High prevalence of mycoplasmas in the genital tract of asymptomatic stallions in Austria

Mycoplasma equigenitalium and M. subdolum have been implicated in genital disorders and infertility of horses. The reported cytopathic effects of M. equigenitalium observed in vitro underscore its potential pathogenic role in reproductive dysfunction in mares. This study was initiated to determine the prevalence of mycoplasmas in the genital tract of stallions in relationship to age, clinical signs, geographic location and semen quality. For this purpose the mycoplasma flora of the genital tract of 116 stallions of the Noric breed was determined by isolation and colony immunoblotting and by polymerase chain reaction (PCR) assays. Of 438 swabs from the genital tract, pre-ejaculatory fluid and semen samples, 352 (80%) samples were positive by PCR and 125 (29%) were positive by culture. Mycoplasmas were isolated predominantly from the fossa glandis and urethra and less frequently from the penis shaft and from semen. M. equigenitalium (89 isolates) and M. subdolum (70 isolates) were the predominant species identified. M. equirhinis and M. felis were detected in 27 and 8 samples, respectively. Comparison of these isolations with clinical signs, semen quality, and age of the stallions revealed no significant correlation. However, geographical location of the stallion significantly correlated with mycoplasma detection. These results suggest that mycoplasmas are present as commensals in the genital tract of stallions. Thus, clinically healthy stallions may present a permanent reservoir for infection of mares via venereal transmission.     

CASE STUDY: Use of Modified Phosphate-Buffered Saline as a Component of Semen Extenders Allows the Use of Transported Semen from Two Stallions

Conception rates for mares bred with transported-cooled and fresh stallion semen were collected over a 4-yr period (1998–2002) for two stallions. Both stallions stood at a commercial breeding farm. Semen from both stallions was used immediately after collection on the farm and after 24 to 48 h of cold storage when transported to locations in the U.S. and Canada. Semen for insemination of mares located on the farm was extended with a commercially available skim milk glucose extender (SKMG). Spermatozoal motility following cold storage for spermatozoa diluted in SKMG extender was unacceptable. Thus, semen from both stallions was centrifuged, and spermatozoa were resuspended in SKMG supplemented with modified PBS. In a previous study, the percentage of motile spermatozoa increased following centrifugation and reconstitution of the sperm pellet in SKMG-PBS as compared with semen dilution in SKMG (Stallion A: 15% vs 47%; Stallion B: 18% vs 43%). In the current study, 22 of 25 (88%) and 3 of 4 (75%) mares conceived with transported-cooled semen from Stallions A and B, respectively. Conception rates for mares inseminated with transported semen did not differ (P>0.05) from those inseminated on the farm with fresh semen. These data illustrate that stallion owners can modify standard cooled semen processing procedures and semen extender composition to improve post-storage spermatozoa motility and to obtain acceptable fertility.     

A study of Klebsiella pneumoniae infection in the uterus of the mare.

Two experiments incorporating 13 mares were conducted for the purpose of producing and monitoring intrauterine infection with Klebsiella pneumoniae. In the pilot study, the infection was produced with strains of K pneumoniae type 68 and type 10 isolated from the genital tract of stallions with a history of breeding problems. In the principal study, K pneumoniae type 68 was used to produce the infection. Tampons and guarded culture swabs were used to obtain uterine samples in the pilot study. In comparing the efficacies of isolation of K pneumoniae with the tampons and isolation with standard guarded culture swab, the tampon proved to be a more reliable means with which to isolate K pneumoniae and was used in the principal study. In both studies, inoculated mares became infected and remained infected at least until the postinoculation estrous cycle was initiated or was completed. Some of the inoculated mares remained infected through more than one estrous cycle. The numbers of K pneumoniae decreased in the uterus of mares after completing the estrous cycle after inoculation. Klebsiella pneumoniae was not demonstrable in frozen tissue sections of uterine biopsy specimens stained by fluorescent antibody technique. Postinoculation sera antibody titers to K pneumoniae, as determined, using the capsule swelling technique, were no higher than 1:8.     

