Multiple pregnancy, a complication of assisted reproduction technology, is associated with poorer maternal and perinatal outcomes. The primary reason behind this is the strategy of replacing more than one embryo during an assisted reproduction technology cycle to maximise pregnancy rates. The solution to this problem is to reduce the number of embryos transferred during in-vitro fertilisation. The transition from triple- to double-embryo transfer, which decreased the risk of triplets without compromising pregnancy rates, was easily implemented. The adoption of a single embryo transfer policy has been slow because of concerns about impaired pregnancy rates in a fresh assisted reproduction technology cycle. Widespread availability of effective cryopreservation programmes means that elective single embryo transfer, along with subsequent frozen embryo transfers, could provide a way forward. Any such strategy will need to consider couples’ preferences and existing funding policies, both of which have a profound influence on decision making around embryo transfer.
Introduction
Assisted reproductive technology (ART) includes procedures that involve in-vitro handling of human gametes or embryos for purposes of establishing a pregnancy . The past 3 decades have witnessed a dramatic increase in the number of women undergoing ART treatment worldwide, with more than 5 million children conceived by means of this treatment . Ovarian stimulation has traditionally been used in ART to produce a number of oocytes, and has the involved replacement of more than one embryo to maximise pregnancy rates. This has led to an increase in the number of twins and higher order multiple pregnancies . Although initially seen as an acceptable outcome of ART, concerns have increasingly been raised about the consequences of such a policy on the health of women and children. Compared with singleton pregnancies, women carrying twins and other multiples are at increased risk of maternal complications, such as preeclampsia, preterm premature rupture of membrane, and caesarean section . A fivefold increase in neonatal mortality occurs after twin delivery compared with singleton delivery, and this risk is higher in higher order multiples .
In many parts of the world, a conscious effort has been made to reduce the rates of iatrogenic multiple pregnancy by means of legislation and policy changes. A favourable downward trend in multiple pregnancy rates has resulted from the introduction of single-embryo transfer (SET) policies in some countries, but this is by no means a universal phenomenon . In this chapter, we review various strategies for reducing multiple pregnancies in in-vitro fertilisation (IVF), including legislative and policy changes.
Burden of the problem: risks of multiple pregnancies after assisted reproductive technology
Data from 376,971 European IVF treatment cycles in 2007 show a multiple birth rate of 22.3% (21.3% twins and 1% triplets), similar to rates in 2005 and 2006 (21.8 and 20.8%, respectively) . Figures from the Society for Assisted Reproduction Technology (SART) registry in the USA, based on 108,130 ART cycles, revealed a multiple birth rate of 35.4, of which 31.8% were twin, 3.5% were triplets, and 0.1% were higher order multiple .
In the UK, data published by the Human Fertilisation and Embryology Authority (HFEA) for the years 2009 and 2010 show a multiple pregnancy rate of 25.4 and 22.2%, respectively. In 1992, for every 100 deliveries after successful ART treatment, 28 were multiple births, whereas the multiple births figures stood at 23 per 100 deliveries after ART treatment in the year 2006.
Burden of the problem: risks of multiple pregnancies after assisted reproductive technology
Data from 376,971 European IVF treatment cycles in 2007 show a multiple birth rate of 22.3% (21.3% twins and 1% triplets), similar to rates in 2005 and 2006 (21.8 and 20.8%, respectively) . Figures from the Society for Assisted Reproduction Technology (SART) registry in the USA, based on 108,130 ART cycles, revealed a multiple birth rate of 35.4, of which 31.8% were twin, 3.5% were triplets, and 0.1% were higher order multiple .
In the UK, data published by the Human Fertilisation and Embryology Authority (HFEA) for the years 2009 and 2010 show a multiple pregnancy rate of 25.4 and 22.2%, respectively. In 1992, for every 100 deliveries after successful ART treatment, 28 were multiple births, whereas the multiple births figures stood at 23 per 100 deliveries after ART treatment in the year 2006.
Risks associated with multiple pregnancies
Multiple pregnancy increases the risk of maternal mortality, estimated to be 14.9 out of 100,000 compared with 5.9 out of 100,000 for singleton births . The incidence of complications of pregnancy, such as pregnancy-induced hypertension, is also increased, and the risk of anaemia is doubled; uterine atony, dystocia, increased operative deliveries, and postpartum haemorrhage are all associated with multiple pregnancy , as is the incidence of depression .
Prematurity occurs in nearly one-half of all multiple pregnancies, and is the main cause of neonatal morbidity and mortality; 42% of twins and 8% of singletons are born before 37 completed weeks . Twins face a six-fold and triplets a 10–20-fold increase in risk of mortality compared with singletons .
