Most children born after assisted reproduction techniques are healthy. The most important risk associated with in-vitro fertilisation is the higher multiple birth rate, which may result in increased child morbidity because several embryos are transferred at each cycle. Single-embryo transfer reduces this risk dramatically. Large registry studies and meta-analyses have indicated a small but statistically significantly increased risk of congenital malformations among children born after in-vitro fertilisation. Other risks, such as preterm birth and low birth weight seem to be associated more with parental characteristics than the in-vitro fertilisation technique. All knowledge about children conceived through in-vitro fertilisation is based on observational studies, with risks of bias and confounders. It is possible that pregnancies and children resulting from in-vitro fertilisation are more carefully monitored and seek health care more frequently, resulting in more health problems being discovered. Continuous follow up of children after in-vitro fertilisation is of great importance, particularly as new techniques are constantly being introduced.
Introduction
In-vitro fertilisation (IVF) is the most successful treatment for infertility, of both female and male origin. The first IVF child was born in England in 1978 ; since then, more than 4 million IVF babies have been born. In many countries, 1–4% of all children born annually are born as a result of IVF. Early IVF use yielded poor pregnancy and delivery rates. In order to achieve better results, several embryos were transferred at each cycle, resulting in higher pregnancy rates. Since then, a dramatic development has taken place in IVF, and the results have improved considerably from a few per cent to around 35–40% pregnancies per treatment or even higher. Follow up of obstetric and child outcomes is of utmost importance in IVF, particularly as new techniques are constantly being introduced.
In this chapter, we summarise published data on child outcomes after IVF, with a focus on safety aspects. Systematic reviews, meta-analyses and large registry studies are reviewed.
In-vitro fertilisation and multiple births
The most important problem associated with IVF has been the high multiple birth rate (MBR), with associated risks for the children. The MBR has decreased in recent years but is still high in most countries. The latest reports from Europe and USA showed an MBR of 20.8% (Europe) and 30.1% (USA) when assessed per delivery. Multiple births, such as high order multiple pregnancies (triplets and more) and also twins are associated with considerably higher risks for preterm birth and low-birth weight, perinatal mortality, morbidity and disability among survivors. The risk increases with the number of fetuses. When the general obstetric outcomes in assisted reproduction, including both singletons and multiples, have been analysed and compared with the general population, much higher rates of pre-term birth and low-birth weight as well as perinatal mortality (around four-to five-fold) were noticed among IVF babies.
In-vitro fertilisation and multiple births
The most important problem associated with IVF has been the high multiple birth rate (MBR), with associated risks for the children. The MBR has decreased in recent years but is still high in most countries. The latest reports from Europe and USA showed an MBR of 20.8% (Europe) and 30.1% (USA) when assessed per delivery. Multiple births, such as high order multiple pregnancies (triplets and more) and also twins are associated with considerably higher risks for preterm birth and low-birth weight, perinatal mortality, morbidity and disability among survivors. The risk increases with the number of fetuses. When the general obstetric outcomes in assisted reproduction, including both singletons and multiples, have been analysed and compared with the general population, much higher rates of pre-term birth and low-birth weight as well as perinatal mortality (around four-to five-fold) were noticed among IVF babies.
The number of embryos transferred
The most important factor influencing the MBR is the number of embryos transferred. The first study showing successful single embryo transfer (SET) came from Finland. The Scandinavian countries, Sweden in particular, have pioneered the reduction in MBR by introducing SET on a large scale. In Sweden, this policy has resulted in an unchanged overall delivery rate, whereas the MBR has decreased from 25% to 5–6% ( Fig. 1 ). The rate of SET is 70–80%.
Several randomised trials and meta-analyses have compared the delivery rates between SET and double embryo transfer (DET). The largest study is from Scandinavia. The studies showed that the pregnancy and delivery rates were significantly higher after DET than SET when only fresh cycles were compared. If a fresh and frozen and thawed SET were combined in the SET group, the delivery rates were comparable to DET. The MBR was dramatically reduced in the SET group, and follow up of children showed a better outcome for the SET group in child morbidity.
Risks with twin pregnancies
The high MBR in IVF implies that one-third of IVF children in Europe and almost one-half of children in the USA were born as multiple birth babies in 2006. Most were twins, but 5% of the IVF children were triplets. Most women with twin pregnancies and their twins do well; however, it is well recognised that maternal and neonatal morbidity and mortality are significantly increased in twin pregnancies compared with singleton pregnancies.