The occurrence and significance of plasma coagulase negative staphylococci from the genital tract of horses

Classification based on biochemical characteristics of 389 strains of plasma-coagulase-negative (plc-) staphylococci isolated from the genital tract of mares and stallions resulted in the following distribution of species: St. sciuri 130 (33.4%), St. equorum 42 (10.8%), St. xylosus 16 (4.1%), St. epidermidis 35 (9.0%), St. simulans 24 (6.2%), St. haemolyticus 33 (8.5%), St. warneri 18 (4.6%), St. lentus 12 (3.1%), St. hyicus 11 (2.8%). Strains of St. cohnii, St. capitis, St. gallinarum, St. saprophyticus and St. hominis have only been found sporadically (a. 1%). 48 (12.3%) strains could not be classified. With regard to species distribution of isolates from stallions and mares. 63.7% of the isolates from stallions belonged to St. sciuri and 9.3% to St. lentus, whereas in isolates of mares these species numbered only 24.9% and 0.4%, respectively. On the other hand the species St. equorum (14.9% vs. 6.8%), St. epidermidis (14.5% vs. 1.7%), St. haemolyticus (14.0% vs. 1.7%), St. warneri (7.7% vs. 0.8%) and St. xylosus (5.9% vs. 2.5%) predominated in mares. St. simulans was found occurring equally in mares and stallions (7.7% vs. 5.9%). Comparing the staphylococcal species of healthy mares and of mares which have not become pregnant after copulation no indication was found for a significant role of certain plc- staphylococci in infertility. All of the 389 isolates were tested for production of protein A, i.e. Fc-fragment binding receptors, using a microenzyme-assay with peroxidase-labelled rabbit immunoglobulin G. With this method cell-bound or extracellular Fc-receptors could not be detected in anyone of the plc- staphylococcal strains.(ABSTRACT TRUNCATED AT 250 WORDS)     

Transmissibility and abortogenic effect of equine viral arteritis in mares

A group of 14 pregnant mares was exposed via contact to 4 mares bred to stallions infected with equine viral arteritis virus. There was a demonstrable febrile response in each donor mare and in 12 of the pregnant mares. All 18 mares became seropositive after exposure. Equine viral arteritis virus was isolated from the nasopharynx of 5 pregnant mares, but not from the donor mares. Ten of the pregnant mares aborted, and virus was isolated from fetal specimens or placenta of 8.     

Serological surveillance of equine viral arteritis in the United Kingdom since the outbreak in 1993

Serological analysis of blood samples submitted to the Animal Health Trust showed that during 1995, 185 of 9203 unvaccinated horses (2.0 per cent) tested positive for antibodies to equine arteritis virus (EAV), and that during 1996, 46 of 8851 unvaccinated horses (0.52 per cent) tested positive. During both years thoroughbreds were the predominant breed tested and only a small proportion of these (<0.3 per cent), consisting predominantly of imported mares, were seropositive. In contrast, among standardbred horses, from which samples were actively solicited in 1995, 84 of 454 (18.5 per cent) were seropositive. Among standardbreds there was a difference in prevalence between types of horses, with 3.7 per cent of racing horses, 25 per cent of non-racing horses and 41 per cent of stallions testing seropositive. Investigations of seropositive stallions identified during 1994 and 1995 demonstrated that clinically inapparent equine viral arteritis (EVA) had occurred previously in the UK. Of 50 seropositive stallions, nearly half were standardbreds and nearly all had been imported from either North America or the European Union. Whether 34 seropositive stallions were shedding virus in their semen was established either by test mating, by the serology of the covered mares, or by investigation by MAFF following the introduction of the Equine Viral Arteritis Order 1995. Nine of the stallions (26.5 per cent) were identified as presumptive shedders of EAV in semen and among specific breeds, viral shedding was identified in six of 15 (40 per cent) standardbreds and three of nine (33 per cent) warmbloods. In contrast with the outbreak of EVA in the UK in 1993, no signs of disease typical of EAV infection were reported during these investigations, even in mares test mated to stallions shedding the virus.     