Gestational age and birth weight are inversely related to the number of fetuses. The mean gestation for twins, triplets and quadruplets is 35, 33 and 29 weeks, respectively. Ninety per cent of triplets and higher order multiples weigh less than 2500 g compared with only 6% of all singletons; 78% of triplets and higher order multiples, and 48% of twins will need admission to neonatal intensive care units compared with 15% of singletons .
Higher rates of congenital anomaly in children conceived through ART have been reported and are more common in multiple pregnancies; the risk of cerebral palsy is increased three- to seven-fold in twins and ten-fold in triplets . Long-term behavioural problems have been found to be higher in children born after multiple births compared with children born as singletons . Prenatal screening poses additional difficulty, and leads to increased anxiety during the antenatal period for women carrying multiples . Multiple gestation requires more frequent clinical and ultrasonographic monitoring compared with singleton pregnancies.
Maternal and neonatal complications in multiple pregnancies are presented in Tables 1 and 2 .
| Complications | Singleton | Twins | Triplets |
|---|---|---|---|
| Preeclampsia (%) | 6 | 17 | 20–39 |
| Gestational diabetes (%) | 3 | 5–8 | >10 |
| Neonatal intensive care unit admission (%) | 15 | 48 | 78 |
| Perinatal mortality (per 1000 live births) | 10 | 27 | 62 |
| Intracranial bleeding (%) | 0.4 | 1.9 | 5.6 |
| Respiratory distress syndrome (%) | 1.6 | 8.0 | 20.4 |
| Complications | Singleton | Multiple pregnancies |
|---|---|---|
| Maternal mortality (per 100,000 birth) | 5.9 | 14.9 |
| Pre-term delivery (<37 weeks) (%) | 8 | 42 (twins) |
| Congenital malformations (%) | 4.8 | 9.2 (twins) |
Healthcare costs rise four-fold in twins and ten-fold in triplets compared with costs associated with singleton pregnancies. The health risks associated with preterm birth are important contributors to increasing costs associated with multiple pregnancies. Data from the USA indicate that the cost per preterm infant is about 51,600 US dollars. In 2005, ART contributed to 4% of all the preterm deliveries in the USA, costing the society over 1 billion US dollars . Chambers et al. studied the cost involved in inpatient admission of singleton and multiple gestation after ART treatment, and compared it with non-ART babies. The costs after ART treatment were 4818 Euros for singleton delivery, 13,890 Euros for twin delivery, and 54,294 Euros for triplet delivery.
Strategies to reduce multiple pregnancies after assisted reproductive technology cycles
Traditionally, transfer of multiple embryos in IVF with the intention of maximising pregnancy rates was the norm. Subsequently, analysis of the HFEA database suggested that, in the presence of four or more embryos, replacement of three rather than two embryos increased the risk of multiples without increasing the live birth rate. This led to a revision of the policy of transferring three or more embryos . It is important to devise a mechanism to determine the optimum number of embryos to be transferred in a woman, depending upon the clinical characteristics, so that the risk of multiple pregnancy can be minimised, without jeopardising her chances of conception.
Three-compared with two-embryos transfer
A Cochrane review by Pandian et al. included only one small trial consisting of 45 women, in which a double-embryo transfer (DET) and a triple-embryo transfer were compared. The cumulative live birth rates (48% v 41%) did not differ significantly in the two groups (four cycles of a DET compared with three cycles of a triple-embryo transfer). The multiple pregnancy rate was significantly higher in the triple-embryo transfer group (30% v 0%; P = 0.05) . Apart from this, available data on outcomes after a DET policy are from observational studies. The earliest convincing set of results in favour of double (compared with triple) embryo transfer are from an analysis of data from 44,236 cycles within the HFEA database. In women with more than four fertilised oocytes, transfer of three (rather than two) embryos did not enhance live birth rates, but significantly increased the risk of multiple pregnancy (OR1.6; 95% CI 1.5 to 1.8) .
Retrospective data from Germany found comparable pregnancy rates after elective transfer of two and three embryos up to the age of 40 years (22% v 22.5%), with multiple pregnancy rate of 16.1% in the double-embryo group compared with 24% in the triple-embryo group .
In their study using a donor model, Licciardi et al. showed similar clinical pregnancy rates after the elective transfer of either two or three embryos (57.5% v 55.8%) in recipients. The multiple pregnancy rate was 40.5% in women who received two embryos and 51.0% in women who received three. No triplets occurred after transfer of two embryos, but the triplet rate in the triple embryo group was 13.8%. These findings are corroborated by results of other observational studies .
