Clinical Theriogenology 2022; 14: 18 Introduction For most livestock species, there is an economical influence to have a certain sex of offspring. Livestock producers may in- crease the probability for progeny of the most desired sex by various means of manipulating the breeding processes. There are many options to preselect sex of offspring, including em- bryo transfer, sex-sorted sperm, and preimplantation genetic testing.1,2 In equine industry, desired sex of offspring is de- pendent on the intended sport. For example, majority of top performing (polo) mares can be inseminated and capable of producing embryos concurrent to their sport careers for trans- fer to broodmare recipients, thereby maintaining financial productivity.3 Colts are more desirable among racing Thor- oughbreds because of higher sale prices and earning potential, and ability to be used as a breeding stallion after a successful career; a Thoroughbred mare can only produce 1 foal per year, whereas a stallion can breed many mares in a year.4 Breed- ing restrictions placed on Thoroughbreds registered in Jock- ey Club also influence the desire for colts. Jockey Club does not approve artificial insemination (AI), cloning, and embryo transfer for Thoroughbreds.5 Registered Thoroughbreds are re- quired to be conceived only by natural breeding (‘live cover’), making it difficult to manipulate the chances for a particular sex of a resulting foal. Given breeding restrictions like those placed on Thorough- breds, breeders consider other methods to increase the prob- ability of desired sex offspring. Period between breeding and ovulation (PBBO) had an influence on resulting foal sex.6 This approach can be a useful tool for breeders of horses such as Influence of period between breeding and ovulation on foal sex ratio Sabrina Hall,a Pablo Jarrin-Yepez,a Liesel Schneider,b Jessica Klabnik-Bradford,a William Riddle,c Woodrow Friend,c Maria Schnobrich,c Tulio Pradoa aCollege of Veterinary Medicine, bDepartment of Animal Science University of Tennessee, Knoxville, TN; cRood and Riddle Equine Hospital, Lexington, KY Abstract Objective was to determine the effect of period between breeding and ovulation (PBBO) on the sex of the foal. Foaling records (n = 288) from 2015 - 2018 breeding seasons and from 2 veterinary hospitals were used. There were 122 (42%) fillies and 166 (58%) colts. Foal sex records were categorized into 3 groups based on PBBO < 24 hours, 24 - 48 hours, and > 48 hours. There was an effect (p < 0.0001) of PBBO on the probability of the foal being a colt. If the PBBO was < 24 hours, the odds for a colt were 23.3 times higher compared to > 48 hours group (OR = 23.6, 95% CI: 3.0, 185, p = 0.003). Additionally, if the PBBO was < 24 hours, the odds for a colt were 4.6 times higher compared to 24 - 48 hours group (OR = 4.6, 95% CI: 2.0, 10.5, p = 0.0003). There were no differences (p = 0.14) between odds for a colt between 24 - 48 and > 48 hours groups. Confounding variables such as sire, age of dam, size of ovulated primary follicle, and type of semen (fresh or chilled) were not significant. Results indicated that the closer a mare is bred to ovulation with fresh semen, it is more likely that she will have a colt. Keywords: Ovulation, sex-ratio, mare, stallion, natural breeding the Thoroughbred where natural breeding is necessary. This can also be useful as a more economical method in preselect- ing foal sex. This retrospective study tested the merit behind the theory that the PBBO has an influence on resulting foal’s sex. We hypothesized that closer a mare is bred to ovulation, the more likely that she will have a colt. Materials and methods Data from 2 veterinary hospitals for 2015 - 2018 breeding seasons were used. Electronic medical records included name of the mare, breed, date of birth, reproductive status, infor- mation regarding ovulation induction medications (type and date), hormones used, date of breeding, date ovulation de- tected, estimated period in hours from breeding to ovulation, number of ovulations, size of largest primary follicle, type of semen (fresh or chilled) used, stallion name, postbreeding treatments, fetal sex, sex of foal, and veterinarian. Breeding records of twin pregnancies were not used. Animals Two-hundred-eighty-eight foaling records were analyzed. There were 279 Thoroughbreds, 5 American Quarter Horses, 2 Paso Finos, 1 Arabian, and 1 Appaloosa mare. Mares bred were between 3 and 22 years. Every Thoroughbred mare was inseminated by natural breeding. Nine mares represented oth- er breeds and were artificially inseminated (fresh or chilled semen). One hundred seven stallions were used to breed 288 Clinical Theriogenology 2022; 14: 19 mares. Stallions were between 3 and 