Cryopreservation of Stallion Spermatozoa with INRA96 and Glycerol

The aim of the present study was to improve success of cryopreservation of stallion spermatozoa. Semen from eleven stallions was collected and frozen in INRA 96 with two different concentrations of glycerol (3.5% and 6.0%) and compared with a control freezing process. The mean post-thaw motility for the eleven stallions of 57.93% (3.5% glycerol) and 66.50% (6.0% glycerol), which was statistically higher (P < 0.05) when compared with the mean post-thaw motility (39.7%) for semen in a control egg-yolk extender (Equipro^® CryoGuard™ Complete, Minitube). The Equipro^® CryoGuard™ Complete is a commercial semen freezing protocol that has been one of the standard processes used in our laboratory for freezing equine spermatozoa. INRA 96 with 6% added glycerol was used in the fertility trial as it provided the highest spermatozoa survival. To evaluate fertility of the frozen semen, eight mares were bred over two cycles with both fresh and frozen semen. The pregnancy rate of mares bred with frozen semen (55.6%) was not statistically different (P > 0.05) from the pregnancy rate of mares bred with fresh semen (55.6%). INRA 96 with 6.0% glycerol improved the survivability of stallion spermatozoa through the cryopreservation process, and subsequent fertility was not different (P > 0.05) from fresh, extended semen.     

Infection patterns in pony mares challenged with the agent of contagious equine metritis 1977

Contagious equine metritis 1977 was reproduced in pony mares using cultures of the Gram-negative coccobacillus aetiologically associated with the disease. Variability in clinical response was observed in the first of 2 experiments, with the presence of semen, either alone or in an extender, appearing to potentiate the pathogenicity of the challenge strain of the organism. The experimental disease was characterised by a variable degree of vaginal discharge and concomitant inflammatory changes involving the vervix and vagina. Although all of the affected mares recovered spontaneously, a high percentage continued to harbour the Gram-negative coccobacillus in their genital tracts for variable periods after challenge. Shedding of the organism was either relatively constant or intermittent and was not solely related to the oestrous period. Cytological examination of smears of cervical and urethral swabs was of diagnostic value only during the clinical phase of the infection. There was evidence that reinfection of mares could occur after an interval of 2 weeks.     

The effect of liposomes composed of phosphatidylserine and cholesterol on fertility rates using frozen thawed equine spermatozoa

The objective of this study was to determine if the addition of liposomes composed of phosphatidylserine (PS) and cholesterol (CH) to equine sperm would improve pregnancy rates after sperm were cryopreserved. Ejaculates from four stallions, collected every other day during May and June were split, treated with PSCH liposomes or HBS (Hepes Buffered Saline) and cryopreserved. Fifty-two mares were bred over eighty estrous cycles with the frozen-thawed semen. The one cycle pregnancy rates of the mares inseminated with semen treated with liposomes (45%) were similar to mares inseminated with control semen (48%; p>0.05). Thus, at least at the levels used, PSCH liposomes added to equine sperm did not improve pregnancy rates of mares inseminated with cryopreserved semen.     

Demographic analysis of the Canadian Standardbred broodmare herd

This study derives demographic parameters for the Canadian Standardbred broodmare herd by analysis of industry mating reports for the years 1986–1989 inclusive. The reference population for the study was all mares bred to Canadian-registered stallions. Number of matings evaluated ranged from 9169 to 9837 in each year. In establishing the size of the national herd, data from registration files for 1986 also were used to estimate the number of Canadian mares bred to US stallions. The national broodmare herd was estimated to consist of 12 237 mares, of which 7.7% annually were bred to US-registered stallions. An estimated 16.7% of the national herd were inactive (not presented for breeding) in any year. Probability of temporary absence from the breeding herd was influenced by age, decreasing from 2 to 6 years, then increasing thereafter. Age-specific culling rates revealed that mares less than 6 years of age were subject to more intense selection pressure than was evident for mares more than 10 years of age. Inconsistent presentation for breeding appeared to be a feature of the management of young broodmares. Overall herd culling rate was estimated to be 10.5%, equivalent to an average breeding life of 9.5 years. It was determined that the average broodmare breeds for 7.9, or 83%, of available years. The results clearly indicate that the reproductive advantage enjoyed by mares 6 years of age results, in part, from more consistent presentation for breeding.     

