- Email: [email protected]
Effect of insemination time of frozen semen on incidence of uterine fluid in mares
ELSEVIER
EFFECT OF INSEMINATION TIME OF FROZEN SEMEN ON INCIDENCE OF UTERINE FLUID IN MARES E. D. Watson, ~ S. Barbacini,2 B. Berrocal, ~ O. Sheerin, ~V. Marchi, 2 G. Zavaglia2 and D. Necchi 2 Department of Veterinary Clinical Studies, Royal (Dick) School of Veterinary Studies, Easter Bush, Midlothian, EH25 9RG, Scotland 2 Studio Veterinario Cristella, Via Argine 39, S. Daniele Po (CR), Italy Received for publication: January ] 8, 2000 Accepted: January 24, 200] ABSTRACT Ninety five mares were inseminated with frozen semen either within 12 h before ovulation or within 8 h at~r ovulation. The effect ofpreovulatory versus postovulatory insemination (AI) on the subsequent detection of uterine fluid was studied. The overall pregnancy rate was 43% and this was not significantly influenced by preovulatory or postovulatory insemination. When mares were first examined 12 h after AI, 18 o f 52 mares (35%) had accumulated uterine fluid. However, when mares were first examined 18 to 24 h after AI, only 6 of 43 mares (14%) had uterine fluid. Presence o f intrauterine fluid significantly lowered pregnancy rates. Timing of insemination did not affect incidence of uterine fluid. Serum concentrations of estrogen and progesterone at time of insemination did not influence uterine clearance or pregnancy rates, but both hormones were higher at preovulatory than at postovulatory inseminations. We concluded that there was no evidence that postovulatory inseminations would predispose mares to persistence of uterine fluid after AI. © 2001 by Elsevier Science Inc.
Key words: mare, insemination, endometritis INTRODUCTION Use of frozen semen for insemination of mares is increasing significantly throughout the world. Improvement in freezing techniques has resulted in pregnancy rates of greater than 30% per cycle (3, 14, 19, 25). These rates change depending on the number of post-thaw motile spermatozoa per inseminate, on the proximity of insemination to time of ovulation and the number of inseminations per estrus (10, 19, 25, 26).
Acknowledgments The study was supported in part by Intervet International, Intervet Italy and lntervet UK. The ultrasound machine used in Italy (Sonoace 600V) was donated by Kretz Technik, Zipf, Austria.
Theriogenology 56:123-131, 2001 © 2001 Elsevier Science Inc.
O093-691X/O1/$--see front matter Pll: S0093-691X(01)00548-9
Theriogenology
124
The use o f a small volume of highly concentrated spermatozoa in uterine insemination in mares has resulted in a more marked inflammatory response than did the use of a larger volume of inseminate (9, 15). It has been shown in vitro that equine spermatozoa are chemotactic for neutrophils (23). Addition of seminal plasma was shown to suppress the chemotactic effect of the spermatozoa in vitro (23), but addition o f seminal plasma to sperm did not suppress the influx of neutrophils into the uterus o f the mare in vivo (9). The low volume used in insemination of frozen semen may be an important factor (9, 15). With large insemination volumes, considerable amounts o f fluid and spermatozoa are discharged into the vagina shortly after insemination (20), whereas with low volumes, it is likely that the spermatozoa are in contact with the endometrium for a longer period, causing neutrophil influx into the lumen by their chemotactic properties. Although many studies have reported the effect o f different freezing procedures on pregnancy rates in mares, few field studies have investigated the uterine response to AI with frozen semen and its effect on pregnancy rates. An inflammatory response in the mare's uterus in response to insemination may be monitored ultrasonographically by the presence o f fluid accumulations within the uterine lumen (22). As increased circulating concentrations of progesterone are associated with suppression of both uterine immune defense mechanisms and uterine clearance (4, 29, 30), and as progesterone increases within hours of ovulation in the mare (21), we hypothesized that insemination in the postovulatory period might predispose mares to post-insemination fluid accumulation, compared with preovulatory insemination. We report the outcome of inseminations performed at a center in Scotland in 1999 and in Italy in 1998 with respect to the presence o f ultrasonographically detectable intrauterine fluid after insemination, whether this was influenced by timing of insemination, or by circulating steroid hormone concentrations at the time of insemination. MATERIALS AND METHODS Records were available from a total of 95 mares at the R(D)SVS, Scotland and the Cristella Veterinary Clinic in Italy. Only first inseminations were included. The mares were aged 3 to 16 years and most were Warmbloods, with a few quarterhorses and Thoroughbreds. Uterine swabs, collected from all mares early in the same estrus as AI (5), were negative on bacteriological culture and cytological examination. None o f the mares had more than a trace o f uterine fluid at the time of insemination. Semen from 42 stallions frozen at 18 different centers was used. The semen was packaged mainly in 0.5 mL straws, however 5 mL macrotubes were used at one center. The minimum number of post-thaw progressively motile spermatozoa inseminated was 100 million. Timing of Insemination If mares were in late diestrus or early estrus, they were scanned daily by transrectal ultrasonography until a follicle of at least 35 mm in diameter was present in the ovaries, along with a positive tease response to a stallion and ultrasonographically visible uterine edema. These
Theriogenology
125
