- Email: [email protected]
Natural outcome and ultrasonic identification of equine fetal twins
Theriogenology 41:1193-1199; 1994
NATURAL OUTCOME AND ULTRASONIC IDENTIFICATION OF EQUINE FETAL TWINS O.J. Ginther and P.G. Griffin Animal Health and Biomedical Sciences Veterinary Science Building University of Wisconsin-Madison Madison, WI 53706, USA Received for publication: September 30, 1993 Accepted: December 22, 1993 ABSTRACT The natural outcome of bilateral twins (one in each uterine horn) that were viable on Day 40 was studied in 15 pony mares. Fetuses were monitored by transrectal ultrasonography until the outcome was determinable. One fetus (2 mares) or both fetuses (8 mares) died (cessation of heartbeats) during Months 2 and 3. Loss of fetuses was externally observed in 4 additional mares during the late fetal stage (Months 8 to 11); 2 mares lost 1 fetus and 2 lost both fetuses. Birth of 2 live foals occurred in only 1 of 15 mares (7%). First day of detected apposition between portions of the 2 allantochorions was studied daily on Days 40 to 51 in 7 of the 15 mares. Apposition was first detected on mean Day 46.3 (standard deviation: ---3.1). In these 7 mares, death of 1 fetus (2 mares) or both fetuses (2 mares) occurred 1 or 2 days after the first day of apposition of allantochorions; no deaths occurred before the day of apposition. Each mare with loss of one fetus during this time gave birth to a living foal. The only structure indicative of twins that was consistently detected ultrasonically during all examinations was the common membrane resulting from the area of apposition between the 2 allantochorions. The common membrane is herein designated as the twin membrane because of its apparent diagnostic potential even in late pregnancy when the presence of the 2 fetuses may not be detected directly. Thickness of the twin membrane reached approximately 3 mm at Month 7 and thereafter gradually increased to 15 mm near the end of pregnancy. Two layers of the membrane (1 for each allantochorion) became distinct at Month 6 and Month 8 with 5.0 MHz and 3.5 MHz transducers, respectively; the 2 layers were separated by a hypoechogenic layer. Key words: Mares, twins, ultrasound Acknowledgements This research was supported by the College of Agricultural and Life Sciences, University of Wisconsin-Madison and by Equiculture, Inc., Cross Plains, WI. The authors thank Lisa Kulick and Tom Roberts for preparation of the figure, Karen Baucus for some of the ultrasound scanning, Wayne Bruenig and Cathy Miller for animal assistance, and Maria Radtke for manuscript preparation.
Copyright © 1994 Butterworth-Heinemann
Theriogenology
1194 INTRODUCTION
Twin equine embryos that are detected ultrasonically during the embryo-mobility phase (Days 11 to 15 after ovulation) become fixed in 1 uterine horn (unilateral fixation) in about 70% of mares (3). The unilaterally fixed twins usually (85% incidence) undergo embryo reduction (natural elimination of I member of the set) before the end of the embryo stage on Day 40. Bilaterally fixed twin embryos, however, do not undergo reduction and therefore enter the fetal stage (>Day 40) intact and viable. Unlike the common occurrence of reduction during the embryo stage, both members of twins that reach the fetal stage are likely to be lost or born weak and undersized (1,6,8). Most studies of twin fetuses have been limited to externally observed abortions or perinatal losses without preceding knowledge on the presence of viable twins. However, diagnosed Thoroughbred twins resulted in abortion (loss of both) in 73%, stillbirths in 11%, and birth of viable single or twin foals in 16% of the mares (1). In a study of farm transrectal palpation records (4), the presence of twins on Days 40 to 42 (n=16) resulted in abortion (loss of both fetuses) in 63%, birth of 2 foals (alive or dead) in 31%, and fetal reduction in only 6% of the mares. In another study (7), only 17 live foals were born from 130 mares carrying twins at approximately Day 42. Studies are lacking on the time of