Timing of delivery of twins should be decided when the benefit of prolonging the pregnancy outweighs the risk of stillbirth. Perinatal mortality of singletons is increased significantly after 42 weeks, whereas perinatal mortality in twins starts to increase significantly after 37 weeks. Recent, large cohort studies have showed significantly higher stillbirth rates near term even in apparently low-risk monochorionic twin pregnancies. Stillbirth risk in monochorionic twins is three-fold higher than in dichorionic twins, and this risk remains high throughout the pregnancy. In uncomplicated monochorionic twins between 32 and 37 weeks, no statistically significant increase of stillbirth occurs between 32 and 37 weeks; these pregnancies are usually monitored until delivery at 37 weeks. The risk of stillbirth in dichorionic twins does not seem to be different between 28 and 38 weeks, justifying a differential policy for the timing of delivery in monochorionic compared with dichorionic twin pregnancies. Therefore, uncomplicated dichorionic twins should be managed expectantly, and delivery can be arranged from 38 weeks. In cases of discordant fetal wellbeing at preterm gestations, timing of delivery should be based mainly on parameters and outlook for the healthy twin balanced against the condition of the compromised fetus. The threshold for early delivery may be lower in monochorionic twins because of the high mortality and morbidity in surviving twins with co-twin death.
Introduction
The management of multiple pregnancies forms an important cornerstone of modern antenatal care. In the past 3 decades, the incidence of multiple pregnancies has increased, mainly because of increasing use of assisted reproduction techniques, with more than one- quarter of in-vitro fertilisation pregnancies resulting in multiple gestations . This increase in the incidence of multiple pregnancies after assisted reproduction techniques has been associated with dizygotic and monozygotic pregnancies . The necessity of having evidence-based management strategies in multiple pregnancies is because they are at a higher risk of complications, such as preterm labour, fetal growth restriction, and preeclampsia, and they are also associated with a significantly increased risk of stillbirth compared with singleton pregnancies . Therefore, timing of delivery in multiple pregnancies is crucial to mitigate the risk of these complications. This chapter will focus on the available evidence in deciding the appropriate timing of delivery in multiple pregnancies, with reference to twin pregnancies.
Accurate dating and assessment of chorionicity
Twin pregnancies are at increased risk of perinatal morbidity and mortality, compared with singleton pregnancies, mainly as a consequence of preterm delivery and fetal growth restriction . An accurate estimation of the gestational age is vital to manage these complications. Considerable evidence suggests that twin pregnancies can be reliably dated using singleton crown–rump length charts between 11 and 14 weeks, and by fetal head circumference thereafter . Biometry of the larger twin is more pragmatic than smaller twin in dating of twin pregnancy, as fetal growth restriction could exist even in an early stage . No uniform policy of dating in in-vitro fertilisation (IVF) pregnancies exists and, as such, some chose the date of oocyte retrieval whereas others use the embryo replacement date for pregnancy dating . To overcome this limitation, crown–rump length measurement between 11 and 14 weeks has been suggested, even in IVF pregnancies .
Perinatal mortality and morbidity among twins are determined by chorionicity, with a higher prevalence of complications in monochorionic compared with dichorionic twins . Therefore, determination of chorionicity is the most important step in managing twin pregnancies. First-trimester markers of two different placental masses and lambda or ‘T’ sign are more reliable in determining chorionicity, as some of the other markers disappear with advancing gestation .
Accurate dating and assessment of chorionicity
Twin pregnancies are at increased risk of perinatal morbidity and mortality, compared with singleton pregnancies, mainly as a consequence of preterm delivery and fetal growth restriction . An accurate estimation of the gestational age is vital to manage these complications. Considerable evidence suggests that twin pregnancies can be reliably dated using singleton crown–rump length charts between 11 and 14 weeks, and by fetal head circumference thereafter . Biometry of the larger twin is more pragmatic than smaller twin in dating of twin pregnancy, as fetal growth restriction could exist even in an early stage . No uniform policy of dating in in-vitro fertilisation (IVF) pregnancies exists and, as such, some chose the date of oocyte retrieval whereas others use the embryo replacement date for pregnancy dating . To overcome this limitation, crown–rump length measurement between 11 and 14 weeks has been suggested, even in IVF pregnancies .
Perinatal mortality and morbidity among twins are determined by chorionicity, with a higher prevalence of complications in monochorionic compared with dichorionic twins . Therefore, determination of chorionicity is the most important step in managing twin pregnancies. First-trimester markers of two different placental masses and lambda or ‘T’ sign are more reliable in determining chorionicity, as some of the other markers disappear with advancing gestation .