24 years. There were 122 (42%) fillies and 166 (58%) colts. Information selected Periods between ovulation and breeding, sire, semen type (fresh or chilled), size of primary follicle, and mare’s age. PBBO was categorized into 3 groups: mares that ovulated < 24 hours after breeding (Group 1), mares that ovulated 24 - 48 hours after breeding (Group 2), and mares that ovulated > 48 hours after breeding (Group 3). Sires were compared within each group regardless of breeding method (natural breeding or AI). Data analyses FREQ procedure was utilized to descriptively analyze fre- quencies of foal sex in relation to other possible explanatory variables. Additionally, FREQ procedure was utilized to assess number of each stallion’s breeding. Each stallion was assigned to groups to test if frequently used stallions tended to yield more male or female foals. Finally, for inferential statistical analyses, separate, single variable generalized linear mixed models with binary distribution and logit link were used in the GLIMMIX procedure of SAS 9.4 (SAS Institute, Cary, NC) to model the probability of a foal being a colt. If variables were individually associated with foal sex, they were added into a model and manually eliminated in a backward fashion to elicit the most parsimonious explanatory model. The IL- INK function was used to convert logits to mean probabilities. Model adjusted probability to yield a colt and standard errors are presented. Significance was set at alpha = 0.05. Records for mares that conceived twins were not used for statistical analysis due to difficulty in identifying those that were fertil- ized first. Results Period between breeding to ovulation Descriptive statistics for the number of colts and fillies born to dams with each PBBO are provided (Table). There was an effect (p < 0.0001) of PBBO on the probability of having a foal colt. If the PBBO was < 24 hours, the odds for having a colt were 23.6 times higher compared to the odds for a colt in the > 48 hours group (OR=23.6, 95% CI: 3.0, 185, p = 0.003). Additionally, if the PBBO was < 24 hours, the odds for a colt birth were 4.7 times higher compared to the odds for a colt in the 24 - 48 hours group (OR = 4.7, 95% CI: 2.0, 10.5, p = 0.0003). There was no difference (p = 0.14) between odds for a colt in 24 - 48 hours and > 48 hours PBBO groups (Figure). Table. Descriptive statistics reporting the number (percentages) of colts and fillies born to dams bred within PBBO categories PBBO (hours) Colts Fillies Total < 24 156 (64%) 65 (36%) 242 24 - 48 9 (28%) 23 (72%) 32 > 48 1 (7%) 13 (93%) 14 Figure. Model-adjusted probabilities for the outcome of a colt foal were different (p < 0.0001) for PBBOs of < 24 hours, 24 - 48 hours, and > 48 hours. Error bars represent 1 SE of the mean, and bars with similar letter are not different at α = 0.05. Clinical Theriogenology 2022; 14: 20 Sire For the 288 foaling records, there were 107 exceptional stal- lions used as sires. Of the 107 sires, 19 were utilized to sire 5 or more foals. These 19 stallions were evaluated to determine if there was an association between sire and foal sex. Sire was not associated (p = 0.87) with the probability for a foal to be a colt. Age of dam Age of dam ranged from 3 to 22 years (mean = 8.56, median = 7.98). Age was considered a possible predictor of foal sex. For the purpose of analysis, mares’ age was categorized into 4 groups: mares ≤ 4 years (category 1); mares 4.1 - 7.9 years (category 2); 8 - 11.9 years (category 3); and > 12 years (cat- egory 4). There were 90, 93, 62, and 45 foals born to mares within the age categories 1, 2, 3, and 4, respectively. There was no relationship (p = 0.76) between mare age and foal sex out- comes. Type of semen/method of insemination There were 2 types of semen utilized in the dataset: fresh and chilled. There were 9 foals conceived using chilled semen compared to 279 conceived with fresh semen via natural breeding. There was no effect (p = 0.15) of semen type on foal sex. Although there was likely insufficient power to detect a difference in proportion of male or female foals to semen type, numerically 6 of 9 (67%) foals conceived with chilled semen were fillies, compared to 116 (42%) fillies of 279 con- ceived via natural breeding. Size of ovulated primary follicle Size of primary follicle ranged from 30 to 52 mm (mean = 40.08 mm, median = 40.0 mm). For analyses, follicle size was categorized into 3 groups based upon distribution of values. Categories 1, 2, and 3 corresponded to sizes 30 - 35, 35.1 - 42.9, > 43 mm, respectively. There was no association (p = 0.28) between primary follicle size category and foal sex. Use of ovulation inducing hormones Ovulation inducing hormones were utilized in 68 