Bacterial Flora of Semen Collected from Danish Warmblood Stallions by Artificial Vagina

Semen samples were collected from 21 Danish Warmblood stallions by the Colorado artificial vagina (Colorado AV, 14 samples) or by the Missouri artificial vagina (Missouri AV, 7 samples). The semen was examined bacteriologically by direct plating (DP) on blood agar plates, and by plating of semen swabs stored in Stuart’s transport media (TM) at 4°C for 1–4 days. No significant differences were observed between results obtained by DP and cultures of identical TM samples. Of the 21 samples examined, only 1 TM (4.8%) and 2 DP samples (9.5%) were sterile, while the rest yielded a predominantly mixed flora comprising 1 to 4 bacterial genera. The natural flora was dominated by coagulase-negative staphylococci (Staphylococcus lentus, S. capitis, S. haemolyticus, S. xylosus) (16/21 = 76%), coryneforms (11/21 = 52%) and alpha-hemolytic streptococci and lactobacilli (7/21 = 33%). Potential venereal pathogens were isolated from 7 stallions (33%). Beta-hemolytic streptococci were found in 4 stallions used for natural service, whereas Pseudomonas aeruginosa serotype 6 (2 samples ) and Klebsiella pneumoniae subsp. pneumoniae capsule type K5 (1 sample) were isolated from 3 stallions used exclusively for artificial insemination. The role of the stallion as a carrier of potential venereal pathogens, and the artificial vagina as a source of contamination, is discussed in the context of mare endometritis.     

Assessing the fertility of two mares cloned from the same founder animal

Two cloned mares, produced from the same sample of skin fibroblasts, were bred during four breeding seasons from their second year of age, as embryo donors, in exactly the same conditions, using the same stallions for both cloned mares. The aim of this study was to test the embryo donor potential of cloned mares and to compare the results obtained from two cloned mares of the same mare with other embryo donor mares (n = 31–39 per breeding season) at the same stud. For both cloned mares, 19 embryos were recovered by 43 collection attempts (44%) (7/22 for one; 12/21 for the other), 16 (84%) pregnancies (5/7 for one, 11/12 for the other) were obtained at day 14 post-ovulation (D₁₄), and 12 (3/7 for one; 9/12 for the other) foals were born. One cloned mare was a less efficient donor mare than the other (p < .05), In control donor mares, 623 embryo collections were performed, with a recovery rate (80%—496/623) significantly higher than for cloned mares. The recovery rate in the subpopulation of 2–5-year-old control donor mares (same age of cloned mares) (89%—127/143) and The recovery rate in the subpopulation of 12 control mares bred with the seven same stallions as clones (55%—17/31), were both higher than for cloned mare (p < .05). The success rate of transfer was not different between embryos produced by cloned mares (84%—16/19) and those produced by control donor mares (79%—392/496). However, the foaling rate per embryo collection was significantly lower for cloned mares (28%—12/43) than for control donor mares (52% - 325/623) (p < .05).     

Evaluation of the field application of PCR in the eradication of contagious equine metritis from Japan

The effectiveness of the polymerase chain reaction (PCR) as a field application test for the eradication of contagious equine metritis (CEM) was evaluated. Seven-thousands five-hundred and thirty-four genital swabs were collected from 4,026 Thoroughbred broodmares and stallions in Japan to test "high risk" horses as well as for general surveillance testing from 1998 to 2001. Bacterial isolation as well as PCR testing of original specimens and cultured specimens was performed for detection of Taylorella equigenitalis from genital swabs. As a result, T. equigenitalis was detected in 12 mares and 1 stallion by PCR, although the bacteria were isolated from only 2 of the PCR-positive mares. CEM-infected and carrier horses were treated by a combination of chemotherapy and surgery. Subsequent follow-up testing over a 3-year period did not detect T. equigenitalis. It was demonstrated that PCR testing was more sensitive than isolation as a method for the detection of T. equigenitalis from genital swabs of horses in the field. It was therefore suggested that a combination of PCR testing and treatment were useful measures in the eradication of CEM from Japan.     