occurrence of fetal death in twin sets after Day 40 and on the reliability of various ultrasonic indicators of the presence of fetal twins, especially during late pregnancy. The present study was done to determine the natural outcome of fetal twins that were viable on Day 40 and to obtain preliminary information on the usefulness of various ultrasonic end points for the diagnosis of fetal twins. MATERIALS AND METHODS The mares were ponies or pony-horse crosses 4 to 18 yr of age and weighing 275 to 400 kg. Transrectal examinations of the reproductive tract were done with linear-array real-time ultrasonic scanners equipped with 3.5 and 5.0 MHz transducers (Tokyo Keiki, LS300, Products Group International, Inc., Boulder, CO). Daily examinations were done during the preovulatory period to determine day of ovulation (Day 0). Mares with bilateral twins (1 embryo in each uterine horn) were selected over a period of 2 yr during routine pregnancy diagnoses on Days 17 to 40. Fifteen mares with 2 viable embryos on Day 40, as determined by heartbeats, were assigned to the study. In 7 mares, examinations were done daily from Day 40 to at least Day 51 to determine the first day of detection of contact or direct apposition between the 2 allantochorions. Examinations, thereafter, were done at sporadic intervals (e.g., every 1 or 2 mo) until the outcome of the twin pregnancy was determinable. In the remaining 8 mares, ultrasonic examinations were made every 3 or 4 d from Day 40 until determination of the final outcome. Final outcomes were categorized as follows: 1) survival of both (birth of 2 live foals); 2) survival of 1 and death of the other (birth of 1 live foal and loss of the other at any time from Day 41 to the end of parturition); and 3) death of both. Death of a fetus was determined by absence of heartbeats during an ultrasonic examination with confirmation at subsequent examinations or by discharge of a dead fetus before the expected time of parturition (
Theriogenology
1195
Videotapes of all ultrasonic examinations were made for the 8 mares that were examined every 3 or 4 d The tapes were studied to obtain preliminary information on the accessibility and identifiability of twins during the late fetal stage and on the ultrasonic end points that seemed most reliable for consistent diagnoses. The common membrane representing the area of apposition between the 2 ailantochorions is designated herein as the twin membrane. RESULTS The outcomes of the Day-40 bilateral twin pregnancies and the month in which the outcome was determinable are shown (Table 1). One or both fetuses died (cessation of heartbeats) in 10 mares (67%) during the early fetal stage (Months 2 and 3). Loss of fetuses was externally observed in 4 additional mares during the late fetal stage (Months 8 to 11); 2 mares lost 1 fetus and 2 lost both fetuses. Birth of 2 live foals occurred in only 1 of 15 mares (7%). Mean first day of detected apposition between portions of the 2 allantochorions in 7 mares was Day 46,3 (standard deviation: +3.1). In these 7 mares, death of one fetus (2 mares) or both fetuses (2 mares) occurred 1 or 2 d after the first day of apposition of allantochorions; no deaths occurred before the day of apposition. Each mare with loss of I fetus during this time gave birth to a living foal. Videotapes taken every 3 to 4 d for the entire fetal stage (Days 40 to term) were available only for the mare that had 2 live foals. The only structure that was consistently detected during all examinations was the common membrane resulting from the area of apposition between the 2 allantochorions (twin membrane; Figure 1).
Table 1. Outcome of Day-40 bilateral twin pregnancies and month in which the outcome was determinable
Month
Days
2 3 4 5 6 7 8 9 10 11
40-60 61-90 91-120 121-150 151-180 181-210 211-240 241-270 271-300 >300
Survival of one, death of one
Survival of both
-. . . . . . . . . . . . . . . . . . . 1 1 (7%)
Death of both
2 . . . . . . . .
7 1
. . . .
. . . .
. . . .
.
.
.
1
2
1 --
4 (26%)
10 (67%)
Theriogenology
1196
.