Mechanism for differing risks of pregnancy loss in monochorionic and dichorionic twins
The higher mortality in monochorionic twins is attributed to the effects of placental vascular characteristics and degree of placental sharing of each twin. Vascular anastomosis between both fetuses at the level of placenta is always present in monochorionic twins, and blood flow in these is often balanced. Up to 15% of monochorionic twins could be complicated by chronic twin-to-twin transfusion syndrome (TTTS) and selective fetal growth restriction caused by haemodynamic imbalance between these anastomoses and unequal placental sharing, respectively . These complications of monochorionic twins are responsible for high early fetal loss rate, and it has been estimated in early research that pregnancy loss rate of monochorionic twins is 12 times higher than dichorionic twins before 26 weeks . With the increasing use and experience of fetoscopic laser techniques, however, early fetal loss rate in monochorionic twins has been significantly reduced ( Fig. 1 ).
Late pregnancy loss in monochorionic twins is not as high as in early pregnancy, but remains higher than dichorionic twins at term. Reasons for term fetal loss in monochorionic twins are not well studied; however, twin anemia–polycythaemia sequence, acute fetal transfusions, congenital anomalies, and hidden fetal growth restriction are thought to be possible contributory factors. Twin anemia–polycythaemia sequence is characterised by large inter-twin haemoglobin differences without signs of twin oligo–polyhydramnios sequence . Twin anemia–polycythaemia sequence may occur spontaneously or after laser surgery for TTTS. The spontaneous form complicates about 3–5% of monochorionic twin pregnancies . The acute form of TTTS is unlikely in the antenatal period, and usually occurs during labour in monochorionic twins. A two- to four-fold increased risk of structural anomalies occurs in monochorionic twins, particularly congential heart disease. The incidence of congenital heart defects is six times higher in monochorionic twins than in dichorionic twins, which may be an additional contributory factor .
Fetal growth restriction is associated with an increased risk of late pregnancy loss in twin pregnancies. Regardless of chorionicity, twins grow at the same rate as singletons up to 32 weeks’ gestation , and thereafter the growth rate is slowed. This may be related to reduced intrauterine physical space or to uteroplacental insufficiency. Discordance in fetal growth of more than 25% is associated with higher mortality in twins . In a large study that included more than 2000 twin pregnancies, D’Antonio et al. showed that perinatal loss in twins with a birth-weight discordance of more than 25% was significantly greater (60.9 per 1000 fetuses) compared with those with a discordance less than 25% (8.6 per 1000 fetuses). Their analysis further showed that birth-weight discordance and gestational age but not chorionicity and individual fetal size percentile were the only independent predictors of perinatal mortality in twin pregnancies .
These data highlight that, in monochorionic twin pregnancies, the mechanism of fetal loss before 26 weeks is primarily related to complications such as TTTS and selective fetal growth restriction. After this gestation, the main determinants of pregnancy loss in monochorionic twins are complications such as twin anemia–polycythaemia sequence, acute fetal transfusions, congenital anomalies, and fetal growth restriction . In dichorionic pregnancies, the predominant factor contributing to fetal loss relates to fetal growth restriction manifested by discordance in fetal size, and this assumes significant importance when the degree of discordance is more than 25%. In addition to fetal loss caused by factors previously mentioned, other important contributors to perinatal mortality in both monochorionic and dichorionic twins is gestational age at delivery .
Risk of fetal and neonatal mortality with advancing gestation in monochorionic and dichorionic twins
The contribution of preterm delivery and prematurity towards neonatal mortality after 32 weeks of gestation is significantly reduced in recent years as a result of improved neonatal care facilities . Prospective risk of fetal death in twin pregnancies, however, increases with advancing gestation, and it ranges between 0.2% and 0.4% per gestational week between 32 and 38 weeks’ gestation . Evidence suggests that the risk of stillbirths in twins equaled that of post-term singletons by 36–37 weeks’ gestation . Therefore, appropriate management of this late gestational age period is critical because twins are known to experience significantly greater stillbirth rates compared with singleton gestations near term. The main limitation of these population-based studies is the lack of data on chorionicity , and, although it is clear that twins have an increased risk of mortality compared with singletons, it also true that the mechanisms and causes of such an increased rate of pregnancy loss differ in chorionicity of the pregnancy. It is, therefore, vital that decisions about the optimal timing of delivery for term and near-term twins should be based on chorionicity.
Dichorionic twins
Evidence on rates of fetal death in dichorionic twins is consistent. Recent, large cohort studies have shown that risk of stillbirth in dichorionic twins remains below 3.5 per 1000 ongoing fetuses between 26 and 36 weeks ( Fig. 2 ). In a large, retrospective cohort study, based on 4912 dichorionic twin pregnancies, Dias et al. reported that the risk of stillbirth in dichorionic twins after 26 weeks remained static at around 1 per 1000 ongoing fetuses in each 2-week epoch, with a total stillbirth rate of 6.5 per 1000 (95% CI 4.8 to 9.8) ongoing fetuses . These data are similar to those reported by Hack et al. , in which the total stillbirth rate after 26 weeks was reported to be 6.6 (95% CI 3.9 to 11.0). Therefore, the current evidence suggests that the stillbirth rate in dichorionic twins is fairly constant after 26 weeks but only up until 38–39 weeks, as evidence shows that rates of fetal death significantly increase once a gestational age of 38–39 weeks is reached . Hack et al. reported that most of late deaths in dichorionic twins in their study were a result of either diagnosed small for gestational age with normal fetal well-being tests, or by undiagnosed small for gestational age . Therefore, it is rational to have expectant management until 38 weeks and plan elective delivery at 38 weeks in uncomplicated dichorionic twin pregnancies. Women who intend to prolong the pregnancy beyond 38 weeks should have fetal growth and wellbeing assessed; if fetal growth restriction is diagnosed, then recommendations should be made for delivery; if it is not diagnosed, and even in the presence of reassuring assessment of fetal wellbeing beyond this gestation, adverse outcome may still be a risk. In the case of dichorionic twin pregnancies, timing of delivery is important. Discordant fetal complications, such as congenital defects, should be individualized, and management focused on optimising outcome for the healthy co-twin.