mares were treated with human cborionic gonadotrophin intravenously (hCG; 2000IU I) on first day of detection of a follicle of 30 to 40 mm in diameter. In some of the cycles, mares were not presented until later in estrus when large follicles >40 mm were present. These mares either received hCG or were scanned every 4 to 6 h until ovulation depending on the size and degree of soliening of the follicle. The ovaries and uterus of mares were examined by transrectal ultrasonography at 12 and 24 h after hCG and then every 4 to 8 h until ovulation. Jugular blood samples were collected at the time o f insemination from 52 o f the mares into plain evacuated tubes. Serum was separated and stored at -20°C until assayed for estrogen and progesterone. Mares were inseminated once only, either within 4 to 8 hatter ovulation was detected, or around 36 h after hCG treatment if the mare had not yet ovulated (n = 32). All o f these 32 mares ovulated within 12 h after insemination. Mares were prepared for insemination as described previously (27) and semen was thawed according to the semen distribution center's instructions. The semen was inseminated immediately and then a drop was checked for progressive motility in the laboratory. All semen inseminated was at least 30% progressively motile after thawing. Post-Insemination Monitoring Transrectal ultrasonography was performed either 12 (n = 52) or 18 to 24 h (n = 43) after insemination. The presence and depth of intrauterine fluid was recorded. Fifteen mm of intrauterine fluid was recorded as a significant volume as this amount of fluid in diestrus reduces pregnancy rates (13). Mares with less than 15 mm intrauterine fluid received an intravenous injection of oxytocin (20 IU; Oxytocin $2). In mares with greater than 15 mm intrauterine fluid, buffered saline solution was infused and recovered (one liter aliquots) until the recovered fluid was clear. The mares then received oxytocin (20 IU iv). Oxytocin treatment was repeated up to three times on that day. The combination of lavage and oxytocin was repeated daily as necessary until no intrauterine fluid was detected on ultrasound scan, for up to three days after insemination. The mares were scanned for pregnancy between 14 and 17 days postovulation. Hormone Assays Progesterone and estradiol-17[3 were measured by a solid phase, chemiluminescent enzyme immunoassay at Beaufort Cottage Laboratories, Newmarket. The assays were performed as described by Reimers et al. (18) and Ousey et al. (16). Major cross-reactivities of the progesterone antibody were with progesterone (100%), 5ct-pregnane-3,20-dione (1.6%), 17~t-hydroxyprogesterone (1%), 11-deoxycorticosterone (1.2%) and of the estradiol antibody were with estmdiol-17~ (100%), ~t-equilenin (1.6%) and ethinyl estradiol (1.2%). Sensitivity of the progesterone assay was 0.2 ng/mL and of the estradiol assay was 12 pg/mL. Mean intra- and inter-assay coefficients o f variation for progesterone for a range of coneentrations were 8.3% and 9.2%, respectively, and for estradiol were 11% and 12.4%, respectively.
Chorulon, Intervet, Cambridge, UK and Milan, Italy 2 Intervet, Cambridge, UK and Milan, Italy
Theriogenology
126
Statistical Analysis Effect of age, intrauterine fluid, and timing of ovulation on pregnancy rates and effects of hCG were analyzed by a Chi-square test. Hormone concentrations were compared using a student's t test. Differences ofP < 0.05 were regarded as significant. RESULTS Forty-one of the 95 inseminations resulted in a pregnancy (43%). None of the mares included was older than 16 years, and age (3 to 9 years, n = 55 versus 10 to 16 years, n = 40) did not influence pregnancy rate (43% for younger mares v 43% for older mares). Timing of insemination (within 12 h before ovulation, compared with within 8 h after ovulation) did not significantly affect pregnancy rates (13/32, 41% with preovulation insemination versus 28/63, 44% with postovulation insemination). Similarly, timing of insemination did not significantly influence accumulation of intrauterine fluid 12 h later (34% of mares accumulated fluid after preovulatory insemination compared with 21% after postovulatory insemination). Transrtctai ultrasonography was performed 12 h after 52 of the inseminations and 18 to 24 h after 43 of the inseminations (Table 1). Of the mares examined 12 h later, 18 (35%) had >_15 mm inlrauterine fluid. Pregnancy rates in these 18 mares were significantly lower (P < 0.05) in mares with fluid compared to mares with no fluid or only a small amount. Of the 18 mares with fluid at 12 h, 10 were in the older age group (56%, 10 to 16 years). This difference was not significant. Only 6 of the mares examined for the first time after 18 to 24 h had >15 mm intrauterine fluid although the difference in detection rates between scanning early (12 h) versus later (18 to 24 h) failed to reach significance (P = 0.07). Of these 6 mares, three became pregnant (50%) compared with 15 (41%) of the mares with little or no fluid. Table 1. Presence of intrauterine fluid (IUF) at 12 and 18 to 24 h after insemination and effects on pregnancy rates First scanned
presence oflUF pregnancy rates in mares with IUF pregnancy rates in mares without 1UF *P < 0.05
12 h post AI
18 to 24 h post AI
18/52 (35%)
6/43 (14%)
5/18 (28%)*
3/6 (50%)
18/34 (53%)*
15/37 (41%)
Concentrations of progesterone and estrogen at time of insemination did not significantly affect pregnancy rates and progesterone concentrations were not significantly higher at the time of insemination in those marts that were inseminated postovulation (Table 2). However, serum
Theriogenology
127
estrogen and progesterone concentrations were significantly higher at the time ofpreovulatory insemination. Concentrations o f circulating estrogen and progesterone at insemination were not related to presence of intrauterine fluid 12 h later (Table 2). Number of large follicles (> 35mm) at time of ovulation did not influence circulating concentrations of estrogen or progesterone (Table 2).
Table 2. Serum concentrations o f estrogen and progesterone at AI
n
Hormone concentrations at time of Al Estrogen pg/mL Progesterone ng/mL
Pregnant Not pregnant
23 29
36.6 + 6.35 43.2 + 8.93
0.21 + 0.03 0.28 + 0.04
Preovulatory AI Postovulatory AI
32 20
48.9 + 8.73 * 26.3 _+2.72 *
0.28 + 0.03 * 0.15 _+0.03 *
18
39.5 _+ 10.99
0.28 + 0.06
34
40.3 _+6.24
0.22 _+0.03
25 38.9 + 7.08 } at AI or } at last scan 27 40.3 _+8.72 Two follicles >_35mm } before AI * Values significantly different (P < 0.01) within columns.