concep- tions. These hormones included human chorionic gonad- otropin (hCG), (2,500 IU; IV hCG, Chorulon: IntervetInc, Unterschleissheim, Germany), deslorelin acetate (1ml IM; Su- croMate, Thorn BioScience LLC, Louisville, KY) and the com- bination of hCG and deslorelin acetate (SucroMate, Thorn BioScience LLC) at the same dose. Number of pregnancies after use of each hormone or hormone combination was 8, 58, and 2, respectively. We analyzed the binary variable (yes or no) of hormone used with regard to foal sex outcomes. There was no association (p = 0.50) of hormone use and foal sex. Full model Although the only variable associated with foal sex was PBBO, we ran the full model including age of dam, size of primary follicle, use of ovulation inducing hormones, and PBBO to determine if there were any major confounding variables. In this full model, only PBBO was importantly associated (p < 0.0001) with foal sex. Other variables did not affect foal sex, nor did they largely impact or confound the effect of PBBO. Discussion Influence of PBBO on foal sex support research conducted in humans and many animal species. Other factors tested, such as the sire and age of dam, did not influence sex ratio, al- though some studies have reported otherwise. Data from sev- eral studies in humans have indicated that if PBBO is < 24 hours, there was either an equal sex ratio or a skew towards males.7 Compared to Y, X sperm are more resilient, making it more likely for them to succeed in achieving fertilization under stressful environmental conditions.6 Vulnerability of Y sperm is attributed to the act of early capacitation and consequent- ly, due to prolonged stay in uterus and oviduct.6 In humans, Y-bearing sperm had a substantially higher motility and swam through cervical mucus more efficiently than X-bearing sperm, and possibly provided more opportunities to fertilize ova around ovulation.7 In deer, decreases and increases in sex ratio were observed based on early and late natural breeding, respectively. Mating that occurred within 36 hours of the on- set of estrus (early), there was a decrease (27%) in number of male fawns. In mating that occurred within 37 - 96 hours after the onset of estrus (late), there was an increase (70%) in num- ber of male fawns. PBBO effect in cattle was inconsistent.6,8 In dairy cattle, there was an insignificant difference in sex ratio in various insemination-ovulation periods; early: 36 - 20 hours prior to ovulation; intermediate: 20 - 28 hours before ovula- tion; late: 8 - 12 hours after ovulation resulted in 50, 50, and 55% females respectively.9 This effect has rarely been studied in the mare, furthermore the sample was not adequate.10 How- ever, sex ratio was not different even when a larger sample size (n = 433) was used.11 Embryos (n = 16) derived from ovula- tions before 48 hours from AI resulted in more male offspring (57% males versus 43% females) than those (n = 10) that ovulated 48 hours after AI (30% males versus 70% females).12 There was a significant difference among those embryos that were fertilized after 48 hours from AI to those that were fertil- ized before 24 hours12 and no significant difference between those that ovulated between 24 and 48 hours. This was specu- lated to be due to unequal amount of resilience among X and Y sperm postbreeding, a possible reason for conflicting results on the impact of length of PBBO on sex ratios, particularly, in cattle that are synchronized for breeding. Result was also not significant in horses that tested this hypothes.1 We suggest that PBBO has significant impact on the sex of the resulting foal, possibly due to the early capacitation of the Y sperm. In horses, there may be an association among individual sires and sire’s breed on secondary sex ratio.13 Foaling records (n = 4,491) from 92 stallions (Thoroughbred, Arabian, and Akhal-Teke) indicated that sire’s breed had an influence on offspring sex ratio. Arabian and Thoroughbred stallions were reported to have higher (63%) or lower (46.1%) offspring sex ratio than the expected 50%. Akhal-Teke stallions had 53.9%, not significantly different from the hypothetical expected 50%.13 Results of the aforementioned study indicated that the probability of conceiving a certain foal sex can be increased Clinical Theriogenology 2022; 14: 21 with the use of semen from stallions that have a proven skewed ratio of progenies, although our results did not sug- gest that individual stallions are more likely to produce a spe- cific sex of foal. This was possibly due to the lower number of stallions with multiple breeding records used in this analysis. Age of the mare did not influence foal sex ratio. However, in a larger a study (n = 59,950) sex ratio decreased