Migration of Embryos in the Anastomosed Uterine Horns of the Pig' 0. C. MILLER AND P. J. DZIUK Department of Animal Science, University of Illinois, Urbana, Illinois ABSTRACT One ovary was removed from 28 gilts and the uterus was transected, closed and anastornosed in one of four patterns. In three of the patterns the uterine horns were jointed near the uterotubal junction and in the fourth pattern the uterus formed one continuous tube from one oviduct to the cervix. Gilts were mated, then killed at 15 to 60 days of gestation when fetuses were located. The proportion of gilts pregnant was lower (P < 0.05) in the two groups requiring passage of sperm through an anastomosis than in the two groups where fertilization did not require that sperm traverse an anastomosis. Ten gilts were pregnant when killed and embryos were distributed in both horns in each gilt. Embryos can pass through an anastomosis and migrate and are spaced quite evenly even after disturbance of the normal anatomical relationships between uterine horns. Intrauterine migration and mixing of embryos is well known to occur in the pig (Corner, '21; Dziuk, Polge and Rowson, '64; Dhindsa, Dziuk and Norton, '67). The normal pig uterus can be considered as one unit with the body connecting the two horns. Krehbiel ('46) surgically anastomosed the two separate uterine horns of the rat and obtained intrauterine migration in some types of joinings. He concluded that in the normal duplex rat uterus, embryo migration was controlled by the individual horn whereas the anastomosed uterus usually acted as one unit. The following experiments were performed to determine whether disturbing normal relationships between the two horns of the pig uterus by anastomoses would affect distribution and migration of embryos. as in type I but sperm had to traverse the anastomosis to reach the oviduct. The two horns formed one continuous tube from oviduct to cervix in type 111. In type IV the cervical end of one horn was anastomosed to the upper end of the other horn. There were eight gilts each of types I and I1 and six gilts each of types I11 and IV. After laparotomy the gilts were penned with a fertile boar in groups of 10 or 12 and mating allowed. Gilts were killed at times ranging from 15 to 60 days after mating when corpora lutea were counted, pregnancy determined and embryos or fetuses identified and located. The uteri of nonpregnant gilts were filled with water to check the anastomoses for patency and leaks. Data were analyzed for statistical significance by chi-square. MATERIALS AND METHODS RESULTS The experimental animals were 28 sexuThe ovaries at autopsy had a mean of 13 ally mature gilts of Duroc, Yorkshire or corpora lutea, confirming previous observaDuroc x Yorkshire breeding. Laparotomies tions that one remaining ovary will comwere done aseptically while the animals pensate (Dhindsa et al., '67). Ten of the were under general anesthesia (Dhindsa 28 gilts were pregnant when killed. Eight et al., '67). One ovary was removed and of the ten had a mean of seven fetuses and the uterus anastomosed or closed with con- the other two had elongated chorionic tinuous sutures of black silk, size 0, in one membranes characteristic of day 15 of gesof four patterns (I, 11, I11 or IV) as shown tation making precise count of embryos diagrammatically in figure 1. In type I the difficult. The proportion of gilts pregnant uterine horns were joined at the ovarian in each treatment group is shown in table end, sperm could go from cervix to oviduct 1 This investigation was supported in part by Public Health Service Research grant HD01987 from the without passing through an anastomosis. National Type I1 gilts had the uterine horns joined velopment.Institute of Child Health and Human DeANAT. REC., 162: 137-138. 137 138 0. C. MILLER A N D P. J. DZIUK TABLE 1 Pregnancy in gilts with uterine anastomoses Type of Number of gilts Number pregnant I 8 6 14 3 anastomosis IV I plus 1v I1 I11 11 DlUS 111 5 8 1 1 21 8 6 14 1 Lower than types I plus N (P < 0.05). 1. There were no significant differences between types I1 and I11 nor between I and IV in the proportion pregnant. A greater proportion of types I and IV combined, were pregnant than types I1 and I11 combined (P < 0.05). The distribution of fetuses is shown in figure 1. At least one fetus was in each uterine horn in all pregnant animals. Embryos and sperm could both pass through anastomoses. There was no indication that embryos had an affinity for one horn or the other. Horns ending blindly at one end did not preclude pregnancy. \Severed And Closed TYPE I TYPE Ip TYPE I TYPE Ip: TYPE TYPE IL: IE TYPE 3 l I TYPE Ip. Fig. 1 Location of fetuses following transection and anastomosis of uteri of gilts. The uteri of some of the nonpregnant gdts contained accumulated fluid and pus which probably prevented pregnancy. DISCUSSION Disruption of the normal anatomical relationships between the two uterine horns and establishment of abnormal ones did not prevent pregnancy. Blind uterine horns per se were compatible with pregnancy which is Merent from naturally occuring blind horns that can not be pregnant and may exert an indirect effect preventing pregnancy from being maintained in the normal horn (Nalbandov, '52). Sperm could pass a n anastomosis but perhaps the difference between groups I and N combined as compared to groups I1 and 111 combined in conception rate was due to more frequent failure of sperm transport across anastomoses. Embryos readily migrated across anastomoses and there was no indication that conception rate was affected by the frequency of failure of embryo migration. Embryos were quite evenly spaced regardless of which end of the horn the embryo entered, confirming the previous observation (Dziuk et al., '64) that uterine migration is not unidirectional. Distribution of embryos seemed normal even though the embryos entered midway between the two ends of the uterus (types I, I1 and IV) rather than at both ends of the uterus as would occur normally. This observation suggests that spacing and migration are random processes and that the pig uterus can be regarded as one continuous unit in spite of abnormal anatomical relationships such as were created in this study. Work reported by Dhindsa and Dziuk ('68) has shown that when one half the uterus is unoccupied before day 14 pregnancy does not continue. Whether failure of maintenance of a unilateral pregnancy was the reason 18 gilts were nonpregnant in this study is not known. All pregnant gdts in this study had at least one fetus in each uterine horn. LITERATURE CITED Comer, G. W. 1921 Internal migration of the ovum. Johns Hopkins Hospital Bull. NO. 32: 78-83. Dhindsa, D. S., and P. J. Dziuk 1968 Influence of varying the proportion of uterus occupied by embryos on maintenance of pregnancy in the pig. J. Animal Sci., 27: 668-672. Dhindsa, D. S., P. J. Dziuk and H. W. Norton 1967 Time of transuterine migration and distribution of embryos in the pig. Anat. Rec., 159: 325-330. Dziuk, P. J., C. Polge and L. E. A. Rowson 1964 Intrauterine migration and mixing of embryos in swine following egg transfer. J. Animal Sci., 23: 3 7 4 2 . Krehbiel, R. H. 1946 Distribution of ova in the combined uteri of unilaterally ovariectomized rats. Anat. Rec., 96: 323-340. Nalbandov, A. V. 1952 Anatomic and endocrine causes of sterility of female swine. Fert. Steril., 3: 100-120.