*=
z
i t txa J¢ I , I
tm p
ir
62
p I ro
•"
?
..~
uc~
i 138
Figure 1.
D
237
E
Ultrasonograms of the twin membrane (apposed layers of the 2 allantochorions) at various days of pregnancy. A 5.0 MHz transducer was used for images A, B, and D and a 3.5 MHz transducer was used for images C, E, and F; scale is in centimeters. Sometimes both amnions as well as the twin membrane were detectable (D). With a 3.5 MHz transducer, the twin membrane began to show a bilaminar appearance at Month 8 (E) and later developed a distinct bilaminar appearance (F). am=amnion; f=fetus; tm=twin membrane; uc=umbilical cord.
Theriogenology
1197 DISCUSSION
The 5.0 MHz transducer was superior to the 3.5 MHz transducer during Month 2 (Days 40 to 60); both fetuses were readily detected and all parts were accessible for detailed study. Beginning on approximately Day 70, the height of the allantoic fluid exceeded the depth of image display for the 5.0 MHz transducer. Often all or a major portion of 1 or both fetuses was not imaged and heartbeats were detectable in both fetuses only occasionally. In contrast, the heart was imaged with the 3.5 MHz transducer in both fetuses until the end of Month 7 (Day 210). In the 1 twin set monitored every 3 to 4 days after Month 7, heartbeats were detected for both fetuses in all examinations during Months 2 to 7, in 6 of 13 examinations during Months 8 and 9 and in 0 of 10 examinations during Months 10 and 11. Thus, even with the 3.5 MHz transducer, the hearts began to be inaccessible after Month 7 because of the increasing size of the fetuses. This study was done in ponies; reachability of fetal parts by transrectal ultrasonography likely would be even more difficult in horses. The eyes seemed good indicators of the relative size of the 2 fetuses. After Month 7, however, the eyes of both fetuses were imaged adequately for measurement in only 47% of examinations. Intrinsic activity and mobility (changes in location, recumbency and presentation) of the fetuses seemed similar to the extensive activity and mobility reported for singletons (5). Closure of both uterine horns during Months 5 to 7, as reported for singleton pregnancies (2), was also common. The uterine horns were not examined after Month 7. Death of fetuses occurred most frequently during Month 2 as indicated by absence of heartbeats; 16 of 24 deaths (67%) occurred in Month 2 and early death of both fetuses in an additional mare occurred sometime between Day 55 and Day 96. The 18 deaths (60% of 30 fetuses) during the early fetal stage involved loss of both fetuses (8 mares) and loss of 1 of the 2 fetuses (2 mares). In mares with frequent examinations and death of both fetuses during the early fetal stage, cessation of heartbeats was detected on the same day in 5 mares and on different days in 2 mares. In 2 mares, absence of heartbeats in 1 fetus was first detected on Days 44 and 51, respectively; the fetus was terminated naturally and effectively without apparent interference with development of the other fetus similar to the efficiency of embryo reduction (3). A single live foal was born in each of the 2 mares with fetal reduction. The 13% incidence of fetal reduction seems similar to the 6% incidence found in an examination of farm transrectal palpation records (4). In the 7 mares with daily examinations on Days 40 to 51, the 2 allantochorions reached the uterine body resulting in direct apposition between the 2 conceptuses (formation of twin membrane) on mean Day 46 (range: Days 42 to 51). Cessation of fetal heartbeats in these 7 mares during Days 40 to 51 occurred in 6 fetuses in 4 mares 1 or 2 days after the detected apposition of allantochorions. No deaths occurred before apposition. This temporal relationship suggests that apposition led to a mechanism (nature unknown) that resulted in death of fetuses within 2 days in 4 of 7 mares. Apparently, therefore, the time of apposition of the allantochorions or formation of the twin membrane is a critical period in the survival of fetal twins. Only 1 of 15 mares (7%) with viable bilateral twins on Day 40 gave birth to 2 living foals. Both foals were strong, but smaller than contemporary singletons. In total, 6 live foals resulted from the 30 fetuses (20%) that were present at Day 40 which is similar to the reported 17% incidence for horse mares diagnosed with twins on Day 42 (7). Two mares had 1 living