European data show that 64% of all embryo transfers in 2001 and 75% in 2007 were either SET or DET. The triplet pregnancy rates were 1.5% and 1% for the year 2001 and 2007, respectively . Data from the USA for 2001 show 37% transfers being either SET or DET, with a corresponding triplet rate of 3.8% .
Thus, despite the lack of well-designed randomised-controlled trials, the overwhelming strength of observational data has led to a change in practice on both sides of the Atlantic, which has resulted in a reduction of triplet pregnancies in IVF. In the UK, the HFEA and Royal College of Obstetricians and Gynaecologists have recommended a DET policy except in exceptional situations, such as multiple IVF failures and advanced age .
Elective single embryo transfer
Elective single embryo transfer (eSET) is the most effective way of minimising the risk of twins in IVF and fulfilling the main objective of infertility treatment: delivery of a healthy baby after an uncomplicated pregnancy.
The shift from a triple- to a double-embryo transfer policy to reduce the risk of triplets and higher order multiples was easier to implement because pregnancy rates were not compromised. The shift to elective SET, however, was difficult to implement owing to significant concerns about a reduction in pregnancy rates.
Pooled data from randomised-controlled trials, which included women with a good prognosis, show that, compared with DET, eSET significantly reduces twinning rates, but also halves live birth rates per fresh IVF cycle . In their large, multicentre, randomized trial ( n = 631 women) comparing DET with SET plus single frozen embryo transfer, Thurin et al. found comparable cumulative live birth rates (42.9% v 38.8%). The multiple pregnancy rates were significantly lower in the SET plus a single frozen thawed embryo group (33.1 v 0.8%; P < 0.001). Vilska et al. highlighted the difference in pregnancy rates after elective SET (29.7%) and non-elective SET (20.2%). They also presented observational data on cumulative pregnancy rates after eSET plus frozen or thawed embryo transfer, and double embryo transfer. The pregnancy rates per oocyte retrieval were 47.3% after SET plus single frozen thawed embryo, and 29.4% after DET. The multiple pregnancy rate was 23.9% in the DET group.
Mclernon et al. , conducted an individual patient data meta-analysis, in which data from eight trials that included 1367 women were pooled. They found that the live birth rate after eSET in a fresh IVF cycle was significantly lower than that after DET (27% v 42%; 0.5 OR CI 0.39 to 0.63), as were the multiple pregnancy rates (2% v 29%; 0.04 OR CI 0.01 to 0.12). They found similar cumulative live births after an additional single frozen embryo transfer (38% v 42%). The odds of a term singleton birth after eSET were five times higher than after DET.
National data from Sweden suggest that liberal use of elective SET in up to 70% of all cases has not resulted in a noticeable decrease in live birth rates in fresh IVF cycles, but has led to a substantial reduction in multiple pregnancies . Karlstrom et al. used regression analysis, and concluded that an eSET rate of 75% would result in a multiple birth rate of less than 5%.
Cost effectiveness of elective single embryo transfer
Fiddelers et al. compared the cost-effectiveness of eSET compared with DET in a randomised population, where the pregnancy rate per embryo transfer was 21% in the eSET group and 40% in the DET group. The societal cost per couple receiving eSET was significantly lower (7334 Euros) compared with 10,924 Euros in the DET group. The incremental cost-effectiveness ratio after DET was 19,096 Euros, indicating the additional cost to the society for extra successful pregnancy after a DET.
Fiddelers et al. , in a systematic review, which included three randomised trials and one observational study, looked at the cost implications after SET and DET, and found that a DET strategy was effective but more expensive in a fresh IVF cycle. Elective SET was more cost-effective only in women with a good prognosis who could go on to have a frozen embryo transfer if a pregnancy did not occur in a fresh cycle. The incremental cost-effectiveness ratio associated with DET ranged from 8399 to 30,571 Euros, indicating the degree of investment needed by society to get an additional pregnancy.
In a health-economic analysis by a Swedish group, the investigators compared the cost to society of two IVF strategies (eSET with frozen compared with DET) until 6 months after delivery . The SET policy resulted in lower average total cost until 6 months after delivery. Significantly fewer maternal and perinatal complications occurred in the eSET group. Although the DET group had more deliveries, the incremental cost per extra delivery was 71,940 Euros. The investigators did not support the routine use of DET over SET.