0.24 +0.03
> 15mm uterine fluid < 15mm uterine fluid
} 12h } post } AI
Single follicle
0.25 + 0.03
DISCUSSION We showed that timing of insemination, either before or after ovulation, did not influence accumulation of uterine fluid 12 h later. Similarly circulating concentrations of steroid hormones at time of insemination were not related to subsequent accumulation of uterine fluid. In the present study, pregnancy rates were similar to those previously reported for insemination of frozen semen (2, 3, 7, 10, 25, 27). The stallion, dose and quality of spermatozoa and number of inseminations all have a profound effect on single cycle pregnancy rates (10, 19, 25). However, in agreement with recent studies, in our mares timing of insemination within a period of 12 h before, or 4 to 8 hatter ovulation did not significantly affect pregnancy rates (2, 3). By contrast, Samper (19) stated that a single preovulatory insemination resulted in higher pregnancy rates (60%) than insemination after ovulation (30%). However, this author was referring to within a period of 12 h postovulation and it has been shown that pregnancy rates fall as the oocyte ages between 6 and 12 h (11). The high pregnancy rate quoted by Samper (19) for a single
128
Theriogenology
preovulatory insemination was also achieved by postovulatory insemination for individual stallions in the present study, but not by other stallions. Other workers quoted similarly high rates for postovulation insemination in research situations using a small number of stallions (8, 12). This high rate is unlikely to be obtained in practice when a large number of commercial stallions are used. More mares first examined by transrectal ultrasonography at 12 h after insemination had intrauterine fluid (35%) than those first examined between 18 and 24 h after AI (14%), although this difference failed to reach statistical significance. The higher incidence 12 h after AI is not surprising as intrauterine inflammation peaks eight hours after AI with fresh semen (6). Time of first detection of fluid after AI did not significantly influence pregnancy rates, and pregnancy rates were not reduced when treatment was not instituted until 18 to 24 h after AI. Indeed it seems likely that a number o f mares were treated unnecessarily after the earlier scan. More work needs to be performed on the effect of timing of first treatment on subsequent pregnancy rates in mares with delayed uterine clearance. However,the overall low pregnancy rate in mares with intrauterine fluid (33%) does suggest that these mares need treatment. Other studies showed that mares with intraluminal fluid within 48 h of mating have lower pregnancy rates than do mares with no intraluminal fluid (20, 31) and mares with fluid at any time in diestrus after insemination with fresh semen have profoundly reduced pregnancy rates (1). Treatment o f mares that retain intraluminal fluid has been shown to aid fluid removal and increase pregnancy rates (17, 24). The current recommendations for treatment regimens are as performed in the present study, that is oxytocin therapy with or without large volume intrauterine lavage, depending on the volume of fluid retained (22). Unfortunately, in the present study we could not leave mares with intrauterine fluid untreated to study the effect o f treatment on pregnancy rates, but the low pregnancy rates achieved in these mares and mares in other studies would suggest that an optimal treatment regimen remains to be devised. In the present study the incidence of moderate to large accumulations of fluid on the day after insemination with frozen semen was similar to that reported after natural mating (16% of cycles) (31). This is surprising in that intrauterine infusion of frozen semen has been shown to result in a significantly greater inflammatory response than natural breeding (9). These latter authors hypothesized that the intensity of neutrophil reaction depended on the concentration and/or volume of inseminate. More recently Troedsson and co-workers (23) showed that seminal plasma suppressed neutrophil migration in vitro and then Nikolakopoulos & Watson (15) showed that larger infusion volumes were associated with a reduced neutrophil response in mares. However in the present study the population of mares was relatively young, and mares with a history of low fertility were not accepted as candidates for insemination with frozen semen. If an unselected population had been used, it is likely that fluid accumulation rates would have been higher. Indeed it has been shown that fluid accumulation after insemination of frozen semen is higher in mares over 16 years old (3).
Theriogenology
129
Concentrations of estrogen and progesterone were significantly higher at the time of insemination when insemination was carried out in the preovulatory period. The high estrogen at this time would derive from the preovulatory follicle, the main source of estradiol 17-13 in the mare. The higher levels of circulating progesterone were unexpected. However it has been shown previously that intrafollicular concentrations of progesterone increase after hCG treatment (28) and so it is possible that the preovulatory follicle secretes more progesterone than the newly ovulated corpus hemorrhagicum. Pregnancy rates were not affected by circulating steroid concentrations at the time of AI. Circulating steroid concentrations at AI did not influence accumulation of intrauterine fluid after insemination and so we had no evidence that postovulatory insemination could predispose to persistent mating-induced endometritis. Indeed in our study progesterone concentrations were higher before ovulation. There was no evidence that mares with two follicles had higher circulating estrogen concentrations than mares with one large follicle. Therefore increased numbers of follicles did not appear to confer mares with higher estrogen levels that might be beneficial in uterine defense and clearance. In conclusion, accumulation of intrauterine fluid in mares alter insemination with frozen semen was not significantly affected by timing of insemination before or after ovulation. Furthermore there was no correlation between circulating concentrations of estrogen and progesterone at time of insemination and post-insemination fluid accumulation. Fewer mares had intrauterine fluid on the day after AI than at 12 h after AI, and so routine treatment as early as 12 h will include some normal fertile mares. REFERENCES 1. Adams GP, Kastelic JP, Bergfelt DR, Ginther OJ. Effect of uterine inflammation and ultrasonically-detected uterine pathology on fertility in the mare. J Reprod Fertil
1997;35(Suppl):445-454. 2. Barbacini S, Gulden P, Marchi V, Zavaglia G. Incidence of embryo loss in mares inseminated before or after ovulation. Equine Vet Educ 1999;11:251-254. 3. Barbacini S, Marchi V, Zavaglia G. Equine frozen semen: results obtained in Italy during the 1994-1997 period. Equine Vet Educ 1999;11:109-112. 4. Evans MJ, Hamer J, Gason LM, Irvine CHG. Factors affecting uterine clearance of inoculated material in mares. J Reprod Fertil 1987;35(Suppl):327-334. 5. Horserace Betting Levy Board. Codes of Practice on Contagious Equine Metritis (CEM), KlebsieUa pneumoniae, Pseudomonas aeruginosa, Equine Viral Arteritis (EVA) and Equid Herpesvirus-I (EHV-I) HBLB (Pub), 2000; 52 Grosvenor Gardens, London. 6. Katila T. Onset and duration of uterine inflammatory response of mares after insemination with fresh semen. Biol Reprod Mono 1995;1:515-517. 7. Katila T, Celebi M, Koskinen E. Effect of timing of frozen semen insemination on pregnancy rates in mares. Acta Vet Scand 1996;37:361-365. 8. Kloppe LH, Varner DD, Elmore RG, Bretzlaff KN, Shull JW. Effect of insemination timing on the fertilizing capacity of frozen/thawed equine spermatozoa. Theriogenology
1988;29:429-439.
