported to be delayed in onset and weaker in susceptible mares (Troedsson, 2008). Weak myometrial contractions were reported to contribute to poor cervical / uterine clearance and lymphatic drainage. The underlying mechanisms for the delay in onset of the uterine contractions and poorer myometrial contractility remain unclear. Rigby et al (2001) reported that the mechanisms controlling intracellular calcium movement in the myometrium may be involved. Studies showed that breeding introduces sperm cells, seminal plasma, bacteria and debris into the uterus. Bacteria, sperm and debris were described as the main agents of inflammation with seminal plasma having a modulating role on uterine inflammation (Troedsson M.H.T., 2001; Troedsson et al., 2008).


Skin and fecal contaminants that are introduced at breeding are a common source of bacteria involved in pathologic endometritis. Post - mating endometritis is most commonly caused by genitally adapted strains of Gram positive bacteria, such as Strep equi zooepidemicus which is found on the skin. Gram negative organism such as E. coli arise from the feces, and mixed infections are usually a combination of both. Gram Positive organisms account for 70% or more of all infections. Bacteria may also be present in the uterus before breeding, or the conditions induced by breeding may facilitate bacterial growth in some mares, and bacteria are introduced by breeding or insemination. There is mounting evidence using in situ hybridization that in some mares bacteria may persist in the endometrium by forming a pathologic biofilm (Brink and Schumacher 2010). Strategies to treat suspect mares that may have biofirms are now in clinical trials (Gores-Lindholm 2009, Ahlschwede et al. 2009). The biofilm hypothesis may explain why some mares appear to be have no uterine problems until about 12 – 24 hours after breeding. Common Causes of Endometritis in Mares include Gram Positive: Streptococcus equi zooepidemicus, Staphylococcus aureus, Actinobacillus, other Strep and Gram Negative: Escherichia coli, Klebsiella pneumonia, Pseudomonas aeuroginosa capsule type 5 and Enterococcus bacteria. Anaerobes such as Bacteroides and fungi (ex. aspergillus, allescheria, morteriella) yeast (Candida) are also involved.


Historically the amount and type of opsonising antibody (IgG) was evaluated as an underlying cause of excess inflammation in mares (Asbury, 1984). This was the basis for uterine therapy using plasma as a source of IgG in subfertile mares. The recognition of Strep as main pathogen led to attempts to make a vaccine against Strep infection. These attempts were unsuccessful in remedying the inflammatory response of susceptible mares (Causey et al., 2006; Weiss et al., 2007). Sperm in utero are all coated in IgG. This IgG is present in seminal plasma and the uterine and tubal secretions (Rumke, 1974; Aguilar and Reyley, 2005). The opsonizing IgG is chemoattractive to polymorphonuclear cells (PMNs), which are predominately neutrophils, to phagocytose these sperm cells.


The IgG-coated sperm activate the complement cascade by cleavage of C5 and release of factors C5a, C5b, and C3b (Troedsson et al., 2001), which in turn activate the rest of the complement cascade within the endometrium and attract PMNs. Specifically, C5a is chemoattractive for neutrophils and induces their activation leading to degranulation, oxidative bursts, and changes in adhesiveness. The C5a also induces other signs of inflammation such as vasoconstriction and increased vascular permeability (Miller and Krangel, 1992). The C3b fragment aids activated PMNs to bind to activated spermatozoa (Troedsson, 2006b). Resistant and susceptible mares have similar levels of IgG and C3 for the first 24 hours, but after 36 hours S mares showed a continual decline whereas R mares showed an increase in levels (Troedsson et al., 1993).