with an increase in mare’s age.14 More fillies were born to mares that were > 15 years compared to younger mares.14 It was suggested that mare’s age had a marked influence on offspring sex ratio.14 Sex ratio of the offspring of the mare have been known to mark- edly deviate from the 1:1 ratio in situations of decreased body condition of the mare at conception.10 Age of the mare could be correlated with a decreased condition or nutritional status that would skew resulting sex ratios at conception.10 There were not enough numbers of AI in our study to com- pare among methods of natural breeding and AI on the im- pact of the resulting sex ratio in horses. However, it has been suggested that sex ratio may not be different between frozen and fresh sperm use in women, and overall, AI appeared to favor male offspring (60%).7 Evidence is accumulating that estrus synchronization or ovulation induction treatments in cattle may influence sex ratio.6,8 Similar reports are not avail- able for mares. Conclusion Closer a mare is bred to ovulation with fresh semen, the more likely that she will have a colt supporting the theory of resil- iency of the X sperm and the long-term vulnerability of the Y sperm within the mare reproductive tract. Other possible factors (sire, age of dam, method of insemination, size of ovu- lated primary follicle, or use of ovulation inducing hormones) may not impact foal sex ratio. Conflict of interest None to declare. References 1. Panarace M, Pellegrini RO, Basualdo MO, et al: First field results on the use of stallion sex sorted semen in a large-scale embryo transfer program. Theriogenology 2014;81:520-525. 2. Herrera C: Clinical applications of preimplantation genetic testing in equine, bovine, and human embryos. J Equine Vet Sci 2016;41:29- 34. 3. Pashen RL, Lascombes FA, Darrow MD: The application of embryo transfer to polo ponies in Argentina. Equine Vet J 1993 (Suppl15);119-121. 4. Chezum B, Wimmer B: Roses or lemons: adverse selection in the market for Thoroughbred yearlings. Rev Econ Stat 1997;79:521-526. 5. The American Stud Book Principal Rules and Requirements. Jockey Club Interactive Registration, 2021. (accessed 7 May 2021). 6. Rorie RW: Effect of timing of artificial insemination on sex ratio. Theriogenology 1999;52:1273-1280. 7. Sampson JH, Alexander NJ, Fulgham DL, et al: Gender after artificial induction of ovulationand artificial insemination. The American Fertility Society 1983;40:481-484. 8. Rorie RW, Lester TD, Lindsey BR, et al: Effect of timing of artificial insemination on genderratio in beef cattle. Theriogenology 1999;52:1035-1041. 9. Roelofs JB, Bouwman EB, Pedersen HG, et al: Effect of time of artificial insemination on embryo sex ratio in dairy cattle. Anim Reprod Sci 2006;93:366-371. 10. Aurich C, Schneider J: Sex determination in horses - current status and future perspectives. Anim Reprod Sci 2014;146:34-41. 11. Davies Morel MCG, Newcombe, JR, Holland SJ: Factors affecting gestation length in the Thoroughbred mare, Anim Reprod Sci 2002;74:175-185. 12. Drexler AP, Prado TM, Amelse LL, et al: Effects of time of insemination relative to time of ovulation on embryonic sex ratio in mares. Clinical Theriogenology 2016;8:357. 13. Gharagozlou F, Akbarinejad V, Youssefi R, et al: Effect of sire- associated factors on secondary sex ratio of offspring in equine. J Equine Vet Sci 2014;34:926-929. 14. Santos MM, Maia LL, Nobre DM, et al: Sex ratio of equine offspring is affected by the ages of the mare and stallion. Theriogenology 2015;84:1238-1245. Clinical Theriogenology 2022; 14: 31 Acknowledgement Dr. Douglas’ Theriogenology residency training program is sponsored by the American Kennel Club Canine Health Foun- dation and the Theriogenology Foundation. References 1. Johnston SD: Performing a complete canine semen evaluation in a small animal hospital. Vet Clin of North Am. Small Anim Pract 1991;21:545-550. 2. Eilts BE, Paccamonti DL, Pinto C: Artificial insemination in the dog. In: Root Kustritz MV, editor. The practical veterinarian: small animal theriogenology. St. Louis: Butterworth-Heineman; 2003. p. 61-95. 3. Johanisson E, Campana A, Luthi R, de Agostini A: Evaluation of ‘round cells’ in semen analysis: a comparative study. Hum Reprod Update 2000;6:404-412. 4. Johnston SD, Root Kustritz MV, Olson PNS: Canine and feline theriogenology. Philadelphia: WB Saunders: 2001. p. 287-306. 5. Barth AD, Bowman PA: The sequential appearance of sperm abnormalities after scrotal insulation or dexamethasone treatment in bulls. Can Vet J 1994;35:93-102. 6. Henning H, Masal C, Herr A, et al: Effect of short-term scrotal hyperthermia on spermatological parameters, testicular blood flow and gonadal tissue in dogs. Reprod Dom Anim 2014;49:145-157.