Theriogenology
130 9. 10.
11.
12. 13. 14. 15. 16.
17.
Kotilainen T, Huhtinen M, Katila T. Sperm-induced leukocytosis in the equine uterus. Theriogenology 1994;41:629-636. Leipold SD, Graham JK, Squires EL, Mclve PM, Brinsko SP, Vanderwall DK. Effect of spermatozoal concentration and number on fertility of frozen equine semen. Theriogenology 1998;49:1537-1543. Lindeberg H, Kuntsi-Vaattovaara H. Kukemuleoia oriin pakastesperman kaytosta siittolassa vuosina 1989-1991. (Experiences of the use of frozen stallion semen on a stud farm in 1989-1991). Suom Ellaaklehti 1992;98:654-660. Martin JC, Klug E, Gunzel AR. Centrifugation of stallion semen and its storage in large volume straws. J Reprod Fertil 1979;27(Suppl):613-622. Newcombe JR. The effect of the incidence and depth of intra-uterine fluid in early dioestrus on pregnancy rate in mares. Pferdeheilkunde 1997;5:545. Newcombe JR. Practical evaluation of the fertilising capacity of frozen-thawed horse semen. Vet Rec 1999;145:46-47. Nikolakopoulos E, Watson ED. Effect of infusion volume and sperm numbers on persistence of uterine inflammation in mares. Equine Vet J 2000;32:164-166. Ousey JC, Rossdale PD, Palmer L, Grainger L, Houghton E. Effects of maternally administered depot ACTH 1-24 on foetal maturation and the timing of parturition in the mare. Equine VetJ 2000;32:489-496. Rasch K, Schoon HA, Sieme H, Klug E. Histomorphological endometrial status and influence ofoxytocin on the uterine drainage and pregnancy rates in mares. Equine Vet J
1996;28:455-460. 18. Reimers J, Salerno VJ, Lamb SV. Validation and application of solid-phase chemiluminescent immtmoassays for diagnosis of endocrine diseases in animals. Comp Haematol lnt 1996;6:170-175. 19. Samper JC. Stallion semen cryopreservation: male factors affecting pregnancy rates. Proc Soc Theriogenology San Antonio 1995;160-165. 20. Squires EL, Barnes CK, Rowley HS, McKinnon AO, Pickett BW, Shideler RK. Effect of uterine fluid and volume of extender on fertility. Proc 35th Ann Conv Am Assoc Equine Pract 1989;25-30 21. Townson DH, Pierson RA, Ginther OJ. Characterization of plasma progesterone concentrations for the distinct luteal morphologies in mares. Theriogenology 1989;32:197204. 22. Troedsson MHT. Therapeutic considerations for mating-induced endometritis. Pferdeheilkunde 1997;13:516-520. 23. Troedsson MHT, Crabo BG, Ibrahim N, Scott M, Ing M. Mating-induced endometritis: mechanisms, ethical importance and consequences. Proc 41 a Mtg AM Assoc Equine Pract, 1995;11-12. 24. Troedsson MHT, Scott MA, Liu IKM. Comparative treatment of mares susceptible to chronic uterine infection. Am J Vet Res 1995;56:468-472. 25. Vidament M, Dupere AM, Julienne P, Evain A, Nove P, Palmer E. Equine fi'ozen semen freezability and fertility field results. Theriogenology 1997;48:907-917. 26. Volkman DH, Van Zyl D. Fertility of stallion semen fi'ozen in 0.5 ml straws. J Reprod Fertil 1987;35(Suppl):143-148.
Theriogenology 27. 28.
131
Watson ED. Artificial insemination in horses. In Practice 1994;17:54-58. Watson ED, Hinrichs K. Changes in the concentrations of steroids and prostaglandin F in preovulatory follicles of the mare after administration of hCG. J Reprod Fertil 1988;84:557-561. 29. Watson ED, Stokes CR, Bourne FJ. The influence of administration of ovarian steroids on the function ofneulrophils isolated from the blood and uterus ofovariectomized mares. J Endocrinol 1987;112:443-448. 30. Watson ED, Stokes CR, David JSE, Bourne FJ. Effect of ovarian hormones on promotion of bactericidal activity by uterine secretions of ovariectomized mares. J Reprod Fertil 1987;79:531-537. 31. Zent WW, Troedsson MHT, Xue J-L. Postbreeding uterine fluid accumulation in a normal population of thoroughbred mares: a field study. Proc. Soc. Theriogenology, Baltimore, 1998;78-79.