In endometritis the predominant cell type is the neutrophils. Neutrophil influx was reported to be associated with prostaglandin secretion and production of nitrous oxide (Alghamdi et al., 2005).The attracted neutrophils have multiple roles including phagocytosis of bacteria, debris, and sperm. No difference in neutrophil function and ability has been detected between R and S mares (Zerbe et al., 2004). It was reported that sperm induced the breakdown of complement and that the complement fragments C5a and that C3b induced neutrophil chemotaxsis (Watson, 1988; Troedsson M.H.T., 2001). The process of phagocytosis by neutrophils of sperm and bacteria releases prostaglandin F2α (PGF2α) which elicits myometrial contractility which are necessary for the expulsion of debris through the cervix. Leukotriene B4 produced from the lipoxygenase pathway is chemotactic to equine neutrophils in vitro and is a potent stimulator of the phagocytic activity.

Seminal plasma

The endometrial inflammatory response was believed to be modulated by seminal plasma. Iit inhibited the chemotaxsis of neutrophils in vitro, was bacteriostatic and decreased the duration of breeding-induced inflammation but not the severity and seminal plasma protects sperm from opsonization (Troedsson 2001, 2002, 2005). One author showed that seminal plasma: decreases uterine contractions when compared to sperm alone, and that neutrophil counts are higher in mares bred with sperm plus seminal plasma rather than sperm alone (Portus et al., 2005). Other authors suggested that the concentrated nature of frozen semen combined with the lack of seminal plasma may trigger more inflammation in mares bred with frozen semen.

Immune responses – female reproductive tract (FRT)

The female reproductive tract has complex immune protection mechanisms. Sperm induced inflammation usually peaked at 6 – 12 hours in resistant mares (Kotilainen, 1994). The mucosal immune system is that portion of the immune system which protects the mucous membranes of the horse, such as the urogenital, gastrointestinal and respiratory systems, from invasion by potentially pathogenic microbes. It protects mucus membranes against infection, prevents the uptake of unwanted antigens, microorganisms, and other foreign materials, and moderates the horse's immune response to that material. Mucosal immunity therefore integrates innate immune mechanisms in protecting a horse from pathogens (Watson and Dixon, 1993; Wira, 2002; Zerbe et al., 2003).

Toll-like receptors and pathogen – associated molecular patterns

The innate immune system has evolved to recognize foreign structures called pathogen-associated-molecular patterns (PAMP's). These conserved molecular patterns of the PAMPs are recognized by receptors called pattern recognition receptors (PRR). Included in the PRR are TLRs. The TLRs are a conserved family of protein receptors, and have been shown to be expressed in macrophages, neutrophils, lymphocytes, endothelial cells and epithelial cells in horses (Suri et al., 2006; Berndt et al., 2009). The downstream signaling that results from the binding of PAMPs to TLRs involves a host of other cytokines and other adaptors that lead to the recruitment of immune cells, and as well as the production of intracellular and secreted anti-microbial factors. It is now established that Toll-like receptor 4 (TLR4) regulates the response to lipopolysaccharide (LPS) while Toll-like receptor 2 (TLR2) does the same for products of Gram positive bacteria. The significance of these receptors and their activators to equine innate immunity is that they concern the main causes of endometritis (Berndt et al., 2009). Clinical responses of mares to Strep and E. coli were dissimilar and different TLRs were likely to be activated by these bacteria (Zerbe et al., 2003).

Pro-inflammatory cytokines

The mediators of the damage caused by bacteria are usually pro-inflammatory cytokines. New insights into the inflammatory responses of the endometrium led researches to investigate the pro-inflammatory cytokine profiles of mares susceptible and resistant to endometritis. Fumuso showed differences between resistant and susceptible mares in the levels of expression of endometrial cytokines including of interleukins (IL) such as IL-1β, IL-6, and Tumour Necrosis Factor alpha (TNF-α) mRNA in non-bred resistant and susceptible mares (Fumuso et al., 2007). This work suggested that mares susceptible and resistant mares might have differential immunologic responses. Interleukin IL-1β, IL-6 and TNF-alpha are recognized as pro-inflammatory, and increases in their expression are associated with increased inflammation. The mediators and pathways that trigger cytokine secretion are presently being elucidated in equine tissues and are known to include activation of TLRs.


