Breeding the ‘difficult mare’

Mares can be classed as ‘difficult’ breeders if suboptimal pregnancy rates are experienced despite adequate reproductive management. This includes breeding with semen of acceptable quality, in an appropriate time frame with relation to ovulation, with any resulting fluid being treated appropriately. Reported per cycle pregnancy rates for natural mating are 63–65% (Allen et al, 2007), with pregnancy rates for chilled semen reported to be between 44–65% per cycle (Squires et al, 2006; Douglas-Hamilton et al, 1984). First cycle pregnancy rates for frozen semen of 51.3% have been published (Loomis, 2001), and around this mark is widely accepted internationally. If a mare fails to get in foal, or if her pregnancy rate per cycle is dramatically below these averages, she may be classed as a difficult breeder.

There are multiple reasons why a mare might be a difficult breeder, and more than one factor may exist in the same mare. For example, an aged mare with uterine degeneration may retain fluid, and this may be compounded by poor vulval conformation (Figure 1). It may be obvious what is causing a mare to be difficult to get in foal, or diagnostics may be required to ascertain the cause. Sometimes a presumptive diagnosis can be made retrospectively, after a mare is bred, with information gained from a post-breeding examination. Before a mare is classified as difficult, her breeding history should be thoroughly reviewed, both to ensure adequate management has been employed and to obtain an indication as to what the problems may be. As much information as possible should be sought, including the number of times she has been bred; the type and quality of semen used; timing of breeding with relation to ovulation; timing and type of ovulation agent used; as well as post-insemination monitoring and management.

Aged mares

For reproductive purposes, a mare can be classed as aged in her mid to late teens, typically 15 years or older, although fertility begins to decline in a linear fashion from approximately 12 years of age. Work has shown that aged mares have significantly reduced per cycle pregnancy rates, as well as foaling rates, the physiology behind which has been reviewed in great detail by Madill (2002). Reasons that older mares can be more challenging include: failure to cycle (in mares ≥20 years); a prolonged transition phase; longer oestrous cycles with greater propensity to fail to respond to ovulating agents; oocyte degeneration; failure of fertilisation; increased early embryonic death; oviductal abnormalities; failure of cervical dilatation, endometrial degeneration (characterised by fibrosis, glandular dilation and gland nesting); fluid accumulation; increased rate of abortion; and poor vulval conformation resulting in contamination of the caudal reproductive tract.

Some of these problems cannot be resolved, such as age-related degeneration of the genetic material in the oocyte. However, action can be taken to limit the ramifications of some of these points in oldesr mares. For example, artificial lighting can be used to advance the onset of the breeding season to allow timely breeding and maximal duration of the breeding season. Assessing and correcting vulval conformation by means of a Caslick's procedure (Figure 2), should be done as early as possible in the breeding season, so as to limit contamination of the caudal reproductive tract when the mare is in oestrus and the cervix is open. This may mean opening of a Caslick's if she is to be bred naturally; however, if breeding by artificial insemination, opening is often not required.

Older mares tend to have slower follicular growth, therefore, after your criteria for giving an ovulation agent have been reached (e.g. a dominant follicle has been detected, with good oedema and a relaxed cervix), waiting and giving the ovulation agent 24 hours later than one would for a younger mare may lead to more reliable ovulation within the appropriate time frame. Using the most fertile semen available is, of course, beneficial. Frozen semen tends to cause a more pronounced post-breeding endometritis than chilled semen, in the majority of mares. Where possible, avoiding frozen semen in aged mares may result in comparatively increased conception rates. Sometimes, however, good-quality frozen semen may supercede poor-quality chilled semen, especially if the chilled semen is coming from stallions in great demand, or is travelling over great distances. When using frozen semen there are fewer restrictions on when the mare can be inseminated, due to reduced reliance on postal and courier services. The author prefers pre-ovulation insemination in older mares, to allow more time to deal with the inevitable post-insemination inflammation before progesterone rises and the cervix closes.