An investigation into the suitability of a commercial real‐time PCR assay to screen for Taylorella equigenitalis in routine prebreeding equine genital swabs

Reasons for performing study: Standard bacteriological methods for identifying Taylorella equigenitalis in cervical smears are time consuming. Therefore, a more rapid real‐time PCR assay was evaluated for its suitability in screening swabs. Objective: To compare the results of a commercially available real‐time PCR assay with routine microbiological culture for the identification of T. equigenitalis, the causative organism of contagious equine metritis, in equine genital swab samples, under ‘field trial’ conditions. Materials and methods: Routine prebreeding genital swabs (n = 2072) collected from Thoroughbred mares and stallions during 2009 were examined together with stored T. equigenitalis positive material. Swabs were cultured for T. equigenitalis using standard microbiological techniques. Bacterial lysates were isolated from the swabs and examined for the presence of a 16S DNA fragment of T. equigenitalis, using a commercial multiplex real‐time PCR assay system. Results: There was complete concordance between positive and negative results obtained by the 2 methods. Real‐time PCR also detected T. equigenitalis DNA from swabs that were negative using standard microbiological culture after 6 months' storage at +4°C but from which T. equigenitalis had been isolated following collection. The sensitivities of realtime PCR and bacterial culture were both 10^?3 (equivalent to 3 colony‐forming units). Conclusion and clinical relevance: Routine bacterial culture of T. equigenitalis requires an incubation period of not less than 7 days before a conclusive negative result can be obtained, whereas bacterial extraction and real‐time PCR assay can be completed in less than 6 h. The commercially‐available PCR assay tested provided a rapid and reliable method for the identification of T. equigenitalis from equine genital swabs and could be usefully employed for the screening of mares and stallions for preseason Horserace Betting Levy Board (HBLB) Code of Practice and in other situations such as for bloodstock sales screening requirements, overcoming the current delays imposed by bacterial culture requirements. Its use could be quality assured by the existing HBLB biannual testing scheme for designated laboratories.     

Vaccination of stallions with a modified live equine viralarteritis virus

Sixteen stallions were vaccinated with modified live virus- EVA vaccine. Paired stallions were assigned a group of four susceptible mares to breedtwice by natural service between 10 and 25 days postvaccination. No significant adverse changes in hematologic parameters or temperatures that could be related to vaccination were observed. Vaccinated stallions showed no overt signsof illness and attempts to isolate the virus from buffy coat cells and semen were unsuccessful. Susceptible mares bred to these stallions between 10 and 25 days postvaccination failed to seroconvert     

Conception rate,uterine infection and embryo quality after artificial insemination and natural breeding with a stallion carrier of Pseudomonas aeruginosa: a case report

Background

Pseudomonas aeruginosa may cause venereal disease and infertility in horses. A Pseudomonas aeruginosa - carrier stallion, often unresponsive to artificial vagina collection, was used to naturally breed mares. Semen collected from the same stallion was also used to perform artificial inseminations. Pregnancy rates, embryo quality and incidence of uterine infection were compared between inseminated or naturally-bred mares.

