The establishment and loss of twins often results in a mare that is barren for a year and the associated economic loss for the owner. The incidence of twin births has been documented as occurring in 1–2% of the equine population (Ginther, 1979), with twinning accounting for 6–30% of abortions in the mare (Jeffcott and Whitwell, 1973; Roberts, 1986; Giles et al, 1993). When twins are present, gestation proceeds normally until the conceptuses begin to compete for endometrial contact area to optimise uteroplacental nutritional exhange. The death or compromise of one fetus affects the remaining fetus, initiating changes in placental hormones leading to premature mammary gland development. Subsequent abortion due to placental insufficiency occurs between 5–9 months of gestation (Roberts, 1982). If abortion occurs after the eighth month, the mare's fertility may be affected negatively the following year. Fortunately, with the use of ultrasound there is better understanding of the mechanisms involved in twinning, earlier identification and improved approaches to twin reduction (Chevalier and Palmer, 1982; Simpson et al, 1982; Bowman, 1987).
Management of twin pregnancies is most successful 13–15 days post-ovulation, when the embryonic vesicles are still mobile (Palmer and Driancourt, 1980; Ginther, 1986; 1989a). Natural twin reduction does not occur before day 11, and it is negligible between days 11 and 15 (Ginther, 1986; Pascoe et al, 1987; Ginther and Bergfelt, 1988). Therefore, twin pregnancies that are detected during the mobility phase (up to day 15) are best managed by manually crushing one embryonic vesicle (Roberts, 1982). After fixation of the embryonic vesicles (>day 16 post ovulation), the success rate of reducing twins to a viable singleton varies greatly among procedures (Table 1).
Table 1. Days of gestation per reduction procedure and success rate of live born singleton foal.
| Procedure | Stage of gestation post ovulation | Success rate |
|---|---|---|
| Manual reduction | <16 days | 93–100% |
| Manual reduction | >20 days | 28–90% |
| Natural reduction | 16–45 days | 18–82% |
| Transvaginal ultrasound-guided aspiration/puncture/injection | 16–62 days | 33–49% |
| Oscillation | 45–55 days | 80% |
| Thoracic compression | 55–65 days | Unknown |
| Craniocervical dislocation | 65–115 days | 63% |
| Transcutaneous ultrasound-guided injection | 130–150 days | 38–56% |
Natural reduction
Natural reduction occurs when one embryo is eliminated without any intervention. The actual mechanism of natural reduction is not known (Ginther, 1989b). Fixation of the embryonic vesicles unilaterally or bilaterally will affect the incidence of natural reduction. Studies have suggested that close apposition of the embryonic vesicles appeared to result in loss of contact between the endometrial and trophoblastic surfaces of the vesicle leading to decreased nourishment and ability to grow (Ginther, 1982; Ginther and Griffin, 1994). Most unilateral reductions occurred before day 26 and the overall incidence of natural reduction, resulting in a single conceptus was found to be 64% (Ginther, 1989b). However, the probability of natural reduction decreases as gestational age increases to day 40, with 53% of unilateral reductions occurring before day 20 and 82% before day 30 (Ginther, 1989a; 1989b). Factors that appear to influence natural reduction include (Ginther, 1989a; 1989b):
- Relative orientation of the embryonic vesicles
- Synchronicity of the ovulatory follicles
- Size disparity between the vesicles.
If the embryo proper is adjacent to the other vesicle it is more likely to be reduced; in addition, if asynchronous ovulations occur with one embryonic vesicle being larger, movement and growth may be impeded leading to a higher likelihood of reduction.