Breeding using artificial insemination (if breed allows) is preferable to fresh cover in mares with a fibrotic cervix, as not only does it minimise contamination of the reproductive tract, but also a stallion may struggle to successfully ejaculate into the uterus of such a mare. Fluid accumulation, due to less effective endometrial drainage and uterine contractility coupled with a fibrotic cervix, is common, and has been described as the ‘older maiden mare syndrome’. In addition, the uterus ‘sagging’ in a multiparous mare may increase the likelihood that she will pool fluid in the dependent parts of the uterus. If a mare is identified as likely to accumulate fluid, either from past attempts at breeding or evidence of pre ovulatory fluid seen on ultrasound (US), aggressive treatment is likely to be necessary both pre and post breeding. Lavaging the uterus with Hartmann's solution immediately before breeding has led to acceptable pregnancy rates (Vanderwall and Woods, 2003; Barker, 2018). Timely post-breeding lavage is also recommended in these mares, using Hartmann's solution or 0.9% saline, from 4 hours post breeding. This, coupled with administration of oxytocin (10–20 IU intravenously (IV), subcutaneously or intramuscularly), will aid uterine contraction and fluid clearance. For mares susceptible to persistent post-breeding endometritis, such therapies may also be combined with intrauterine antibiotic administration, if indicated by previous uterine culture and sensitivity results. If there is a positive culture on pre-breeding endometrial swabs, it may be preferable to use a cycle treating the mare, and breed her on a subsequent cycle, after a clean endometrial swab. Despite being widely used in the past, it is not best practice to routinely administer prophylactic intrauterine antibiotic treatment post breeding, in the interest of responsible antibiotic use, so clinical judgement needs to be used.

If an aged mare consistently fails to get in foal, or to maintain her pregnancy, a uterine biopsy should be taken. Mares with a biopsy Grade 3 (Kenney and Doig, 1986) traditionally have a poor prognosis (<10%) for carrying a foal to term (Table 1). Many suggest that degeneration cannot be reversed, yet there is some evidence supporting the use of physical and chemical curettage (Ricketts, 1985; Bracher et al, 1991). Although often considered as a last resort, kerosene for the purposes of chemical curettage (Bracher et al, 1991; Crabtree et al, 2012; Scoggin, 2016), has been found to be of benefit in older broodmares and embryo donors with advanced uterine degeneration, failing to conceive (personal communication, Crabtree, 2017).

It is important to consider assisted reproductive techniques in older mares. Embryo flushing bypasses some but not all of the problems encountered in an aged mare, and may be a cost-effective way of achieving a pregnancy from an older mare. Problems resulting in an inhospitable uterine environment for embryo development, such as an ineffective endometrium, are bypassed by flushing the embryo and transferring it to a younger recipient. Older mares will, however, have reduced embryo recovery rates, and reduced pregnancy rates post-transfer, as well as higher embryonic death rates by 50–60 days of gestation when compared to a younger donor mare (Table 2) (Madill, 2011). Although embryo flushing is widely available and negates many problems faced by breeding from an older mare, it is important that the client is aware of the poorer potential pregnancy rates. Oocyte collection and transfer or intracytoplasmic sperm injection may be considered, although again the quality of the oocytes collected will inevitably have an impact on the success. Work has clearly shown that reduced oocyte viability is a factor causing reduced fertility in older mares (Carnevale and Ginther, 1995). It is important that the client is informed regarding the likely pregnancy rates and costs, to ensure that they have appropriate expectations. In the future, these latter processes will likely become more widely available and as such less costly. Often, well-managed assisted techniques will be more cost-effective than multiple attempts to breed a difficult mare with repeated negative results.

Mares prone to fluid accumulation

This group consists of both mares prone to persistent breeding-induced endometritis and those with chronic infectious endometritis; clinically it can be difficult to distinguish between the two. Samples can be taken from the endometrium to help form a diagnosis, using a double-guarded endometrial swab, which can be used for cytological examination as well as for culture. A cytobrush is also useful when collecting a sample for cytological examination; however, a swab or cytobrush will only sample a small area of the endometrium, therefore low-volume lavage using 100 mL of saline or Hartmann's may be preferable (LeBlanc et al, 2007). As discussed, pre-breeding uterine lavage can prove both diagnostic and therapeutic, and should be considered in mares that have pre-breeding uterine fluid discernible on US scan. If good clinical technique is used, including appropriate mare preparation and adequate fluid recovery, there are minimal down sides to performing pre-breeding lavage, as time and cost implications are small (Vanderwall and Woods, 2003; Barker 2018). The author prefers to perform a 1-litre Hartmann's lavage as a minimum, and if deemed appropriate then samples can be submitted for cytology and culture (Figure 3) (Barker 2018). It is likely that the results of culture will not be available in time for the current cycle whereas cytology can be; nevertheless the information is useful to determine actions and/or treatments for subsequent cycles. Gross or cytological examination of the fluid retrieved may actually preclude the mare from breeding that cycle, and instead treatment may be instigated to ensure she is clean for subsequent cycles. If pre-breeding lavage is performed, oxytocin can be administered IV up to an hour before breeding, due to its short half-life. Systemic antibiotic use may be considered and started in advance of breeding, if previous culture and sensitivity results support their use. Oral trimethoprim-sulfadiazine is most often used and has good penetration of the urogenital tract.