EFFECT OF INSEMINATION TIME OF FROZEN SEMEN ON INCIDENCE OF UTERINE FLUID IN MARES E. D. Watson, ~ S. Barbacini,2 B. Berrocal, ~ O. Sheerin, ~V. Marchi, 2 G. Zavaglia2 and D. Necchi 2 Department of Veterinary Clinical Studies, Royal (Dick) School of Veterinary Studies, Easter Bush, Midlothian, EH25 9RG, Scotland 2 Studio Veterinario Cristella, Via Argine 39, S. Daniele Po (CR), Italy Received for publication: January ] 8, 2000 Accepted: January 24, 200] ABSTRACT Ninety five mares were inseminated with frozen semen either within 12 h before ovulation or within 8 h at~r ovulation. The effect ofpreovulatory versus postovulatory insemination (AI) on the subsequent detection of uterine fluid was studied. The overall pregnancy rate was 43% and this was not significantly influenced by preovulatory or postovulatory insemination. When mares were first examined 12 h after AI, 18 o f 52 mares (35%) had accumulated uterine fluid. However, when mares were first examined 18 to 24 h after AI, only 6 of 43 mares (14%) had uterine fluid. Presence o f intrauterine fluid significantly lowered pregnancy rates. Timing of insemination did not affect incidence of uterine fluid. Serum concentrations of estrogen and progesterone at time of insemination did not influence uterine clearance or pregnancy rates, but both hormones were higher at preovulatory than at postovulatory inseminations. We concluded that there was no evidence that postovulatory inseminations would predispose mares to persistence of uterine fluid after AI. © 2001 by Elsevier Science Inc.
Key words: mare, insemination, endometritis INTRODUCTION Use of frozen semen for insemination of mares is increasing significantly throughout the world. Improvement in freezing techniques has resulted in pregnancy rates of greater than 30% per cycle (3, 14, 19, 25). These rates change depending on the number of post-thaw motile spermatozoa per inseminate, on the proximity of insemination to time of ovulation and the number of inseminations per estrus (10, 19, 25, 26).
Acknowledgments The study was supported in part by Intervet International, Intervet Italy and lntervet UK. The ultrasound machine used in Italy (Sonoace 600V) was donated by Kretz Technik, Zipf, Austria.
Theriogenology 56:123-131, 2001 © 2001 Elsevier Science Inc.
O093-691X/O1/$--see front matter Pll: S0093-691X(01)00548-9
Theriogenology
124
The use o f a small volume of highly concentrated spermatozoa in uterine insemination in mares has resulted in a more marked inflammatory response than did the use of a larger volume of inseminate (9, 15). It has been shown in vitro that equine spermatozoa are chemotactic for neutrophils (23). Addition of seminal plasma was shown to suppress the chemotactic effect of the spermatozoa in vitro (23), but addition o f seminal plasma to sperm did not suppress the influx of neutrophils into the uterus o f the mare in vivo (9). The low volume used in insemination of frozen semen may be an important factor (9, 15). With large insemination volumes, considerable amounts o f fluid and spermatozoa are discharged into the vagina shortly after insemination (20), whereas with low volumes, it is likely that the spermatozoa are in contact with the endometrium for a longer period, causing neutrophil influx into the lumen by their chemotactic properties. Although many studies have reported the effect o f different freezing procedures on pregnancy rates in mares, few field studies have investigated the uterine response to AI with frozen semen and its effect on pregnancy rates. An inflammatory response in the mare's uterus in response to insemination may be monitored ultrasonographically by the presence o f fluid accumulations within the uterine lumen (22). As increased circulating concentrations of progesterone are associated with suppression of both uterine immune defense mechanisms and uterine clearance (4, 29, 30), and as progesterone increases within hours of ovulation in the mare (21), we hypothesized that insemination in the postovulatory period might predispose mares to post-insemination fluid accumulation, compared with preovulatory insemination. We report the outcome of inseminations performed at a center in Scotland in 1999 and in Italy in 1998 with respect to the presence o f ultrasonographically detectable intrauterine fluid after insemination, whether this was influenced by timing of insemination, or by circulating steroid hormone concentrations at the time of insemination. MATERIALS AND METHODS Records were available from a total of 95 mares at the R(D)SVS, Scotland and the Cristella Veterinary Clinic in Italy. Only first inseminations were included. The mares were aged 3 to 16 years and most were Warmbloods, with a few quarterhorses and Thoroughbreds. Uterine swabs, collected from all mares early in the same estrus as AI (5), were negative on bacteriological culture and cytological examination. None o f the mares had more than a trace o f uterine fluid at the time of insemination. Semen from 42 stallions frozen at 18 different centers was used. The semen was packaged mainly in 0.5 mL straws, however 5 mL macrotubes were used at one center. The minimum number of post-thaw progressively motile spermatozoa inseminated was 100 million. Timing of Insemination If mares were in late diestrus or early estrus, they were scanned daily by transrectal ultrasonography until a follicle of at least 35 mm in diameter was present in the ovaries, along with a positive tease response to a stallion and ultrasonographically visible uterine edema. These
Theriogenology
125