Methods

P. aeruginosa was isolated from swabbing of the penis, prepuce and distal urethra of the stallion. Before being bred or inseminated, clitoral/vestibular samples were collected from all mares, and cultured for isolation of P. aeruginosa. At the first observed estrus, endometrial swabs were also collected. All mares subjected to natural mating (NS) were re-evaluated for P.aeruginosa by culture of clitoral and endometrial swabs. Artificial inseminations (AI) were performed either with fresh-extended semen (11 AI/7 mares) or frozen semen (10 AI/7 mares). The stallion was also used to breed 3 mares (4 services). For embryo collection, 2 mares were inseminated with fresh-extended semen (1 AI/mare), and 2 additional mares were inseminated with frozen semen (2 AI/mare). Two mares were naturally-bred with a total of 9 services, for embryo collection. All mares were examined after AI or natural service (NS), for uterine pathologies. Embryo recoveries were attempted passing a catheter with inflatable cuff connected to a sterile flexible 2-way flushing catheter, through the cervix. Flushed media was recovered into an Em-Con filter, and embryos searched using a stereoscope. Embryos were graded from 1 (excellent) to 4 (degenerated/dead).

Results

Pregnancy rates obtained after NS was 50% per cycle. However, more than half of the NS resulted in uterine disease, while uterine pathology was seen only in 22% of the time following AI. Half of the mares bred by NS got positive to P. aeruginosa. Percentage of embryo recovery rates was identical after AI or NS (66.7%). The 4 embryos recovered after AI were classified as Grade 1, while after NS only 2 out of the 6 recovered embryos were Grade 1.

Conclusion

a) there was no evidence of reduced fertilization after AI or NS, b) a numerically higher incidence of uterine disease was noticed after NS, c) venereal transmission of P. aeruginosa after NS was confirmed, d) a lower percentage of G1 embryos may be obtained after NS. Overall, the data supports the indication for P. aeruginosa-carrier stallions to be bred by AI rather than by NS, and raises the possibility that P. aeruginosa may affect embryo quality.     

Male, female and management risk factors for non-return to service in Dutch mares

The "effect" of stallion, mare and management-related factors on the odds of pregnancy per cycle in the horse were identified and quantified from the breeding records of Dutch Warmblood (n=4491), Friesian (n=1467) and Shetland-pony mares (n=3267) mated either naturally or by artificial insemination to one of the 88 stallions between 1992 and 1996. A mare was considered to be pregnant when she did not return to oestrous within 28 days of the last insemination. For Dutch Warmblood horses, the percentage of mares that did not return for service within 28 days (NR28) varied between studfarms and ranged from 61 to 82%. The NR28 for mares inseminated with fresh semen ranged from 67 to 74% and for mares inseminated with frozen/thawed semen this percentage was 59. Mares served at a second cycle had lower odds not to return than mares served at the third or subsequent cycle (OR=0.84). For Friesian horses, the NR28 for young mares was higher than that for older mares. Mares served before 1 May in any year had lower odds of non-return than mares served after 1 July (OR=0.69). The NR28 of mares inseminated once per cycle was 6% lower than that of mares inseminated three times or more per cycle. For Shetland ponies, the NR28 also varied between studfarms and ranged from 62 to 78%. Stallions < or =3 years old had lower odds of non-return compared to older stallion (> or =11) (OR=0.57). Mares served before 1 July had lower odds of non-return. Other significant factors for this breed were age of the mare, cycle number and insemination frequency. Stallion factors accounted for 5.9, 2.0 and 14.7% of the variation in the NR28 for Dutch Warmblood, Friesian horses and the Shetland ponies, respectively.     

Epizotiology and phylogeny of equine arteritis virus in hucul horses

The aim of the study was to determine the situation of equine arteritis virus (EAV) infections in hucul horses. A total of 176 horses (154 mares and 22 stallions) from the biggest hucul horse stud in Poland were tested. Antibodies against EAV were detected in 97 (55.1%) horses. The EAV seroprevalence among mares was 53.2% while in stallions - 68.2%. The percentage of positive mares increased with their age, thus amongst the mares of less than 2 years of age the percentage was 32.5%, while in the group of 3-5 years old increased to 59.4% and in the mares in the age of 6-10 years and older than 10 years 89.5% and 95% were seropositive, respectively. Among 11 seropositive stallions five were supposed to be shedders of EAV with their semen. It is likely that those persistently infected stallions were the reservoirs of the virus in the stud. Genetic studies using of ORF5 gene showed high homology between the viruses detected in the semen of those stallions what suggested lateral transmission between the stallions sharing the same stable. Persistent infection in an immature stallion, which has not yet been used for breeding, was established as a result of infection via respiratory route. Phylogenetic analysis confirmed that all hucul viruses shared the same ancestor and as most of EAV strains dominating in Polish horse population belonged to the European origin EAV subgroup (EU-1).     