Major maternal complications in twin pregnancies include hypertensive disorders and eclampsia, antepartum and postpartum haemorrhage, increased rates of caesarean section and admission to intensive care units. Preterm and very preterm birth and low and very low birth weight are the major causes of neonatal mortality and morbidity. Twins have three times higher mortality, and the risk of being born with a very low birth weight (less than 1500 g) is 10-fold for a twin compared with a singleton. At least 60% of all twins are also born before the 37th week of pregnancy. Twins are also at an increased risk of later disabilities, and the risk of cerebral palsy in twins is about five times that of singletons.
Dizygotic and monozygotic twins
Twin pregnancies are divided into two major types: monozygotic and dizygotic.
Monozygotic twinning is a result of the splitting of one fertilised ovum during the first 2 weeks of embryogenesis, whereas dizygotic twins originate from the fertilisation of two ova by different spermatozoa. The rate of monozygotic twinning is believed to be fairly constant around the world and over time (3–4 per 1000 pregnancies). The incidence of dizygotic twinning is affected by many factors, such as maternal age, parity and race; heredity is more variable, ranging from 3–40 per 1000 pregnancies. In white people, about 30% of spontaneous twin pregnancies are monozygotic and 70% are dizygotic. Most twins resulting from assisted reproductive technology are dizygotic, and only 1–2% are monozygotic. Although the monozygotic twinning rate is lower in ART twins, it is still more than twofold compared with the rate in the general population (3–4 per 1000). Increased frequency of monozygotic twinning has been observed after ovulation induction alone and after in-vitro fertilisation, in the latter case especially when it is associated with blastocyst culture or with zona pellucida manipulation for promoting hatching of blastocysts (assisted hatching).
According to the number of layers in the septum between the amniotic sacs, twin placentas are categorised as monochorionic and dichorionic. Dizygotic twins are always dichorionic, but monozygotic twins can be either dichorionic or monochorionic. Recent studies have shown that it is chorionicity rather than zygosity that determines the outcome in twin pregnancies, with monochorionic twins being at higher risk than dichorionic twins. Thus, when comparing twins conceived through assisted reproductive technology and twins in the general population, type of zygosity and chorionicity must be taken into account. Three meta-analyses have shown a higher risk of preterm birth and admissions to neonatal intensive care units in twins conceived through assisted reproductive technology compared with spontaneous twins. The most recent meta-analysis also showed a higher rate of low birth weight in IVF twins than in spontaneous twins.
In a recent Swedish registry study, only dizygotic IVF twins ( n = 1545 twin pairs) were compared with dizygotic non-IVF twins ( n = 8675 twin pairs). After adjustment for year of birth, maternal age, parity and smoking, IVF twins had significantly higher rates of very preterm birth (less than 32 weeks) and neonatal jaundice. No significant difference in low birth weight, birth weight difference within twin pairs or respiratory problems was seen.
Complications after single embryo transfer
Singleton babies resulting from SET are also at an increased risk of adverse outcome compared with singletons from the general population. Singletons born after IVF have twice as high rates of pre-term birth and low-birth weight. The rates of very preterm birth (less than 32 weeks) and very low-birth weight (less than 1500 g) are three times higher than in the general population. The rate of small for gestational age is also increased by a factor of 1.5 ( Table 1 ). The reasons for the poorer outcome for IVF singletons are not fully understood, but there are indications that it is not the IVF technique per se but more maternal characteristics such as age, parity and years of involuntary childlessness that explain, at least to some extent, the higher risks. In addition, some studies have shown that children born after spontaneous conception but after a period of infertility are at higher risk of adverse outcomes. Maternal characteristics, however, do not seem to explain the increased risks completely, and it is still being discussed whether hormonal stimulation and culture conditions could contribute to adverse outcomes. Two studies, one from Norway and one from Denmark, on perinatal outcome after assisted fertilization, used a sib-ship design, in which the same mother had given birth to one child after spontaneous conception and one child after assisted reproductive technology. In contrast to the Norwegian results, which after adjustment for maternal confounders found no difference in outcome, the larger Danish study found that the risk of low birth weight and pre-term birth was higher in siblings conceived through assisted reproductive technology after fresh embryo transfer than in the spontaneously conceived siblings after adjustment for maternal age and birth order. Maternal complications such as pre-eclampsia, placenta praevia and gestational diabetes occur more often in IVF pregnancies.