Theriogenology
1198
foal accompanied by a dead fetus; the living foals were undersized and weak. In 2 other mares, late loss of both fetuses occurred during Months 8 and 10. The results are similar to those reported (1,6-8) in that externally observed fetuses and foals were dead, undersized, or weak; that is the rate of perinatal losses was high. It appears that fetuses which survive the critical period involving membrane apposition during the early fetal stage are again challenged during the late fetal stage; these later losses likely involve nutritive competition between the 2 conceptuses (6). The membrane representing the apposed surfaces of the 2 allantochorions (twin membrane) was readily detected during all ultrasonic examinations for all twin sets after the first day of detected apposition. The twin membrane was the only detected structure that was diagnostic of twins even in late pregnancy. The 2 layers of the twin membrane were indistinct initially, and the structure appeared unilayered similar to the amnion until Month 6 with the 5.0 MHz transducer; thereafter, the bilaminar nature of the membrane became clear (2 hyperechogenic layers with a hypoechogenic central layer). The double layering became faintly detectable with the 3.5 MHz transducer during Month 8 and thereafter became increasingly obvious. The thickness of the twin membrane was initially less than 2 mm; therefore short segments were not detectable and the membrane had the appearance of a broken line. The thickness reached approximately 3 mm at Month 7 and thereafter gradually increased to 15 mm near the end of pregnancy. The twin membrane was especially prominent in late pregnancy because of its thickness and distinct double layers separated by a hypoechogenic layer. The characteristic layering was a diagnostic convenience in late pregnancy, because sometimes only short segments of the membrane were found. Another advantage of the twin membrane as a diagnostic criterion was that it was accessible because a portion of it was located in the dorsal area of the large gravid uterus of late pregnancy. The amnion was closely applied to the twin membrane in some areas, and imparted a serrated (loose) surface to the twin membrane. During the early fetal stage when the amniotic and twin membranes were similar in thickness, a distinction between them was readily made; the amniotic membranes were tortuous (loose) and close in proximity to fetal parts. Often areas were found where 3 membranes were visible in a single image --- the 2 amniotic membranes on each side of the twin membrane. Later (Month 8 with the 3.5 MHz transducer) the twin membrane was identifiable by its double-layered appearance unlike any other membrane. Although the number of late twin sets was limited in the present study, the consistent results provide at least a preliminary indication that the characteristics of the twin membrane provide a reliable indicator of the presence of twins. No other indicator (e.g., detection and identification of all or parts of 2 fetuses) seemed to approach the usefulness of the twin membrane. A study of the diagnostic usefulness of the twin-membrane criterion under farm conditions is needed. REFERENCES 1. DeskurS. Twinningin Thoroughbred mares in Poland. Theriogenology1985;23:711-718. 2. Ginther OJ. Equine fetal kinetics: Allantoic-fluidshifts and uterine-horn closures. Theriogenology 1993;40:241-256. 3. GintherOJ. Twin embryos in mares II: Post fixationembryo reduction. EquineVet J 1989;21:171-174. 4. GintherOJ, Douglas RH. The outcome of twin pregnanciesin mares. Theriogenology1982;18:237-244. 5. GintherOJ, Griffin PG. Equinefetal kinetics: Presentationand location. Theriogenology1993;40:1-11.
Theriogenology
1199
6: Jeffcott LB, Whitwell K. Twinning as a cause of foetal and neonatal loss in the Thoroughbred mare. J Comp Path 1973;83:91-106. 7. Pascoe RR. Methods for the treatment of twin pregnancy in the mare. Equine Vet J 1983;15:40-42. 8. Whitwell KE. Investigations into fetal and neonatal losses in the horse. Vet Clin N A m e r : Large Anim Prac 1980;2:313-331.