Mares prone to fluid accumulation should also be lavaged in a timely fashion post breeding, and lavaging 4–6 hours post breeding does not appear to negatively affect pregnancy rates (Brinsko et al, 1990). Paccamonti and Crabtree (2019) describe a protocol for managing susceptible mares involving lavage 4–8 hours post breeding, coupled with uterine ecbolics such as oxytocin and prostaglandin analogues. Although work has suggested prostaglandin analogues can be used up to 12 hours post ovulation without a decline in plasma progesterone (Troedsson et al, 2001), the author tends to avoid its use post ovulation, as in practice it can be difficult to determine the exact time of ovulation.

Mares that develop large quantities of fluid around the time of breeding are likely to continue to show a moderate to marked oedema pattern within the uterus, and in this case it is important to consider the use of anti-inflammatory corticosteroids around this time. Known problem mares may be treated with prednisolone at a dose of 1 mg/kg orally SID started 24–72 hours prior to expected ovulation, and treatment may be continued for 24–72 hours post ovulation (Paccamonti and Crabtree, 2019).

It may be that oedema reduces normally prior to ovulation but then after breeding increases to maximal grade once more (Figure 4). In this instance a one-off treatment with dexamethasone (40 mg for a 500 kg horse) 6–12 hours post breeding may be used, although the individual's laminitis risk must be considered and discussed with the owner (Bucca et al, 2008; Paccamonti and Crabtree, 2019). Oxytocin should be used to improve uterine clearance, and may be administered as often as every 2 hours, starting 4–6 hours post breeding. This can be combined with exercise to aid fluid clearance, which is why a mare on box rest can be a particular challenge. As little as 10 minutes of walk work appears to aid clearance of fluid from the uterus, and in the author's experience this is appreciable on US scan. Sequential US examinations and repeated lavage should be performed as necessary, although they should ideally not be performed more than 72 hours after ovulation. Post-ovulation, progesterone levels begin to rise, taking the mare out of oestrus, ultimately closing the cervix ahead of the arrival of an embryo into the uterus at approximately 5.5 days. It is important to consider assisted reproductive techniques in mares with excessive fluid accumulation that fail to scan in foal at 14 days, as it is likely that an inhospitable environment within the uterus is contributing to early embryonic death, prior to detection of the pregnancy by US.

Some mares may have no evidence of pre-ovulatory fluid, yet after breeding have marked fluid accumulation. Although some of these mares may be suffering from breeding-induced endometritis, recent work suggests a proportion of mares suffer from latent Streptococcus spp. infections, which may become activated after breeding or intrauterine treatments (Petersen et al, 2015). Latent streptococcal infections may exist in the form of persister cells, deep in the endometrium or glands, whereby the bacteria have such a low metabolic rate they are not affected by appropriate antibiotic treatment, even when antibiotics achieve the minimum inhibitory concentration. A relatively new product (bActivate™, bActivate, Copenhagen) designed to reactivate latent streptococcal infections so that they may be treated, has shown some promising results (Peterson et al, 2015). It is a relatively expensive treatment, therefore cases should be chosen carefully.

Biofilms consisting of a heterogeneous community of different bacterial species, surrounded by an extracellular matrix, that coexist in a symbiotic relationship (Walker, 2008), are produced by some Gram-negative bacteria (notably Pseudomonas aeruginosa and Escherichia coli) and fungi. A biofilm can reduce the efficacy of intrauterine treatments by acting as a physical barrier, conferring antibiotic resistance. Mucolytics such as dimethylsulphoxide (DMSO) or acetylcysteine may be used in uterine lavage fluid to help break down the biofilm and render antibiotic treatment more effective (Barker, 2018). Gores-Lindholm et al (2009) concluded that infusion of mucolytic (N-acetylcysteine) agents pre-breeding resulted in higher than expected pregnancy rates.