Treatment with steroids such as prednisone and dexamethasone were investigated as an adjunct to other management procedures in mares( Bucca, Carli et al. 2008). Neutrophil function in dexamethasone treated mares at 24 hrs post treatment was normal (Ohman, Klein et al.). Glucocorticoids were investigated because of their potential to have positive effects on decreasing pro-inflammatory cytokines. There are reports of the use of glucocorticoids (dexamthasone 25 mg, or prednisolone 0.2 mg/kg) to modulate the endometrial inflammatory response in mares (Bucca et al., 2008; LeBlanc and Causey, 2009). The exogenous administration of glucocorticoids has been shown to regulate TLR levels in certain animals and tissues. Glucocorticoids decrease inflammation and suppress the immune system by diffusing through the lipid membrane into the cytoplasm and binding to glucocorticoid response elements on DNA which inhibit the production of IL-1 and IL-6 and other cytokines (Almawi et al., 1996) (Brattsand and Linden, 1996). The cytokine IL-1 is elevated by immunostimulants such as MCWE. Glucocorticoids also inhibit the downstream effect of interleukins which have a positive feedback loop on further interleukin production and on the recruitment of T-cells (Almawi et al., 1996). A true evaluation of the efficacy of glucocorticoid treatment alone on endometritis is still pending (Dell' Aqua Jr et al., 2006). In other studies there was no effect of dexamethasone treatment on pregnancy rate (Bucca et al., 2008).

Immune stimulants

There is currently interest in investigating the use of immunomodulation as a therapy for pathologic endometritis. An understanding of which mares to treat with immune stimulants, how long they should be treated and with what substances remains to be elucidated. The MCWE was reported to decrease IL-1 mRNA expression (IL-1 is believed to be secreted by activated macrophages). The authors suggest that the MCWE restored homeostatic regulation of local inflammation (Fumuso et al., 2005). Killed preparations of Propionibacterium acnes have been associated with higher levels of cell mediated CD-8 T cell activation and interferon-gamma secretion, and has been termed a biologic response modifier. The MCWE product elevates IL-1. Both the MCWE and the Propionibacterium acnes preparations are labeled for use in cases of endometritis as it "normalizes" a mare's inflammatory response to insemination (Rohrbach et al., 2007). There are few reports however of these immunomodulatory products being used as a sole therapy in mares with endometritis. All of these products may be associated with transient fever and or malaise.

Diagnostic and therapeutic approach

Generally therapeutic approaches involve an investigation of the mare's history, examination of her for predisposing factors that lead to endometritis, an evaluation of her current condition by looking at her perineum, and using: vaginoscopy, ultrasonography, cytology, culture and or biopsy. Mares that are susceptible to endometritis tend to have poor tone, shortened inter-estrus intervals and early onset heavy endometrial edema when they have small follicles. Some mares with endometritis show some signs of heat, but stallions will often refuse to breed them or they will not stand. The principles of therapy involve the elimination predisposing factors, treating infection/inflammation, recognizing that mechanical clearance is a problem, and managing and treating mares to reduce the inflammatory burden (one breeding per heat, lavage and ecbolics such as oxytocin). Mares with pre-existing bacterial infections or inflammation should be treated to eliminate debris and bacteria. The perineum should be examined carefully to examine the shape, length, tilt and position of the vulva in relationship to the pelvis. Diagnosis of endometritis is made by palpation to determine uterine and cervical tone, ultrasound to see f there is free intrauterine fluid, and to confirm the mare is not pregnant. Perform a vaginal to check for pneumovagina and to look for discharge. Obtain a free uterine fluid sample, using a swab/ low volume lavage sample for cytology and culture. A finding >10% neutrophils on an endometrial swab cytology in combination with isolation of bacterial pathogen by uterine culture or low volume lavage is characteristic of endometritis. Endometrial biopsy or biopsy culture, biopsy cytology, or hysteroscopically guided biopsy is also a useful tool to diagnose endometritis. Mares with chronic infertility and no obvious clinical signs should be sampled using a low volume lavage method (60 – 100 mls PBS in followed by 20IU oxytocin). A proportion of the lavage fluid (10mL) may be centrifuged to make a cellular pellet that is used for cytology. A Diff Quik and Gram stain of the cytology and culture will allow you to determine if there is clear evidence of infection (>10% neutrophils and microorganisms).