NATURAL OUTCOME AND ULTRASONIC IDENTIFICATION OF EQUINE FETAL TWINS O.J. Ginther and P.G. Griffin Animal Health and Biomedical Sciences Veterinary Science Building University of Wisconsin-Madison Madison, WI 53706, USA Received for publication: September 30, 1993 Accepted: December 22, 1993 ABSTRACT The natural outcome of bilateral twins (one in each uterine horn) that were viable on Day 40 was studied in 15 pony mares. Fetuses were monitored by transrectal ultrasonography until the outcome was determinable. One fetus (2 mares) or both fetuses (8 mares) died (cessation of heartbeats) during Months 2 and 3. Loss of fetuses was externally observed in 4 additional mares during the late fetal stage (Months 8 to 11); 2 mares lost 1 fetus and 2 lost both fetuses. Birth of 2 live foals occurred in only 1 of 15 mares (7%). First day of detected apposition between portions of the 2 allantochorions was studied daily on Days 40 to 51 in 7 of the 15 mares. Apposition was first detected on mean Day 46.3 (standard deviation: ---3.1). In these 7 mares, death of 1 fetus (2 mares) or both fetuses (2 mares) occurred 1 or 2 days after the first day of apposition of allantochorions; no deaths occurred before the day of apposition. Each mare with loss of one fetus during this time gave birth to a living foal. The only structure indicative of twins that was consistently detected ultrasonically during all examinations was the common membrane resulting from the area of apposition between the 2 allantochorions. The common membrane is herein designated as the twin membrane because of its apparent diagnostic potential even in late pregnancy when the presence of the 2 fetuses may not be detected directly. Thickness of the twin membrane reached approximately 3 mm at Month 7 and thereafter gradually increased to 15 mm near the end of pregnancy. Two layers of the membrane (1 for each allantochorion) became distinct at Month 6 and Month 8 with 5.0 MHz and 3.5 MHz transducers, respectively; the 2 layers were separated by a hypoechogenic layer. Key words: Mares, twins, ultrasound Acknowledgements This research was supported by the College of Agricultural and Life Sciences, University of Wisconsin-Madison and by Equiculture, Inc., Cross Plains, WI. The authors thank Lisa Kulick and Tom Roberts for preparation of the figure, Karen Baucus for some of the ultrasound scanning, Wayne Bruenig and Cathy Miller for animal assistance, and Maria Radtke for manuscript preparation.
Copyright © 1994 Butterworth-Heinemann
Theriogenology
1194 INTRODUCTION
Twin equine embryos that are detected ultrasonically during the embryo-mobility phase (Days 11 to 15 after ovulation) become fixed in 1 uterine horn (unilateral fixation) in about 70% of mares (3). The unilaterally fixed twins usually (85% incidence) undergo embryo reduction (natural elimination of I member of the set) before the end of the embryo stage on Day 40. Bilaterally fixed twin embryos, however, do not undergo reduction and therefore enter the fetal stage (>Day 40) intact and viable. Unlike the common occurrence of reduction during the embryo stage, both members of twins that reach the fetal stage are likely to be lost or born weak and undersized (1,6,8). Most studies of twin fetuses have been limited to externally observed abortions or perinatal losses without preceding knowledge on the presence of viable twins. However, diagnosed Thoroughbred twins resulted in abortion (loss of both) in 73%, stillbirths in 11%, and birth of viable single or twin foals in 16% of the mares (1). In a study of farm transrectal palpation records (4), the presence of twins on Days 40 to 42 (n=16) resulted in abortion (loss of both fetuses) in 63%, birth of 2 foals (alive or dead) in 31%, and fetal reduction in only 6% of the mares. In another study (7), only 17 live foals were born from 130 mares carrying twins at approximately Day 42. Studies are lacking on the time of occurrence of fetal death in twin sets after Day 40 and on the reliability of various