Monochorionic twins
The risk of fetal death in monochorionic twins is higher than in dichorionic twins, and this risk is not just confined to early pregnancy; higher rates have been documented even after 32 weeks ( Fig. 2 ). Relative low prevalence of monochorionic twins has resulted in a paucity of epidemiological evidence on which to base clinical decisions about the optimal timing of monochorionic twin birth to avoid intrauterine fetal death. The optimal timing of delivery in monochorionic twins has been much debated, with some suggesting expectant management in uncomplicated monochorionic twins until 36 weeks, and others recommending an early delivery at around 32 weeks ( Table 1 ), owing to presumed risk of unexplained late fetal loss after this gestation . Tul et al. evaluated 387 monochorionic twin pregnancies that were delivered after 24 weeks in Slovenia between 1997 and 2007, and reported that the prospective risk of stillbirth per pregnancy after 34 weeks of gestation was as high as 16.5 (95% CI 9.0 to 30.2). The weakness of this study was that it lacked a standardised ultrasound surveillance protocol for monochorionic twins, and the frequency of assessments was based on decisions of the attending obstetrician rather than the current evidence-based fortnightly assessments, thus introducing a degree of bias in their results . In contrast, in another retrospective study by Hack et al. , the investigators reported that the delivery of monochorionic twins should be at or before 37 weeks, as the prospective risk of stillbirth after 36 weeks for monochorionic twins is significantly increased 21.6 (8.4 to 54.3), and is almost nine times higher than in dichorionic twins (2.3, 95% CI 0.8 to 6.8) at that gestation . In this study, the investigators reported a standardised protocol for the monitoring of twin pregnancies consisted of fortnightly scans for growth, amniotic fluid and Doppler assessments . They reported that, in the group of monochorionic twins, six stillbirths occurred after 32 weeks gestation, but they were all caused by complications of TTTS . The same group examined a larger cohort of twin pregnancies in a further study and reported that, after excluding cases with TTTS, the risk of stillbirth was low, between 32 and 36 weeks . Six neonatal deaths occurred, three early and three late, most of which were caused by complications of prematurity related to delivery before 36 weeks. This led them to conclude that, in uncomplicated monochorionic twins, planned delivery should be undertaken at or after 36 weeks as, before these gestations, the complications of prematurity can be substantial . Considerable evidence suggests that late preterm delivery is not without risks, and is known to significantly increase the risk of neonatal morbidity and mortality , thus adding further support to the practice of delivering uncomplicated monochorionic twins at or after 36 weeks.
| Reference | Number of monochorionic twins | Country | Prospective risk of stillbirths at 32 weeks (per 1000 ongoing fetuses) | Prospective risk of stillbirths after 36 weeks (per 1000 ongoing fetuses) |
|---|---|---|---|---|
| Dias et al., 2012 | 549 | UK | 6.09 (2.8–13.2) | 3.37 (0.9–12.2) |
| Farah et al., 2012 | 144 | Ireland | 0.0 (0.0–14.8) | 16.7 (5.7–47.8) |
| Tul et al., 2011 | 387 | Slovenia | 7.42 (3.2–17.2) | 10.90 (4.7–25.3) |
| Hack et al., 2011 | 465 after excluding TTTS | Netherlands | 5.4 (2.3–12.5) | 1.7 (0.3–9.4) |
| Smith et al., 2010 | 234 | USA | 4.18 (1.1–15.1) | 0.00 (0.0–14.9) |
| Domingues et al., 2009 | 111 | Portugal | 6.33 (1.1–35.0) | No data |
| Lee et al., 2008 | 189 | USA | 0.00 (0.0012.1) | 8.30 (1.5–45.7) |
| Lewi et al., 2008 | 178 | Belgium and Germany | 3.11 (0.5–17.4) | 0.00 (0.0–20.5) |
| Hack et al., 2008 | 188 | Netherlands | 6.83 (1.9–24.5) | 21.60 (8.4–54.3) |
| Simoes et al., 2006 | 193 | Portugal | 0.00 (0.0–11.4) | 0.00 (0.0–22.0) |
| Barigye et al., 2005 | 151 | UK | 7.19 (2.0–25.8) | 5.38 (0.9–29.8) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
