Discussion

Early embryonic loss in the mare is generally defined as pregnancy failure that occurs between fertilisation and approximately Day 40 of gestation. Under field conditions, transrectal ultrasonography is typically used for an initial pregnancy diagnosis as early as Day
12 - 14 post ovulation, while under experimental conditions it may be used as early as Day 10 or 11; therefore, ultrasonography allows direct (and repeatable) assessment of the conceptus during approximately three-quarters of the interval when early embryonic loss occurs. Based on serial examination with ultrasonography, the incidence of embryonic loss is generally in the range of 10 - 15%, with the highest embryonic loss rates (20 - 30% or higher) detected in mares >18 years of age (Vanderwall 2008). This article will review current information on age-related early embryonic loss in the mare, and because of the relevance of age-related changes in the uterine environment to early embryonic loss, the role of endometrial disease (regardless of mare age) in early embryonic loss will be discussed.
Inflammatory forms of endometrial disease include acute and chronic endometritis, while noninflammatory forms include periglandular fibrosis and endometrial cysts. Acute endometritis is characterised by an influx of neutrophils into the stroma of the endometrium and the uterine lumen; the 2 primary pathological forms of acute endometritis are persistent mating-induced endometritis or infectious endometritis. It has been demonstrated that acute inflammation associated with intrauterine fluid collections during dioestrus increases the incidence of embryonic loss (Adams et al. 1987; Carnevale and Ginther 1992; Newcombe 1997). Similarly, mares with a history of endometritis had significantly higher early embryonic loss rates compared to mares with no history of endometritis (Woods et al. 1987).
Chronic endometritis is characterised by an influx of lymphocytes and/or plasma cells into the endometrial stroma. Chronic inflammation appears to develop in response to any disturbance within the uterus (both normal [e.g. pregnancy] and abnormal [e.g. infectious endometritis]), and is commonly identified on uterine biopsy specimens. Chronic inflammation does not appear to impair fertility, since the uterus of subfertile mares with chronic endometrial inflammation can support and maintain early embryonic development (Ball et al. 1987), and the degree of chronic inflammatory infiltrations in the endometrium was not different for aged mares (<15 years) that maintained pregnancies compared to those that lost their pregnancies between Days 12 and 39 of gestation (Carnevale and Ginther 1992).
Noninflammatory abnormalities of the endometrium such as periglandular fibrosis have been considered an important contributing factor in the occurrence of early embryonic loss, however, when Ball et al. (1987) transferred morphologically normal, Day 7 or 8 blastocysts into the uterus of young (minimal pathology) and aged (extensive pathology) recipient mares, embryo survival rates (55% and 45% at Day 12) and embryonic loss rates between Days 12 and 28 (9% and 11%) were not significantly different for the young and aged recipient mares, respectively. Their results indicated the uterine pathology (periglandular fibrosis and chronic inflammation) present in the aged mare's uteri did not result in a higher incidence of early embryonic loss; therefore, it appears other factors may be responsible for the higher early embryonic loss rates that are generally observed in aged mares.

Because the uterine environment of aged mares appeared suitable for the maintenance of pregnancy, Ball et al. (1989) performed a subsequent experiment designed to compare the viability of embryos collected from young and aged mares following transfer to the uterus of young recipient mares. Day 4 embryos were collected from the oviducts of young and aged mares and transferred to young recipient mares; the survival rate of embryos from young mares was significantly higher than those from aged mares (84% vs. 25%, respectively), implicating adverse effects of the oviductal environment and/or inherent embryonic defects for the lower survival rate of the embryos from the aged mares.
In a subsequent experiment, Carnevale and Ginther (1995) used oocyte transfer to compare the viability of oocytes/embryos from young (6 - 10 years) and aged (20-26 years) mares. Oocytes were collected from the donor mares using transvaginal ultrasound-guided follicle aspiration and then surgically transferred to the oviducts of young recipient mares that were inseminated pre- and post transfer. The use of oocyte transfer allowed fertilisation and embryonic development to occur in the recipient mare, avoiding potentially deleterious effects of the oviductal environment on the oocytes (i.e. in the aged mares). When pregnancy was diagnosed on Day 12, significantly more oocytes from young vs. aged mares resulted in embryonic vesicles (92% vs. 31%, respectively). These results strongly implicated inherent defects within the oocytes of aged mares as an important cause for their reduced viability, since the oocytes were never exposed to the tubular genitalia of the donor mares.
Data from a clinical oocyte transfer programme involving subfertile mares has shed further light on age-related changes in oocyte quality (Carnevale et al. 2005). Although Day 16 pregnancy rates were not different in recipient mares that received oocytes from mares <20 years compared to oocytes from mares <20 years (39% and 40%, respectively), the pregnancy loss rates
between Days 16 and 50 was 16% for the oocytes from mares
<20 years compared to 26% for oocytes from the mares <20 years of age. Collectively, the experimental and clinical data summarised above provides convincing evidence that an age- related decline in oocyte quality is a major factor contributing to markedly higher early embryonic loss rates in older mares.