ultrasonic indicators of the presence of fetal twins, especially during late pregnancy. The present study was done to determine the natural outcome of fetal twins that were viable on Day 40 and to obtain preliminary information on the usefulness of various ultrasonic end points for the diagnosis of fetal twins. MATERIALS AND METHODS The mares were ponies or pony-horse crosses 4 to 18 yr of age and weighing 275 to 400 kg. Transrectal examinations of the reproductive tract were done with linear-array real-time ultrasonic scanners equipped with 3.5 and 5.0 MHz transducers (Tokyo Keiki, LS300, Products Group International, Inc., Boulder, CO). Daily examinations were done during the preovulatory period to determine day of ovulation (Day 0). Mares with bilateral twins (1 embryo in each uterine horn) were selected over a period of 2 yr during routine pregnancy diagnoses on Days 17 to 40. Fifteen mares with 2 viable embryos on Day 40, as determined by heartbeats, were assigned to the study. In 7 mares, examinations were done daily from Day 40 to at least Day 51 to determine the first day of detection of contact or direct apposition between the 2 allantochorions. Examinations, thereafter, were done at sporadic intervals (e.g., every 1 or 2 mo) until the outcome of the twin pregnancy was determinable. In the remaining 8 mares, ultrasonic examinations were made every 3 or 4 d from Day 40 until determination of the final outcome. Final outcomes were categorized as follows: 1) survival of both (birth of 2 live foals); 2) survival of 1 and death of the other (birth of 1 live foal and loss of the other at any time from Day 41 to the end of parturition); and 3) death of both. Death of a fetus was determined by absence of heartbeats during an ultrasonic examination with confirmation at subsequent examinations or by discharge of a dead fetus before the expected time of parturition (
Theriogenology
1195
Videotapes of all ultrasonic examinations were made for the 8 mares that were examined every 3 or 4 d The tapes were studied to obtain preliminary information on the accessibility and identifiability of twins during the late fetal stage and on the ultrasonic end points that seemed most reliable for consistent diagnoses. The common membrane representing the area of apposition between the 2 ailantochorions is designated herein as the twin membrane. RESULTS The outcomes of the Day-40 bilateral twin pregnancies and the month in which the outcome was determinable are shown (Table 1). One or both fetuses died (cessation of heartbeats) in 10 mares (67%) during the early fetal stage (Months 2 and 3). Loss of fetuses was externally observed in 4 additional mares during the late fetal stage (Months 8 to 11); 2 mares lost 1 fetus and 2 lost both fetuses. Birth of 2 live foals occurred in only 1 of 15 mares (7%). Mean first day of detected apposition between portions of the 2 allantochorions in 7 mares was Day 46,3 (standard deviation: +3.1). In these 7 mares, death of one fetus (2 mares) or both fetuses (2 mares) occurred 1 or 2 d after the first day of apposition of allantochorions; no deaths occurred before the day of apposition. Each mare with loss of I fetus during this time gave birth to a living foal. Videotapes taken every 3 to 4 d for the entire fetal stage (Days 40 to term) were available only for the mare that had 2 live foals. The only structure that was consistently detected during all examinations was the common membrane resulting from the area of apposition between the 2 allantochorions (twin membrane; Figure 1).
Table 1. Outcome of Day-40 bilateral twin pregnancies and month in which the outcome was determinable
Month
Days
2 3 4 5 6 7 8 9 10 11
40-60 61-90 91-120 121-150 151-180 181-210 211-240 241-270 271-300 >300
Survival of one, death of one
Survival of both
-. . . . . . . . . . . . . . . . . . . 1 1 (7%)
Death of both
2 . . . . . . . .
7 1
. . . .
. . . .
. . . .
.
.
.
1
2
1 --
4 (26%)
10 (67%)
Theriogenology
1196
.
*=
z
i t txa J¢ I , I
tm p
ir
62
p I ro
•"
?