| Complication | Odds ratio and relative risk | 95% Confidence interval | References |
|---|---|---|---|
| Preterm delivery (<37 weeks) | 1.95 | 1.73–2.20 | Jackson et al., 2004 |
| 2.04 | 1.80–5.28 | Helmerhorst et al., 2004 | |
| 1.98 | 1.77–2.22 | McGovern et al., 2004 | |
| 1.93 | 1.36–2.74 | McDonald et al., 2005 | |
| 1.84 | 1.54–2.21 | McDonald et al., 2009 | |
| Preterm delivery (<32–33 weeks) | 3.10 | 2.00–4.80 | Jackson et al., 2004 |
| 3.27 | 2.03–5.28 | Helmerhorst et al., 2004 | |
| 2.49 | 0.86–7.21 | McGovern et al., 2004 | |
| 2.99 | 1.54–5.80 | McDonald et al., 2005 | |
| 2.27 | 1.73–2.97 | McDonald et al., 2009 | |
| Low birth weight (<2500 g) | 1.77 | 1.40–2.22 | Jackson et al., 2004 |
| 1.70 | 1.50–1.92 | Helmerhorst et al., 2004 | |
| 1.40 | 1.01–1.95 | McDonald et al., 2005 | |
| 1.60 | 1.29–1.98 | McDonald et al., 2009 | |
| Very low birth weight (<1500 g) | 2.70 | 2.31–3.14 | Jackson et al., 2004 |
| 3.00 | 2.07–4.36 | Helmerhorst et al., 2004 | |
| 3.78 | 2.29–5.75 | McDonald et al., 2005 | |
| 2.65 | 1.83–3.84 | McDonald et al., 2009 | |
| Small for gestational age (birth weight < 10th percentile) | 1.60 | 1.25–2.04 | Jackson et al., 2004 |
| 1.40 | 1.15–1.71 | Helmerhorst et al., 2004 | |
| 1.59 | 1.20–2.11 | McDonald et al., 2005 | |
| 1.45 | 1.04–2.00 | McDonald et al., 2009 | |
| Caesarean section | 2.13 | 1.72–2.63 | Jackson et al., 2004 |
| 1.54 | 1.44–1.66 | Helmerhorst et al., 2004 | |
| 1.81 | 1.41–2.32 | McDonald et al., 2005 | |
| Perinatal mortality | 2.19 | 1.61–2.98 | Jackson et al., 2004 |
| 1.68 | 1.11–2.55 | Helmerhorst et al., 2004 | |
| 2.40 | 1.59–3.63 | McDonald et al., 2005 | |
| Congenital malformation | 1.29 | 1.01–1.67 | Rimm et al., 2004 |
| 1.29 | 1.21–1.37 | Hansen et al., 2005 | |
| 1.41 | 1.06–1.88 | McDonald et al., 2005 | |
| Neonatal care | 1.60 | 1.30–1.96 | Jackson et al., 2004 |
| 1.27 | 1.16–1.40 | Helmerhorst et al., 2004 | |
| 1.36 | 1.20–1.54 | McDonald et al., 2005 | |
| Cerebral palsy | 1.82 | 1.31–2.52 | Hvidtjorn et al, 2009 |
| Cancer | 1.4 a | 0.6–3.1 | Bruinsma et al., 2000 |
| 0.9 a | 0.4–2.0 | Klip et al., 2001 | |
| 1.4 a | 0.9–2.0 | Källén et al., 2005 | |
| 1.42 b | 1.09–1.87 | Källén et al., 2010 |
a Standardised incidence ratio.
Recent reports concerning maternal and child outcomes have shown an improvement for IVF pregnancies and IVF children. In a large Swedish registry study of IVF singletons, the preterm birth rate was 7.2% compared with 5% in the general population, and the low birth weight rate was 5% compared with 3.2%. In this study, the outcome for singletons after SET (both elective SET and non-elective SET) was similar to the outcome for singletons after DET. A better outcome for the SET singletons might have been expected, as the ‘vanishing twin’ phenomenon, predicting a worse outcome, only operates in singletons from DET. In addition, previous smaller studies indicated a better outcome for SET singletons.
Different in-vitro fertilisation techniques
Intracytoplasmic sperm injection
Intracytoplasmic sperm injection (ICSI) was a major advancement in IVF. Today, around 50% or even more of IVF cycles use ICSI. Since the introduction of ICSI, concerns about its safety have persisted, as ICSI is a more invasive procedure than conventional IVF, natural selection of sperm is set aside, and the use of sperm of lower quality may imply a greater risk of chromosomal and genetic abnormalities. No adverse effects of ICSI on perinatal outcomes, however, are indicated compared with IVF. Published studies indicate either better or similar perinatal outcome in ICSI. One suggested reason for a better outcome for ICSI pregnancies is a healthy female partner.