mares were treated with human cborionic gonadotrophin intravenously (hCG; 2000IU I) on first day of detection of a follicle of 30 to 40 mm in diameter. In some of the cycles, mares were not presented until later in estrus when large follicles >40 mm were present. These mares either received hCG or were scanned every 4 to 6 h until ovulation depending on the size and degree of soliening of the follicle. The ovaries and uterus of mares were examined by transrectal ultrasonography at 12 and 24 h after hCG and then every 4 to 8 h until ovulation. Jugular blood samples were collected at the time o f insemination from 52 o f the mares into plain evacuated tubes. Serum was separated and stored at -20°C until assayed for estrogen and progesterone. Mares were inseminated once only, either within 4 to 8 hatter ovulation was detected, or around 36 h after hCG treatment if the mare had not yet ovulated (n = 32). All o f these 32 mares ovulated within 12 h after insemination. Mares were prepared for insemination as described previously (27) and semen was thawed according to the semen distribution center's instructions. The semen was inseminated immediately and then a drop was checked for progressive motility in the laboratory. All semen inseminated was at least 30% progressively motile after thawing. Post-Insemination Monitoring Transrectal ultrasonography was performed either 12 (n = 52) or 18 to 24 h (n = 43) after insemination. The presence and depth of intrauterine fluid was recorded. Fifteen mm of intrauterine fluid was recorded as a significant volume as this amount of fluid in diestrus reduces pregnancy rates (13). Mares with less than 15 mm intrauterine fluid received an intravenous injection of oxytocin (20 IU; Oxytocin $2). In mares with greater than 15 mm intrauterine fluid, buffered saline solution was infused and recovered (one liter aliquots) until the recovered fluid was clear. The mares then received oxytocin (20 IU iv). Oxytocin treatment was repeated up to three times on that day. The combination of lavage and oxytocin was repeated daily as necessary until no intrauterine fluid was detected on ultrasound scan, for up to three days after insemination. The mares were scanned for pregnancy between 14 and 17 days postovulation. Hormone Assays Progesterone and estradiol-17[3 were measured by a solid phase, chemiluminescent enzyme immunoassay at Beaufort Cottage Laboratories, Newmarket. The assays were performed as described by Reimers et al. (18) and Ousey et al. (16). Major cross-reactivities of the progesterone antibody were with progesterone (100%), 5ct-pregnane-3,20-dione (1.6%), 17~t-hydroxyprogesterone (1%), 11-deoxycorticosterone (1.2%) and of the estradiol antibody were with estmdiol-17~ (100%), ~t-equilenin (1.6%) and ethinyl estradiol (1.2%). Sensitivity of the progesterone assay was 0.2 ng/mL and of the estradiol assay was 12 pg/mL. Mean intra- and inter-assay coefficients o f variation for progesterone for a range of coneentrations were 8.3% and 9.2%, respectively, and for estradiol were 11% and 12.4%, respectively.
Chorulon, Intervet, Cambridge, UK and Milan, Italy 2 Intervet, Cambridge, UK and Milan, Italy
Theriogenology
126
Statistical Analysis Effect of age, intrauterine fluid, and timing of ovulation on pregnancy rates and effects of hCG were analyzed by a Chi-square test. Hormone concentrations were compared using a student's t test. Differences ofP < 0.05 were regarded as significant. RESULTS Forty-one of the 95 inseminations resulted in a pregnancy (43%). None of the mares included was older than 16 years, and age (3 to 9 years, n = 55 versus 10 to 16 years, n = 40) did not influence pregnancy rate (43% for younger mares v 43% for older mares). Timing of insemination (within 12 h before ovulation, compared with within 8 h after ovulation) did not significantly affect pregnancy rates (13/32, 41% with preovulation insemination versus 28/63, 44% with postovulation insemination). Similarly, timing of insemination did not significantly influence accumulation of intrauterine fluid 12 h later (34% of mares accumulated fluid after preovulatory insemination compared with 21% after postovulatory insemination). Transrtctai ultrasonography was performed 12 h after 52 of the inseminations and 18 to 24 h after 43 of the inseminations (Table 1). Of the mares examined 12 h later, 18 (35%) had >_15 mm inlrauterine fluid. Pregnancy rates in these 18 mares were significantly lower (P < 0.05) in mares with fluid compared to mares with no fluid or only a small amount. Of the 18 mares with fluid at 12 h, 10 were in the older age group (56%, 10 to 16 years). This difference was not significant. Only 6 of the mares examined for the first time after 18 to 24 h had >15 mm intrauterine fluid although the difference in detection rates between scanning early (12 h) versus later (18 to 24 h) failed to reach significance (P = 0.07). Of these 6 mares, three became pregnant (50%) compared with 15 (41%) of the mares with little or no fluid. Table 1. Presence of intrauterine fluid (IUF) at 12 and 18 to 24 h after insemination and effects on pregnancy rates First scanned
presence oflUF pregnancy rates in mares with IUF pregnancy rates in mares without 1UF *P < 0.05
12 h post AI
18 to 24 h post AI
18/52 (35%)
6/43 (14%)
5/18 (28%)*
3/6 (50%)
18/34 (53%)*
15/37 (41%)
Concentrations of progesterone and estrogen at time of insemination did not significantly affect pregnancy rates and progesterone concentrations were not significantly higher at the time of insemination in those marts that were inseminated postovulation (Table 2). However, serum
Theriogenology
127
estrogen and progesterone concentrations were significantly higher at the time ofpreovulatory insemination. Concentrations o f circulating estrogen and progesterone at insemination were not related to presence of intrauterine fluid 12 h later (Table 2). Number of large follicles (> 35mm) at time of ovulation did not influence circulating concentrations of estrogen or progesterone (Table 2).
Table 2. Serum concentrations o f estrogen and progesterone at AI
n
Hormone concentrations at time of Al Estrogen pg/mL Progesterone ng/mL
Pregnant Not pregnant
23 29
36.6 + 6.35 43.2 + 8.93
0.21 + 0.03 0.28 + 0.04
Preovulatory AI Postovulatory AI
32 20
48.9 + 8.73 * 26.3 _+2.72 *
0.28 + 0.03 * 0.15 _+0.03 *
18
39.5 _+ 10.99
0.28 + 0.06
34
40.3 _+6.24
0.22 _+0.03
25 38.9 + 7.08 } at AI or } at last scan 27 40.3 _+8.72 Two follicles >_35mm } before AI * Values significantly different (P < 0.01) within columns.