..~
uc~
i 138
Figure 1.
D
237
E
Ultrasonograms of the twin membrane (apposed layers of the 2 allantochorions) at various days of pregnancy. A 5.0 MHz transducer was used for images A, B, and D and a 3.5 MHz transducer was used for images C, E, and F; scale is in centimeters. Sometimes both amnions as well as the twin membrane were detectable (D). With a 3.5 MHz transducer, the twin membrane began to show a bilaminar appearance at Month 8 (E) and later developed a distinct bilaminar appearance (F). am=amnion; f=fetus; tm=twin membrane; uc=umbilical cord.
Theriogenology
1197 DISCUSSION
The 5.0 MHz transducer was superior to the 3.5 MHz transducer during Month 2 (Days 40 to 60); both fetuses were readily detected and all parts were accessible for detailed study. Beginning on approximately Day 70, the height of the allantoic fluid exceeded the depth of image display for the 5.0 MHz transducer. Often all or a major portion of 1 or both fetuses was not imaged and heartbeats were detectable in both fetuses only occasionally. In contrast, the heart was imaged with the 3.5 MHz transducer in both fetuses until the end of Month 7 (Day 210). In the 1 twin set monitored every 3 to 4 days after Month 7, heartbeats were detected for both fetuses in all examinations during Months 2 to 7, in 6 of 13 examinations during Months 8 and 9 and in 0 of 10 examinations during Months 10 and 11. Thus, even with the 3.5 MHz transducer, the hearts began to be inaccessible after Month 7 because of the increasing size of the fetuses. This study was done in ponies; reachability of fetal parts by transrectal ultrasonography likely would be even more difficult in horses. The eyes seemed good indicators of the relative size of the 2 fetuses. After Month 7, however, the eyes of both fetuses were imaged adequately for measurement in only 47% of examinations. Intrinsic activity and mobility (changes in location, recumbency and presentation) of the fetuses seemed similar to the extensive activity and mobility reported for singletons (5). Closure of both uterine horns during Months 5 to 7, as reported for singleton pregnancies (2), was also common. The uterine horns were not examined after Month 7. Death of fetuses occurred most frequently during Month 2 as indicated by absence of heartbeats; 16 of 24 deaths (67%) occurred in Month 2 and early death of both fetuses in an additional mare occurred sometime between Day 55 and Day 96. The 18 deaths (60% of 30 fetuses) during the early fetal stage involved loss of both fetuses (8 mares) and loss of 1 of the 2 fetuses (2 mares). In mares with frequent examinations and death of both fetuses during the early fetal stage, cessation of heartbeats was detected on the same day in 5 mares and on different days in 2 mares. In 2 mares, absence of heartbeats in 1 fetus was first detected on Days 44 and 51, respectively; the fetus was terminated naturally and effectively without apparent interference with development of the other fetus similar to the efficiency of embryo reduction (3). A single live foal was born in each of the 2 mares with fetal reduction. The 13% incidence of fetal reduction seems similar to the 6% incidence found in an examination of farm transrectal palpation records (4). In the 7 mares with daily examinations on Days 40 to 51, the 2 allantochorions reached the uterine body resulting in direct apposition between the 2 conceptuses (formation of twin membrane) on mean Day 46 (range: Days 42 to 51). Cessation of fetal heartbeats in these 7 mares during Days 40 to 51 occurred in 6 fetuses in 4 mares 1 or 2 days after the detected apposition of allantochorions. No deaths occurred before apposition. This temporal relationship suggests that apposition led to a mechanism (nature unknown) that resulted in death of fetuses within 2 days in 4 of 7 mares. Apparently, therefore, the time of apposition of the allantochorions or formation of the twin membrane is a critical period in the survival of fetal twins. Only 1 of 15 mares (7%) with viable bilateral twins on Day 40 gave birth to 2 living foals. Both foals were strong, but smaller than contemporary singletons. In total, 6 live foals resulted from the 30 fetuses (20%) that were present at Day 40 which is similar to the reported 17% incidence for horse mares diagnosed with twins on Day 42 (7). Two mares had 1 living