Manual reduction
Manual reduction is most successfully performed before fixation of the embryonic vesicles (<16 days post-ovulation). When twin conceptuses are observed after fixation, manual reduction can be attempted, however is difficult without disrupting both. If separation or isolation of one is possible, 90% of unilateral twins can be manually disrupted between 17–20 days (Bowman, 1987; Pascoe and Stover, 1989). Three-quarters of bilateral twins may be successfully reduced to a singleton pregnancy by disrupting one vesicle before 30 days of gestation. It is important to only disrupt, not rupture the vesicle by slipping the membranes, creating flocculence within the vesicle fluid. If flocculence or irregularity occurs within the vesicle, most will be resorbed within 2–3 days. Repeat ultra-sound is warranted to ascertain continued growth of the remaining vesicle and resorption of the disrupted embryonic vesicle. Manual manipulation transrectally, with disrupting, not rupturing, one of the vesicles between 28 and 42 days, results in 28% of the mares having a single viable foal (McKinnon and Rantanen, 1998).
Transvaginal ultrasound-guided reduction
Selective reduction of dizygotic twin pregnancies using transvaginal ultrasound-guided aspiration, puncture or injection has been examined in mares (Bracher et al, 1993; Macpherson and Reimer, 2000; Journée et al, 2013). The technique involves a 5 or 7.5 MHz transvaginal ultrasound transducer placed inside the mare's anterior vagina. The operator's arm is then placed in the rectum for manipulation of the reproductive tract. The pregnancy is secured, and the transducer is manipulated until the pregnancy is imaged on the ultrasound. An assistant passes a sterile needle through the channel in the transducer casing. A sharp stab of the needle is made for passage of the needle through the vaginal and uterine walls into the yolk or allantoic space. After ultrasonographic identification of the echogenic needle tip in the yolk or allantoic space, a 60 ml luer-tip syringe or suction pump is connected to the needle and the embryonic vesicle is punctured, fluid aspirated or procaine penicillin, potassium chloride or amikacin injected. For transvaginal ultrasound-guided aspiration, orientation influences when aspiration is discontinued, and the volume of fluid removed. For unilateral twins, aspiration is discontinued when there is danger of aspirating the fetal membranes, the conceptus can no longer be visualised, or it is no longer possible to obtain fluid. When performing the procedure on a bilateral twin, complete fluid evacuation is ideal. Trauma to the treated fetus is not a concern with bilateral twins and may be advantageous.
The success rate of dizygotic twin reduction through transvaginal ultrasound-guided aspiration is highly variable and dependent on many factors. There appears to be an advantage to perform the procedure before 36 days of gestation, particularly in the case of unilateral twins. When comparing transvaginal ultrasound-guided aspiration to transvaginal ultrasound-guided aspiration combined with puncture of the fetus, a significant negative effect was detected on the live foaling rate (23%) for mares in which transvaginal ultrasound-guided aspiration was combined with fetal puncture, versus (61%) for mares that had transvaginal ultrasound-guided aspiration performed alone (Klewitz et al, 2013). This effect may have resulted from increased manipulation of the uterus to ensure precise positioning of the conceptus against the probe to perform multiple punctures without movement. Another larger study determined that 33% of transvaginal ultrasound-guided reductions resulted in a single live foal but there was no significant difference based on fetal puncture (42.9%) versus fluid aspiration (29.3%) (Klewitz et al, 2013). The authors concluded that fetal puncture does not offer significant advantages over fluid aspiration.
Transvaginal ultrasound-guided aspiration for unilateral dizygotic twins has significant limitations due to the close proximity of the embryos or fetuses to the associated membranes. Penetration or puncture of the adjacent vesicle can occur if the fetal membranes are not seen in the imaging plane. Additionally, when fluid is withdrawn from a unilateral twin, the adjacent vesicle tends to pull away from the endometrium. Fluid may also leak from the incompletely evacuated vesicle causing the vesicle membranes to separate from the endometrium (Raggio et al, 2008; Journée et al, 2013).
With bilateral twin pregnancies, there is significantly less likelihood that penetration of the conceptus and surrounding membranes will occur. Therefore, more aggressive aspiration and manipulation of the fluid and fetus may be performed. Also, the time limitation seen with unilateral twins is not as stringent. Age, parity, size of the mare, position and tone of the uterus can all negatively impact the success of the procedure. Therefore, performing the procedure in older, multiparous mares after 45 or 50 days can be challenging and less successful.