0.24 +0.03
> 15mm uterine fluid < 15mm uterine fluid
} 12h } post } AI
Single follicle
0.25 + 0.03
DISCUSSION We showed that timing of insemination, either before or after ovulation, did not influence accumulation of uterine fluid 12 h later. Similarly circulating concentrations of steroid hormones at time of insemination were not related to subsequent accumulation of uterine fluid. In the present study, pregnancy rates were similar to those previously reported for insemination of frozen semen (2, 3, 7, 10, 25, 27). The stallion, dose and quality of spermatozoa and number of inseminations all have a profound effect on single cycle pregnancy rates (10, 19, 25). However, in agreement with recent studies, in our mares timing of insemination within a period of 12 h before, or 4 to 8 hatter ovulation did not significantly affect pregnancy rates (2, 3). By contrast, Samper (19) stated that a single preovulatory insemination resulted in higher pregnancy rates (60%) than insemination after ovulation (30%). However, this author was referring to within a period of 12 h postovulation and it has been shown that pregnancy rates fall as the oocyte ages between 6 and 12 h (11). The high pregnancy rate quoted by Samper (19) for a single
128
Theriogenology
preovulatory insemination was also achieved by postovulatory insemination for individual stallions in the present study, but not by other stallions. Other workers quoted similarly high rates for postovulation insemination in research situations using a small number of stallions (8, 12). This high rate is unlikely to be obtained in practice when a large number of commercial stallions are used. More mares first examined by transrectal ultrasonography at 12 h after insemination had intrauterine fluid (35%) than those first examined between 18 and 24 h after AI (14%), although this difference failed to reach statistical significance. The higher incidence 12 h after AI is not surprising as intrauterine inflammation peaks eight hours after AI with fresh semen (6). Time of first detection of fluid after AI did not significantly influence pregnancy rates, and pregnancy rates were not reduced when treatment was not instituted until 18 to 24 h after AI. Indeed it seems likely that a number o f mares were treated unnecessarily after the earlier scan. More work needs to be performed on the effect of timing of first treatment on subsequent pregnancy rates in mares with delayed uterine clearance. However,the overall low pregnancy rate in mares with intrauterine fluid (33%) does suggest that these mares need treatment. Other studies showed that mares with intraluminal fluid within 48 h of mating have lower pregnancy rates than do mares with no intraluminal fluid (20, 31) and mares with fluid at any time in diestrus after insemination with fresh semen have profoundly reduced pregnancy rates (1). Treatment o f mares that retain intraluminal fluid has been shown to aid fluid removal and increase pregnancy rates (17, 24). The current recommendations for treatment regimens are as performed in the present study, that is oxytocin therapy with or without large volume intrauterine lavage, depending on the volume of fluid retained (22). Unfortunately, in the present study we could not leave mares with intrauterine fluid untreated to study the effect o f treatment on pregnancy rates, but the low pregnancy rates achieved in these mares and mares in other studies would suggest that an optimal treatment regimen remains to be devised. In the present study the incidence of moderate to large accumulations of fluid on the day after insemination with frozen semen was similar to that reported after natural mating (16% of cycles) (31). This is surprising in that intrauterine infusion of frozen semen has been shown to result in a significantly greater inflammatory response than natural breeding (9). These latter authors hypothesized that the intensity of neutrophil reaction depended on the concentration and/or volume of inseminate. More recently Troedsson and co-workers (23) showed that seminal plasma suppressed neutrophil migration in vitro and then Nikolakopoulos & Watson (15) showed that larger infusion volumes were associated with a reduced neutrophil response in mares. However in the present study the population of mares was relatively young, and mares with a history of low fertility were not accepted as candidates for insemination with frozen semen. If an unselected population had been used, it is likely that fluid accumulation rates would have been higher. Indeed it has been shown that fluid accumulation after insemination of frozen semen is higher in mares over 16 years old (3).
Theriogenology
129
Concentrations of estrogen and progesterone were significantly higher at the time of insemination when insemination was carried out in the preovulatory period. The high estrogen at this time would derive from the preovulatory follicle, the main source of estradiol 17-13 in the mare. The higher levels of circulating progesterone were unexpected. However it has been shown previously that intrafollicular concentrations of progesterone increase after hCG treatment (28) and so it is possible that the preovulatory follicle secretes more progesterone than the newly ovulated corpus hemorrhagicum. Pregnancy rates were not affected by circulating steroid concentrations at the time of AI. Circulating steroid concentrations at AI did not influence accumulation of intrauterine fluid after insemination and so we had no evidence that postovulatory insemination could predispose to persistent mating-induced endometritis. Indeed in our study progesterone concentrations were higher before ovulation. There was no evidence that mares with two follicles had higher circulating estrogen concentrations than mares with one large follicle. Therefore increased numbers of follicles did not appear to confer mares with higher estrogen levels that might be beneficial in uterine defense and clearance. In conclusion, accumulation of intrauterine fluid in mares alter insemination with frozen semen was not significantly affected by timing of insemination before or after ovulation. Furthermore there was no correlation between circulating concentrations of estrogen and progesterone at time of insemination and post-insemination fluid accumulation. Fewer mares had intrauterine fluid on the day after AI than at 12 h after AI, and so routine treatment as early as 12 h will include some normal fertile mares. REFERENCES 1. Adams GP, Kastelic JP, Bergfelt DR, Ginther OJ. Effect of uterine inflammation and ultrasonically-detected uterine pathology on fertility in the mare. J Reprod Fertil
1997;35(Suppl):445-454. 2. Barbacini S, Gulden P, Marchi V, Zavaglia G. Incidence of embryo loss in mares inseminated before or after ovulation. Equine Vet Educ 1999;11:251-254. 3. Barbacini S, Marchi V, Zavaglia G. Equine frozen semen: results obtained in Italy during the 1994-1997 period. Equine Vet Educ 1999;11:109-112. 4. Evans MJ, Hamer J, Gason LM, Irvine CHG. Factors affecting uterine clearance of inoculated material in mares. J Reprod Fertil 1987;35(Suppl):327-334. 5. Horserace Betting Levy Board. Codes of Practice on Contagious Equine Metritis (CEM), KlebsieUa pneumoniae, Pseudomonas aeruginosa, Equine Viral Arteritis (EVA) and Equid Herpesvirus-I (EHV-I) HBLB (Pub), 2000; 52 Grosvenor Gardens, London. 6. Katila T. Onset and duration of uterine inflammatory response of mares after insemination with fresh semen. Biol Reprod Mono 1995;1:515-517. 7. Katila T, Celebi M, Koskinen E. Effect of timing of frozen semen insemination on pregnancy rates in mares. Acta Vet Scand 1996;37:361-365. 8. Kloppe LH, Varner DD, Elmore RG, Bretzlaff KN, Shull JW. Effect of insemination timing on the fertilizing capacity of frozen/thawed equine spermatozoa. Theriogenology
1988;29:429-439.