Cryopreservation
Another important step forward was the introduction of freezing and thawing of embryos. Recently, a new technique, vitrification, has been developed, which also seems to work for oocytes. This will give young women suffering from malignant diseases, without a partner, and treated with cytotoxic agents (often resulting in ovarian failure), a chance to achieve a pregnancy. Freezing of oocytes before treatment with cytotoxic drugs might then be possible. A further development is freezing of ovarian tissue before cytotoxic treatment, and later auto-transplantation. This technique is still under development and only rather few children have been born by this method.
Studies on child outcome after freezing of embryos have generally been reassuring. Most studies have been carried out on cleavage-stage embryos and after slow freezing. Few studies have been published concerning obstetric outcome after vitrification. For cleaved embryos and after slow freezing, controlled studies have indicated a better or at least as good obstetric outcome, measured as preterm birth and low-birth weight, compared with children born after fresh cycles.
Three large registry studies, all from Scandinavia confirmed better outcome for children from cryo-cycles for most variables compared with fresh cycles. Compared with singletons from the general population, however, the outcome for the cryo-singletons was less good. Some concerns have been raised in these Scandinavian studies, all reporting a higher large for gestational age rate compared with fresh cycles and with the general population. In addition, a recent Danish sib-ship study, which included sibling pairs where one singleton was born after fresh and the other after freezing and thawing revealed that, after adjustment for maternal age and birth order, the higher mean birth weight in cryo singletons remained, even with the same mother. The interpretation of these findings is unclear but the cryo techniques may induce changes in the early embryo stages and in intrauterine growth potential. Macrosomic babies are at increased risk of adverse perinatal outcomes such as stillbirth, birth asphyxia, shoulder dystocia, hypoglycemia, respiratory distress and perinatal mortality. Macrosomia has also been reported as a risk factor for childhood cancer.
In animals, the association between IVF and ‘large offspring syndrome’, in particular related to cattle and sheep, is well known and closely related to severe malformations. No increase in malformations has been noticed for cryo-children compared with children from fresh IVF cycles, and there is no reason to assume that the higher large for gestational age rate in cryo-children is related to the ‘large offspring syndrome’ observed in animals.
Few studies on vitrification of oocytes, cleaved embryos and blastocysts have been published. One recent study compared child outcome after vitrified blastocysts, fresh blastocysts and slow freezing of early cleavage-stage embryos. No significant differences were reported in the rates of preterm birth, very preterm birth, low birth weight, very low birth weight or mortality between the groups. A higher small for gestational age rate was reported in singletons born after fresh blastocysts compared with children born after vitrified blastocysts. Studies on child follow up after vitrification of oocytes are still rare. The largest study reported on 200 children. The mean birth weight for singletons was 2920 g. The low birth weight rate among singletons was 18% and the prematurity rate was 26%. There was no control group.
Blastocysts
Culture of embryos 5–6 days, to the blastocyst stage, instead of 2–3 days, is clinical routine in many countries, particularly in the USA. Many studies, including a Cochrane review, have shown higher pregnancy and delivery results after blastocyst transfer compared with transfer day 2–3. Disadvantages with the blastocyst technique, in addition to more work load, is that some women will not receive an embryo transfer as no embryos survive to the blastocyst stage. In addition, when culturing to days 5–6, fewer embryos will be left for freezing. Nevertheless, the Cochrane review showed higher delivery rates after blastocyst transfer compared with days 2–3. No difference in the total delivery rates, including fresh and frozen cycles, have however been observed between the blastocyst and early cleavage-stage strategies. The introduction of vitrification as a method for freezing blastocysts with improved results might increase the use of blastocyst culture. Few studies have been published concerning child follow up after blastocyst transfer. In the Swedish follow-up study, a small but significantly increased risk for preterm birth and malformations was detected for singleton children born after blastocyst transfer compared with children born after days 2–3 transfer. In another large registry study from Australia, no differences were found in preterm birth or low birth weight.
Preimplantation genetic diagnostics
Preimplantation genetic diagnostics (PGD) is an advanced method for diagnosing severe genetic diseases at the embryo stage. An embryo biopsy is carried out in order to determine structural and numerical chromosomal imbalances, specific monosomic defects and aneuploidy. Using fluorescence in-situ hybridisation, polymerase chain reaction, or both, a decision is made concerning whether the blastomere has a normal or an abnormal karyotype. Only normal embryos are transferred. Compared with traditional fetal diagnosis, PGD allows women to avoid late abortion with negative psychological consequences. The disadvantage is the need for IVF. More than 5000 children have been born after PGD. Follow-up of children has shown good results, with no indication that the PGD technique adds any extra risks compared with IVF without PGD.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