Puncture of the embryonic vesicle or injection of amikacin sulphate (1 ml) has also been attempted at 28 days of gestation. Although not significantly different from transvaginal ultra-sound-guided puncture, transvaginal ultrasound-guided injection successfully reduced 40% of twin pregnancies (Raggio et al, 2008). By not aspirating the yolk sac or allantoic fluid, the downfalls of transvaginal ultrasound-guided aspiration may be diminished. This technique may therefore prove to be a better method for twin reduction, however the numbers in this study are small so further trials are needed.
Oscillation
Excessive oscillation of a dizygotic twin can be attempted between 45–50 days of gestation. Fetal termination by oscillation either alters umbilical blood flow or dislocates the fetus from the umbilicus (Beavers et al, 2017). The designated fetus is oscillated quickly at 0.5–2 oscillations per second, stopping blood flow or decreasing blood rate or frequency. One to four sessions may be needed to observe altered umbilical blood flow. Success is determined when a heartbeat or umbilical blood flow is not identified with colour flow doppler ultrasonography. All mares (5/5) were discharged with a single viable pregnancy in a clinical study (Beavers et al, 2017). Difficulties with this procedure include isolating an individual fetus and producing enough turbulence to sever the cord without affecting the remaining fetus.
Thoracic compression
Ultrasound guided manual crushing of the thorax of a twin against the brim of the pelvis has been described (Crabtree, 2018; McCue, 2021). A recent research abstract described the heart stopping, demonstrated by colour flow during compression (Arnold, 2023). However, fetal compression may be needed on successive days until the selected fetus is dead (Arnold, 2023). The likelihood of success was significantly higher in artificially inseminated mares (73%) than in mares carrying in vitro produced embryos (0%) (Arnold, 2023). While this procedure is being performed in practice, its use has only been documented in a report with reduction of monozygotic twins (Peere et al, 2022). More data need to be compiled to determine the success rate for larger numbers of mares to produce a single, healthy, normal-sized foal.
Another procedure recently reported uses the 55–65 day gestational period by manually transrectally crushing the thorax against the brim of the pelvis (McCue, 2021). The heart will stop, demonstrated by color flow during compression. When compression ceases however, the heart often starts up again – and may need numerous compressions over approximately 15 minutes. In addition, complete cessation of the heart may not be accomplished after one treatment and multiple sequential days may be needed. If compression is too strong or too long, rupture of the chorioallantois is possible and the release of fetal fluids can affect the remaining fetus. While this procedure is being performed in practice, its use has only been documented in a couple of reports; with monozygotic twins and the other with two dizygotic twins (Chopin, 2017; Peere et al, 2022). More data need to be compiled to determine the success rate for a single, healthy, normal-sized foal in dizygotic twins.
Craniocervical dislocation
Craniocervical is described as the dislocation of the first cervical vertebrae from the cranium, disrupting the ligamentous attachments and severing the spinal cord. This procedure can be performed using a transrectal, transabdominal or colpotomy approach between 60 and 110 days of gestation to produce a single foal (Wolfsdorf et al, 2005; Basset et al, 2021). The basis for this procedure is to eliminate one twin prior to complete placental formation, allowing the remaining fetus to use the entire endometrial surface for nutrient and oxygen exchange and to grow to its full potential.
All three approaches have a similar success rate (63–64%) for having a normal-sized healthy single foal (Wolfsdorf et al, 2005; Basset et al, 2021). The positive factors associated with this procedure are that that there is no penetration of the uterus, a high success rate relative to other procedures at this stage and a normal size foal as a result of reduction prior to complete placental formation. The negative factors include potential rectal damage (transrectal), incisional infection (flank), colpotomy (intestinal prolapse) and abortion of both fetuses. Fetal viability should be evaluated every 2–4 weeks to establish normal growth of the continuing fetus and demise of the other, which can take up to 8 weeks.