Unexplained infertility

Perhaps the most frustrating situation is when everything appears to go well, yet the mare fails to get in foal after multiple attempts. First the mare's history must be thoroughly reviewed to ensure that adequate breeding management has been employed. A thorough examination of the reproductive tract, to include palpation and US, with manual and visual inspection of the vulva, vestibule and cervix, should be performed. Defects in the cervix may easily be missed if only a visual examination is performed; although uncommon, these can be hugely detrimental to a mare's fertility. Suitable laboratory samples should be taken to rule out subclinical endometritis, and an endometrial biopsy should be considered if not already taken. Hysteroscopy can be performed as part of the diagnostic work-up, as certain pathologies may not be diagnosed any other way, for example uterine adhesions.

If no abnormalities are found to explain the infertility, it can be a challenge to know what to do next. Anecdotally, supplementation with altrenogest has been used in mares that failed to conceive with no obvious cause; however, there is a distinct lack of evidence to support this as a therapy and this needs to be conveyed to owners, who often view it as a ‘golden solution’ to early pregnancy loss. It is possible to test the mare's progesterone level even in the face of altrenogest treatment, as altrenogest does not cross react on many of the progesterone assays (check with the laboratory being used).

Changing stallions may result in a pregnancy, and this should be attempted in the first instance if possible. However, if several cycles with reasonable quality semen, of known fertility, from more than one stallion, do not result in a pregnancy, it may be worth considering a ‘test breed’. This involves using semen as fresh as possible from a local health-tested stallion with known good fertility. When a test breed is performed, the mare may be flushed for an embryo, or left to be scanned for pregnancy at 14 days. If an embryo is recovered or the mare is scanned in foal then it is clear that she is able to produce a pregnancy, and the stallion should be changed or the breeding techniques previously employed must be fine-tuned. Some clients consider moving a mare to the country/state where the chosen stallion resides, to allow insemination with semen that is as fresh as possible.

If no embryo is recovered or no pregnancy is achieved after a test breed, then other less common problems should be considered. Blockage of the oviducts resulting in failure of fertilisation or embryo release is one such problem, and is a diagnosis of exclusion rather than a definitive diagnosis. Treatment involving flushing the oviducts with PGE2 (prostaglandin E2) gel by laparoscopy has resulted in a 93% conception rate in mares bred the same or the subsequent breeding seasons (Allen et al, 2006), and a 55% conception rate was reported in mares that had been barren for an average of 2 years prior to treatment (Martynski et al, 2015). Obviously case selection is paramount in these instances, as it is an invasive and expensive treatment; however, recent work has suggested application of a PGE1 (misoprostol) onto or around the oviductal papillae via a uterine catheter is also potentially effective (Alverenga and Segabinazzi, 2018), and this is clearly easier to perform in the field.

There are multiple other uncommon reasons why a mare may be unable to produce a pregnancy, such as a chromosomal abnormality, but they are outwith the scope of this article.

Conclusions

When careful analysis of a mare's breeding history is coupled with a thorough examination and appropriate diagnostics, it is usually possible to ascertain the most likely cause of infertility in difficult breeders. This enables an appropriate plan to be formulated, with the hope of achieving a pregnancy. However, even with the most thorough systematic management some mares fail to conceive, and in these instances test breeding can be considered, or assisted reproductive techniques employed, depending on the breed and budget available.

KEY POINTS

• Keep it simple and be systematic.
• Consider vulval conformation, examine the cervix both visually and digitally.
• Examine previous breeding records to ensure adequate reproductive management has been employed.
• Encourage open and realistic discussion regarding expectations with the mare owner from the offset.
• Perform diagnostics, low volume lavage, uterine biopsy and hysteroscopy where appropriate.
• Consider the potential for latent infections/biofilm causing unexplained infertility.
• Consider a test breed using good quality, fresh, local semen from a stallion known to be health tested and fertile.
• Do not overlook assisted reproductive techniques if breed allows.
• Remember you will not achieve pregnancies in 100% of mares!