Quantitative assessment of the risks of reducing the routine swabbing requirements for the detection of Taylorella equigenitalis

The transmission of contagious equine metritis (CEM) on stud farms in Britain, Ireland and other European countries is prevented by following the recommendations in the Horserace Betting Levy Board's Code of Practice on CEM. A quantitative risk assessment was undertaken to estimate the likely impact of removing the recommendation, from the 2002 code, to culture endometrial or cervical swabs microaerophilically for the presence of Taylorella equigenitalis, the causative organism. The scientific literature was reviewed for evidence about the anatomical distribution of T. equigenitalis at different times after infection and it was found that, in chronically infected mares, the organism was detectable in the clitoral swabs of nearly 93 per cent, but in the cervical swabs of only 31 per cent. In contrast, in acutely infected mares, the organism was detectable in the clitoral swabs of nearly 69 per cent, but in the cervical swabs of 84 per cent. By using these results, a quantitative risk assessment was undertaken, assessing the likely effects of removing the recommendation that swabs from the cervix of low-risk mares should be cultured for T. equigenitalis. The results were sensitive to the prevalence of the infection, but when it was low, there appeared to be few benefits in continuing to culture cervical swabs routinely. However, such swabs are vital when the disease is suspected.     

Progesterone and equine chorionic gonadotropin concentrations around the time of pregnancy loss in mares impregnated by donkeys or stallions

The objective of this study was to compare the incidence of pregnancy loss of mares carrying a mule embryo with that of mares carrying a horse embryo. The possible causes of such mortality were evaluated through serial ultrasonographic evaluations and hormonal monitoring, paying special attention to the role of premature regression of the endometrial cups and its relation to inadequate luteal function.

Twenty-eight mares impregnated by stallions and 19 mares impregnated with donkey semen were evaluated ultrasonographically every week from day 20 to day 150 of pregnancy. The viability of the product was assessed each time, and the diameter of the embryonic vesicle was measured from day 25 to day 60. Blood samples for progesterone and equine chorionic gonadotropin (eCG) determination were taken every week.

Both progesterone and eCG concentrations during normal pregnancies wegre significantly lower in the mares inseminated with donkey semen than in the mares impregnated by stallions (P<.05). In 7 of the mares carrying mule conceptuses to term, the concentrations of eCG remained basal throughout the study. In the other animals from this group, the levels of this hormone did increase but returned to baseline much earlier (on day 77 of pregnancy) than in the mares served by stallions (on day 126 day of pregnancy). There was no significant difference between the growth rate of embryonic vesicles of mares carrying mule embryos and that of mares carrying horse embryos (P>.05).

The incidence of pregnancy loss was significantly higher (P<.05) in mares carrying a mule embryo (36.8 %) than in mares carrying a horse embryo (21.4%); it occurred on average on day 93 of pregnancy in mares carrying mule embryos and on day 43 on mares carrying horse embryos. There was only 1 case in which pregnancy loss was associated with concentrations of both eCG and progesterone that were much lower than the average for the normal pregnancies of the same group, and this was in a mare carrying a horse embryo. The most frequent cause of pregnancy loss was premature luteal regression due to primary luteolysis, as evaluated via peripheral progesterone concentrations. This occurred in 2 mares carrying horse embryos and in 4 mares carrying mule embryos. Three mares carrying mule embryos and 1 carrying a horse embryo had abortions that were not preceded or accompanied by any alteration in progesterone or eCG levels and were thus classified as fetal deaths of non-endocrine origin.

It is concluded that the incidence of pregnancy loss is higher in mares carrying a mule embryo than in mares carrying a horse embryo. However, this is not due to the low progesterone concentrations associated with the premature regression of the endometrial cups that occurs in mares with interspecific pregnancy.