Most types of bacteria that cause pathologic endometritis will significantly decrease the mare's chance of conception, if she is bred when she is infected. A cytology should be obtained at the same time a culture is obtained, in order to interpret the culture. The cytology, especially in mares infected with Gram positive organisms will typically have >10% neutrophils, and may have visible Gram positive cocci. The cytology will be helpful in sorting out false positive samples (where environmental contaminants are grown) but no inflammation is present. False negatives where there is a causative organism but inflammation is minimal, are usually associated with Gram negative infections, especially Pseudomonas. It is preferable to treat uterine infections before attempting breeding or rebreeding. Mares with a high degree of endometrial irritation may need a cycle off from breeding after treatment. The disturbance in the uterine milieu prevents pregnancy.

Pathologic endometritis may be diagnosed after breeding so there is race between the therapy and the infection to restore the uterus to a condition conducive to pregnancy. Microbial therapies are usually initiated by identifying the type of bacteria on a cytology in these cases. The majority of the infections are due to Gram positive cocci, so the presence of cocci in a cytology generally means a Strep or Staph infection, which are usually sensitive to penicillin. Controversy continues over whether to use intrauterine or intramuscular therapy, however usually better MIC levels are achieved over time with IM treatment. Appropriate antibiotic therapy should be identified or confirmed by culture/sensitivity, and antibiotics should be administered systemically for 5 – 7 days to avoid microbial resistance. Intrauterine antibiotics result in uterine invasion, they are costly and they may cause super- nfection, however in some mares with pathologic biofilms they may be more effective (LeBlanc and Causey, 2009). The expensive antibiotics are sometimes used IU, however IU treatment may not eliminate the organisms in other reservoirs in the mare such as the vagina, clitoral sinus or clitoral fossa. Note there are mares and owners that do not have the temperament for once or twice daily IM injections for 5 to 7 days. Intensify the management if the mare repeatedly experiences endometritis. Prebreeding saline uterine lavage in estrus a day before breeding, followed by oxytocin 20 - 40 IU. Start antibiotic therapy. Limit the number of breedings (use hCG when the preovulatory follicle is between 35 - 40 mm). Lavage at 4 - 6 hours post breeding as above with oxytocin 20 - 40 IU. Viable sperm are in uterine tube by 4 hours post-mating. Follow post breeding for fluid accumulation, may need to lavage 1-2x daily until the uterus is clean. May need to use IU antibiotics once a day for 2 – 3 days. Perform or Close Caslick after concluding post-breeding therapy.

Other therapies include

     • Chemical curettage (100 mls kerosene) administer to mare in diestrus then give PG, mechanical curettage; medical grade DMSO 300 mls cycle before breeding; infuse 60mls of acetyl cysteine diluted to a 10% solution day before breeding. Give oxytocin 4 – 6 hrs later, breed the next day. If cervix fails to dilate crush 2 - 100μg Cytotec PGE tablets. Place in sterile lube, mix, inject into cervical canal and spread onto external cervical os.

     • Other natural methods: Expose mare to sexual stimulation while still in heat to cause endogenous oxytocin release, or use low dose prostaglandin pre ov (125 μg cloprostenol), or carbetocin (0.175 mg IV) pre or post ov

     • Treat with Regumate (10cc 0.44 mg/kg) if tone remains poor by day 3 post mating

     • Referral to a theriogenologist is always an option. Don't forget it is okay to advise not breeding some of these mares. We need to cull more animals out of the breeding herd. We geld a large number of stallions, but we spay very few mares! Advice on broodmare selection is an important part of a veterinarian's role in management.

References available on request

Related Videos
© 2024 MJH Life Sciences

All rights reserved.