Fiddelers et al. also compared the cost-effectiveness of seven embryo-transfer strategies in IVF using a Markov model. The strategies included three cycles of eSET cycles for all patients, DET cycles for all patients, and standard treatment cycles, in which eSET was offered for women younger than 38 years of age and DET for the remainder. Overall, a choice between three cycles of eSET, or DET, or standard treatment cycles was suggested, depending on societal willingness to pay for an additional pregnancy. If the willingness to pay limit was 7350 Euros, then three eSET would be preferable. At a threshold of 15,250 Euros, the option would be three cycles of standard treatment cycles, whereas, if the limit were further raised to 17,000 Euros, then a strategy of three DET cycles could be used.
Effective cryopreservation
As the data above show, the success of an eSET policy depends on the availability of an effective and reliable cryopreservation service to ensure maximum rates of cumulative live birth per woman after each oocyte retrieval. With an effective cryopreservation programme, cumulative delivery rate of over 50% can be achieved after an elective single embryo transfer programme .
In their study, which included analysis of 20,244 cycles in 7,244 patients, Kato et al. have shown that use of elective single embryo policy accompanied by an efficient cryopreservation programme yields satisfactory live birth rates for women up to the age of 45 years. Schnorr et al. , in a retrospective analysis, evaluated the effect of a cryopreservation programme on multiple pregnancy rates. They developed a theoretical model that yielded a cumulative pregnancy rate of 77% and twin pregnancy rate of less than 20%, with no triplets after limiting the number of embryos transferred to two.
Higher costs after the use of more liberal eSET and subsequent frozen cycles is a concern. In their retrospective analysis of data from two time periods (DET v SET period), Veleva et al. found significantly higher cumulative live birth rate per oocyte retrieval and reduced multiple pregnancy rates after SET compared with DET. The incremental cost-effectiveness ratio was calculated and a term live birth in the elective SET period was found to be 19,889 Euros less expensive than in the DET period .
Single blastocyst transfer
Retrospective data from Australia found that the highest live birth and healthy baby rates (46% and 38%) for women aged less than 35 years occurred in those undergoing single blastocyst transfer .
In another retrospective analysis from the USA, single blastocyst transfer when compared to double blastocyst transfer in women under 40 years resulted in similar pregnancy rates (63% v 61%), but significantly reduced rates of twins .
Guerif et al. , in a prospective study, compared single elective cleavage stage transfer on day 2 with single blastocyst transfer in 478 couples. The delivery rate per fresh IVF cycle was significantly higher in single blastocyst transfer compared with cleavage stage (day 2) transfer. The cumulative delivery rates per couple after fresh and frozen transfer were similar in both the groups.
Multifetal reduction
Procedures, such as multifetal reduction, have been carried out to reduce triplet or higher order pregnancies to twins or rarely, singletons. Such procedures involving termination of one normal fetus is undertaken to reduce the chances of miscarriage of all the fetuses or very premature delivery with its attendant neonatal morbidity and mortality . Less commonly, fetal reduction from twins to singleton has been advocated, although it carries risk of miscarriage and has significant ethical implications.
In a retrospective study comprising 255 women with trichorionic triplet pregnancies, 185 women underwent triplet reduction, and the remaining 70 opted for expectant management . The investigators found reduction in incidence of very premature delivery and very low birth weight infant by one-third, although the risk of fetal loss was significantly higher in women who underwent fetal reduction compared with those who had expectant management (15% v 5%).
In a prospective study, which included 148 women with triplet pregnancies, Boulot et al. compared expectant management ( n = 83) with fetal reduction to twins ( n = 65). The incidence of prematurity before 28, 32 and 34 weeks ( P < 0.002) and low birth weight infants ( P < 0.001) was all significantly less in the reduced twins group. The study indicated fetal reduction as an effective method to improve pre-term delivery rates and fetal growth.
In a systematic review, Papageorghiou et al. combined data from six studies (embryo reduced [ n = 482] v expectant [ n = 411]). Embryo reduction was associated with decreased risk of preterm delivery but increased risk of miscarriage. It was calculated that seven (95% CI 5 to 9) embryo reductions were required to prevent one early preterm delivery, whereas the number of reductions that would cause one miscarriage was 26 (95% CI 14 to 193).
Although the studies do support fetal reduction from triplets to twins, some data advocate reduction from twins to singleton . In a retrospective case-control study, Evans et al. found no significant change in pregnancy course and outcome after reduction to singleton compared with twin pregnancy. Garel et al. evaluated the psychological outcomes in women after multifetal reduction. One-third experienced persistent depressive symptoms, such as sadness and guilt but, after 2 years, most had overcome the emotional pain associated with the procedure.