Theriogenology
1198
foal accompanied by a dead fetus; the living foals were undersized and weak. In 2 other mares, late loss of both fetuses occurred during Months 8 and 10. The results are similar to those reported (1,6-8) in that externally observed fetuses and foals were dead, undersized, or weak; that is the rate of perinatal losses was high. It appears that fetuses which survive the critical period involving membrane apposition during the early fetal stage are again challenged during the late fetal stage; these later losses likely involve nutritive competition between the 2 conceptuses (6). The membrane representing the apposed surfaces of the 2 allantochorions (twin membrane) was readily detected during all ultrasonic examinations for all twin sets after the first day of detected apposition. The twin membrane was the only detected structure that was diagnostic of twins even in late pregnancy. The 2 layers of the twin membrane were indistinct initially, and the structure appeared unilayered similar to the amnion until Month 6 with the 5.0 MHz transducer; thereafter, the bilaminar nature of the membrane became clear (2 hyperechogenic layers with a hypoechogenic central layer). The double layering became faintly detectable with the 3.5 MHz transducer during Month 8 and thereafter became increasingly obvious. The thickness of the twin membrane was initially less than 2 mm; therefore short segments were not detectable and the membrane had the appearance of a broken line. The thickness reached approximately 3 mm at Month 7 and thereafter gradually increased to 15 mm near the end of pregnancy. The twin membrane was especially prominent in late pregnancy because of its thickness and distinct double layers separated by a hypoechogenic layer. The characteristic layering was a diagnostic convenience in late pregnancy, because sometimes only short segments of the membrane were found. Another advantage of the twin membrane as a diagnostic criterion was that it was accessible because a portion of it was located in the dorsal area of the large gravid uterus of late pregnancy. The amnion was closely applied to the twin membrane in some areas, and imparted a serrated (loose) surface to the twin membrane. During the early fetal stage when the amniotic and twin membranes were similar in thickness, a distinction between them was readily made; the amniotic membranes were tortuous (loose) and close in proximity to fetal parts. Often areas were found where 3 membranes were visible in a single image --- the 2 amniotic membranes on each side of the twin membrane. Later (Month 8 with the 3.5 MHz transducer) the twin membrane was identifiable by its double-layered appearance unlike any other membrane. Although the number of late twin sets was limited in the present study, the consistent results provide at least a preliminary indication that the characteristics of the twin membrane provide a reliable indicator of the presence of twins. No other indicator (e.g., detection and identification of all or parts of 2 fetuses) seemed to approach the usefulness of the twin membrane. A study of the diagnostic usefulness of the twin-membrane criterion under farm conditions is needed. REFERENCES 1. DeskurS. Twinningin Thoroughbred mares in Poland. Theriogenology1985;23:711-718. 2. Ginther OJ. Equine fetal kinetics: Allantoic-fluidshifts and uterine-horn closures. Theriogenology 1993;40:241-256. 3. GintherOJ. Twin embryos in mares II: Post fixationembryo reduction. EquineVet J 1989;21:171-174. 4. GintherOJ, Douglas RH. The outcome of twin pregnanciesin mares. Theriogenology1982;18:237-244. 5. GintherOJ, Griffin PG. Equinefetal kinetics: Presentationand location. Theriogenology1993;40:1-11.
Theriogenology
1199
6: Jeffcott LB, Whitwell K. Twinning as a cause of foetal and neonatal loss in the Thoroughbred mare. J Comp Path 1973;83:91-106. 7. Pascoe RR. Methods for the treatment of twin pregnancy in the mare. Equine Vet J 1983;15:40-42. 8. Whitwell KE. Investigations into fetal and neonatal losses in the horse. Vet Clin N A m e r : Large Anim Prac 1980;2:313-331.