Theriogenology
130 9. 10.
11.
12. 13. 14. 15. 16.
17.
Kotilainen T, Huhtinen M, Katila T. Sperm-induced leukocytosis in the equine uterus. Theriogenology 1994;41:629-636. Leipold SD, Graham JK, Squires EL, Mclve PM, Brinsko SP, Vanderwall DK. Effect of spermatozoal concentration and number on fertility of frozen equine semen. Theriogenology 1998;49:1537-1543. Lindeberg H, Kuntsi-Vaattovaara H. Kukemuleoia oriin pakastesperman kaytosta siittolassa vuosina 1989-1991. (Experiences of the use of frozen stallion semen on a stud farm in 1989-1991). Suom Ellaaklehti 1992;98:654-660. Martin JC, Klug E, Gunzel AR. Centrifugation of stallion semen and its storage in large volume straws. J Reprod Fertil 1979;27(Suppl):613-622. Newcombe JR. The effect of the incidence and depth of intra-uterine fluid in early dioestrus on pregnancy rate in mares. Pferdeheilkunde 1997;5:545. Newcombe JR. Practical evaluation of the fertilising capacity of frozen-thawed horse semen. Vet Rec 1999;145:46-47. Nikolakopoulos E, Watson ED. Effect of infusion volume and sperm numbers on persistence of uterine inflammation in mares. Equine Vet J 2000;32:164-166. Ousey JC, Rossdale PD, Palmer L, Grainger L, Houghton E. Effects of maternally administered depot ACTH 1-24 on foetal maturation and the timing of parturition in the mare. Equine VetJ 2000;32:489-496. Rasch K, Schoon HA, Sieme H, Klug E. Histomorphological endometrial status and influence ofoxytocin on the uterine drainage and pregnancy rates in mares. Equine Vet J
1996;28:455-460. 18. Reimers J, Salerno VJ, Lamb SV. Validation and application of solid-phase chemiluminescent immtmoassays for diagnosis of endocrine diseases in animals. Comp Haematol lnt 1996;6:170-175. 19. Samper JC. Stallion semen cryopreservation: male factors affecting pregnancy rates. Proc Soc Theriogenology San Antonio 1995;160-165. 20. Squires EL, Barnes CK, Rowley HS, McKinnon AO, Pickett BW, Shideler RK. Effect of uterine fluid and volume of extender on fertility. Proc 35th Ann Conv Am Assoc Equine Pract 1989;25-30 21. Townson DH, Pierson RA, Ginther OJ. Characterization of plasma progesterone concentrations for the distinct luteal morphologies in mares. Theriogenology 1989;32:197204. 22. Troedsson MHT. Therapeutic considerations for mating-induced endometritis. Pferdeheilkunde 1997;13:516-520. 23. Troedsson MHT, Crabo BG, Ibrahim N, Scott M, Ing M. Mating-induced endometritis: mechanisms, ethical importance and consequences. Proc 41 a Mtg AM Assoc Equine Pract, 1995;11-12. 24. Troedsson MHT, Scott MA, Liu IKM. Comparative treatment of mares susceptible to chronic uterine infection. Am J Vet Res 1995;56:468-472. 25. Vidament M, Dupere AM, Julienne P, Evain A, Nove P, Palmer E. Equine fi'ozen semen freezability and fertility field results. Theriogenology 1997;48:907-917. 26. Volkman DH, Van Zyl D. Fertility of stallion semen fi'ozen in 0.5 ml straws. J Reprod Fertil 1987;35(Suppl):143-148.
Theriogenology 27. 28.
131
Watson ED. Artificial insemination in horses. In Practice 1994;17:54-58. Watson ED, Hinrichs K. Changes in the concentrations of steroids and prostaglandin F in preovulatory follicles of the mare after administration of hCG. J Reprod Fertil 1988;84:557-561. 29. Watson ED, Stokes CR, Bourne FJ. The influence of administration of ovarian steroids on the function ofneulrophils isolated from the blood and uterus ofovariectomized mares. J Endocrinol 1987;112:443-448. 30. Watson ED, Stokes CR, David JSE, Bourne FJ. Effect of ovarian hormones on promotion of bactericidal activity by uterine secretions of ovariectomized mares. J Reprod Fertil 1987;79:531-537. 31. Zent WW, Troedsson MHT, Xue J-L. Postbreeding uterine fluid accumulation in a normal population of thoroughbred mares: a field study. Proc. Soc. Theriogenology, Baltimore, 1998;78-79.