Given the widespread availability of embryo cryopreservation, the entire exercise of trying to achieve a pregnancy through replacement of multiple embryos followed by fetal reduction would seem to many to be an inefficient way of achieving the desired outcome of a single healthy baby. Apart from the clinical and economic issues, the psychological consequences and ethical concerns surrounding selective fetal reduction limit its widespread applicability as a policy to reduce multiple birth.
Legislation
In countries in which IVF is publically funded, the complications of iatrogenic multiple pregnancy, along with the attendant economic costs of obstetric and neonatal care, underline the need to introduce legislation to minimise this risk.
In the USA, SART guidelines have noted the fact that improved ovarian stimulation strategies and culture conditions have yielded embryos with higher implantation potentials and encouraged limiting the number of embryos transferred to reduce multiple pregnancy rates. The first such SART guideline in 1998 limited the maximum number of embryos transferred to three in women under the age of 35 years. Over the years, guidance has been further refined, and the maximum transferable number was brought down to one – two in 2004 . After introduction of these guidelines, the multiple births after ART cycle showed a downward trend until 2006, and the higher order multiple birth showed reduction until 2008. The recent American Society for Reproductive Medicine practice committee guideline has recommended transferring one to two embryos at cleavage stage or single blastocyst in a woman younger than 35 years with a favourable prognosis .
In the 1990s, the UK, and Germany enacted legislation restricting the maximum number of embryos transferred to three in a treatment cycle. In 2001, HFEA in the UK restricted the maximum number of embryos transferred to two in women under the age of 40 years with no exception . In 2008, HFEA issued guidelines for elective single embryo transfer.
The Belgian government linked the reimbursement for IVF cycles to restriction in number of embryos transferred. A retrospective study from one of the university centres in Belgium analysed the data 1 year before and after introducing legislation, which restricted the transferred embryo numbers . They found no significant difference in the live birth rate during the two time periods (37.5% v 32.5%). The multiple pregnancy rates per clinical pregnancy reduced from 26% (pre legislation) to 8% (post legislation).
The Swedish experience with legislation restricting the transferred embryos number has been analysed by Saldeen and Sandstrom . In 2003, all the women being treated with ART in the country (public or private) underwent eSET, unless risk of twinning was low. Data from three periods (pre- and post-legislation and the intervening transitional period) show that the eSET rate was 25% (pre-legislation), 55% (transitional) and 73% post-legislation. No significant difference was reported in clinical pregnancy rates during the three periods. The twinning rate was significantly reduced in the post-legislation (6%) period compared with the pre-legislation (23%) period.
Couple’s preferences
Couples often have strong preferences around embryo-transfer strategies, and information provision and counselling plays an important role in influencing their decision making. Since IVF treatment entails enormous psychological, emotional and financial costs, many couples, if given a choice, would prefer to complete their family in one treatment cycle. Pinborg et al. conducted a survey of women who had twin babies after ART. The control group consisted of women with naturally conceived twins, and women with a single baby after ART treatment. Among the women with ART twins, 85% preferred twins as their first child compared with 62% of mothers with ART singletons and 60% of mothers with naturally conceived twins. The important reason for preferring twins was a perception that there was a single window of opportunity. Elective SET was acceptable to fewer than one-quarter of women who had conceived through IVF.
A Canadian study found that two-fifths of IVF patients wanted multiple births . The desire was more strongly associated with increased duration of infertility and previous ART treatment. It was negatively associated with having children before starting treatment and recognition about neonatal morbidity. The authors concluded that the role of patient education is important in reducing multiple pregnancies by influencing decision making.
Better communication about the risks of multiple pregnancy does not necessarily change a couple’s opinion . In a Dutch randomised-controlled trial, couples undergoing IVF were exposed to a multifaceted intervention involving different approaches to help them choose between eSET and DET . The four elements of the intervention were (1) information booklet; (2) reimbursement offer; (3) extra session with nurse coordinator, and (4) extra phone call before oocyte pick up to clarify any doubts. The control arm received standard IVF care. In the intervention group, 43% of couples underwent SET compared with 32% in the control group ( P = 0.05) in the first treatment cycle. The investigators suggested that a multifaceted decision-making strategy that improves knowledge could be an important tool in promoting eSET.
The reluctance to opt for eSET could also be influenced by funding issues. Couples undergoing self-funded cycles are less keen to adopt eSET as a policy compared with couples undergoing publicly funded IVF.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