Transcutaneous ultrasound-guided injection
The use of transcutaneous ultrasonography to aid in twin reduction in the mare was pioneered by Rantanen and Kincaid (1988). In experienced hands, an average of 50% of mares undergoing transcutaneous twin reduction will deliver one live foal (McKinnon, 2011). The suggested time to perform this procedure is between 115 and 130 days of gestation (Rantanen and Kincaid, 1988). The procedure is performed in the standing, heavily sedated mare to promote movement of the fetuses into the cranial abdomen for easier accessibility and to minimise fetal movement during the procedure. The most accessible fetus is selected for reduction, or when possible, the smaller fetus is targeted. The use of a biopsy guide to perform the procedure is at the discretion of the operator. Once the needle is passed into the peritoneal space, the needle tip is located on the ultrasound image, advanced through the uterine wall and into the fetal heart using a quick, thrusting motion. Free flow of blood from the needle after removal of the stylet indicates needle placement within the fetal heart. Potassium chloride or procaine penicillin are injected into the fetal heart, thorax or abdomen (McKinnon and Rantanen, 1988). McKinnon reported that fetal death took slightly longer when injecting penicillin into the thorax or abdomen than directly into the heart. The reported advantages of using procaine penicillin verses potassium chloride are:
- The possible risk of iatrogenic infection is reduced
- There is better visualisation of the agent as it is injected
- Fetal death occurs even in the absence of cardiac placement.
A disadvantage of using procaine penicillin is that it may take up to a few days for the fetus to die (this is also dependent on injection site) (McKinnon and Rantanen, 1988; Rantanen, 1990; McKinnon, 2011).
Cardiac activity of the treated fetus is monitored immediately after the procedure. The fetus does not always die immediately, in which case the mare is monitored over subsequent days to assess the status of both the treated and untreated fetuses. As with craniocervical dislocation, the terminated fetus is mummified and delivered in a small placental sac along with the live fetus. The terminated fetus rarely interferes with the development of the remaining fetus. However, because the procedure is performed after placental formation is complete, some have speculated that placental insufficiency may result in small, unthrifty foals.
Monozygotic twins
The above procedures mainly describe the success rates associated with dizygotic twins. However, it has become evident that an increasing number of monozygotic twins have been identified post-fixation with transfer of single in-vitro produced embryos (1.6%) (Dijkstra et al, 2020). Some of the above procedures have been attempted on a limited numbers of monozygotic twins to include transvaginal ultrasound-guided aspiration, thoracic compression, craniocervical dislocation and transcutaneous ultrasound-guided injection. In a report of 8 monozygotic twins, with the afore mentioned procedures used, only had one single normal-sized foal born after thoracic compression (Sper et al, 2012). Additional cases have been reported to have success using craniocervical dislocation and transcutaneous ultrasound-guided injection (Basset et al, 2021; Peere et al, 2022). However, with the small number of successful reductions, the optimum technique has yet to be identified.
Conclusions
Reduction of a twin is most successful during the mobility stage. However, if identification occurs at a later point in gestation, there are numerous procedures, depending on the time period, to perform twin reduction. These include transvaginal ultrasound-guided aspiration, puncture or injection, compression of the thorax, oscillation, craniocervical dislocation and transcutaneous ultra-sound-guided injection. However, when the above procedures are not an option or fail to produce a single viable conceptus, aborting both conceptuses is an option of last resort. Most mares that are carrying twins will naturally abort between 7 and 9 months. More importantly, mares can occasionally have an increased risk for dystocia, cervical tears, retained placentas and associated complications. Although some mares successfully carry twin pregnancies to term, the risk of complications warrants termination of one or both fetuses depending on the individual mare's situation. Knowing what options are available can help the veterinarian identify which procedure may be most appropriate for the mare and client.
KEY POINTS
- It is favourable to identify twin pregnancies before embryo fixation.
- Manual reduction before fixation has the highest success rate of single born foal.
- Different stages of gestation present different opportunities to use various procedures.
- Post-fixation oscillation or thoracic compression are the least invasive methods and potentially have the best result for a normal-sized healthy single foal.
- Craniocervical dislocation has the highest success rates after 65 days of gestation of delivering a normal-sized healthy single foal.