Correspondence: Professor P.G. Crosignani, Clinica Ostetrica e Ginecologica I, Facolta de Medicina e Chirurgia, Universita Degli Studi di Milano, Via Commenda, 12-20122 Milano, Italy. E-mail: piergiorgio.crosignani{at}unimi.it
![]() |
Abstract |
---|
Key words: assisted reproduction/embryo transfer/multiple pregnancy/ovarian stimulation/pregnancy rates
![]() |
Introduction |
---|
This workshop first considered the maternal risks and the perinatal and long-term outcomes of multiple pregnancies and then proposed possible means of reducing them.
![]() |
Spontaneous twinning |
---|
Monozygotic twinning
Spontaneous embryo division between day 0 and day 4 post-fertilization results in diamniotic, dichorionic, diplacental MZ twins, division between days 4 and 8 post-fertilization in diamniotic, dichorionic, monoplacental MZ twins and division after day 8 in monoamniotic, monochorionic, monoplacental MZ twins. Even later (incomplete) division gives rise to conjoined twins. An excess of monozygotic twins occurs in mothers who are themselves one of monozygotic twins (Lichtenstein et al., 1996), whereas there is no evidence for a paternal effect on monozygotic twinning (Lichtenstein et al., 1998
). This suggests that there is a genetic component to the increased likelihood of an embryo to divide, and that this genetic component is expressed in the mother rather than in the fertilized oocyte. However, other factors may be of importance in general or in specific assisted reproductive treatment procedures, as reflected in the recently reported increased incidence of MZ twinning as a consequence of ovarian stimulation (Derom et al., 1991
) and following the transfer of blastocysts after human IVF/ICSI (Rijnders et al., 1998
).
Dizygotic twinning
Fertilization of two simultaneously ovulated oocytes results in dizygotic twinning. The incidence of excessive (>1 follicle of >11 mm diameter) follicular recruitment is ~10% (3/31 cycles; Martin et al., 1991a) in mothers with singleton pregnancies and in mothers of MZ twins, whereas it is 31% (24/77 cycles; Martin et al., 1991a
) in mothers of DZ twins. Mothers of DZ twins have significantly higher basal FSH concentrations and FSH pulse frequency (Lambalk et al., 1998
). This is accompanied by elevation of follicular phase oestradiol, inhibin and LH concentrations (Martin et al., 1991b
), suggesting that the primary cause of multiple ovulation in humans is not a decrease in inhibin secretion from the ovary, that the increased secretion of FSH and LH may be caused by elevated secretion of, or sensitivity to, GnRH, and that the elevated inhibin and oestradiol concentrations are a response to the resulting increased FSH release by the pituitary. Additional evidence for this phenomenon of extragonadally determined DZ twinning is the finding that in populations with high twinning rates (e.g. Western Nigeria: 50/1000 maternities) the basal FSH concentration in women is higher than in women from populations with low twinning rates (e.g., Scotland: 12/1000 maternities; Nylander, 1981). There is a slight tendency for twins to have been conceived during summer as compared with singletons [odds ratio (OR) 1.3; Dionne et al., 1993]. When stratified by concordant sex versus discordant sex, more discordant sex twin pairs were conceived during summer than corresponding singletons (OR 1.7), whereas no such finding was observed for concordant sex twins (OR 1.1; Dionne et al., 1993). This may reflect the effect of increased light during summer on the pineal gland, resulting in decreased melatonin production, and consequent decreased inhibition of pituitary FSH release. This again suggests that multiple ovulation is extragonadally determined.
Conclusion
MZ twinning is a less frequent cause of twinning in the human (0.4% of maternities) than DZ twinning (1.2%). The DZ twinning rate is related to a wide variety of endogenous and exogenous factors and appears to be mediated by pituitary FSH release. The MZ twinning rate is relatively constant, but may, in assisted reproductive treatment procedures, be increased, e.g. by late (blastocyst) embryo transfer.
![]() |
Iatrogenic twinning |
---|
Induction of ovulation
Treatment of women with anovulatory infertility is highly successful (Baird, 1993). Whenever possible, the treatment should aim to restore the feedback system which selects a single follicle for ovulation. Thus, treatment of hyperprolactinaemic women with dopamine agonists and hypogonadotrophic hypothalamic amenorrhoea with pulsatile gonadotrophin-releasing hormone (GnRH) is associated with a high incidence of single ovulations. The first line treatment of women with normogonadotrophic anovulation, including polycystic ovarian syndrome (PCOS), is clomiphene or other anti-oestrogens (Adashi, 1996
). The incidence of twins is increased to 10% and that of triplets ~1%. Treatment with gonadotrophins should be confined to those who are resistent to clomiphene, i.e. remain anovulatory, because the risk of multiple pregnancy is considerable (~25%) with conventional `step up' regimens (Fauser and van Heudsen, 1997). The low dose `step up' or `step down' regimens which aim to maintain the physiological principles of follicle selection, result in a high incidence of monovulation but at the price of slightly lower pregnancy rates (White et al., 1996
).
In-vitro fertilization
IVF was initially attempted after recovery of the oocyte from the single pre-ovulatory follicle. However, it was soon realized that the pregnancy rate per cycle could be increased by replacing more than one embryo in the uterus after collecting many oocytes from multiple follicles (Edwards et al., 1996). A major factor affecting the likelihood of a multiple birth was the number of embryos replaced and subsequent analysis has demonstrated that an increase in the number of embryos transferred invariably results in a higher likelihood of multiple birth, but without necessarily improving the overall success rate of IVF (Templeton and Morris, 1998
).
Some studies have suggested that there is an improvement in birth rate with an incremental increase in the number of embryos replaced, but the UK study indicated that a more important factor was the number of oocytes fertilized and hence embryos available for transfer. In all situations where more than four embryos were available for transfer, there was no greater birth rate for women receiving three embryos than for those receiving two, but there was a considerable increase in the rate of multiple births, and especially triplets. A decision analysis based on these data has demonstrated that in most circumstances two embryos can be transferred without anticipating any significant diminution in the expected pregnancy or birth rates.
In several countries three-embryo transfers are now almost unknown, with a resultant reduction in the triplet rate. Furthermore, there are now data addressing the issue of one-embryo transfers (Vilska et al., 1999). Many centres now limit the number of embryos replaced (two or three) in order to reduce the risk of multiple births (Templeton and Morris, 1998
). The remaining `spare' embryos are frozen and replaced in subsequent cycles if pregnancy does not occur.
While treatment with gonadotrophins in combination with GnRH analogues gives a high yield of oocytes, there are several disadvantages, including the risk of hyperstimulation, the unpredictability of response and the necessity to store spare embryos. Recent strategies, designed to limit the number of follicles stimulated and improve the quality of the oocytes, include a natural cycle in combination with GnRH antagonists to prevent the LH surge or the use of clomiphene and small doses of gonadotrophins (Diedrich et al., 1994). It is possible to maintain a pregnancy rate per cycle of ~20% using these regimens because of improvements in methods of egg collection (e.g. transvaginal ultrasound), and advances in methods of fertilization (e.g. ICSI and embryo culture).
However, the reduced number of oocytes results in fewer embryos available for freezing. This approach would be more valuable if we had non-invasive methods of assessing the developmental competence of individual embryos.
Intrauterine insemination in combination with ovulation induction
In couples with unexplained infertility and male subfertility, the pregnancy rate per cycle is increased following intrauterine insemination (IUI), particularly if treatment is combined with ovarian stimulation (ESHRE Workshop, 1996). It is likely that the increased pregnancy rate is related to the larger number of eggs available for fertilization. Unfortunately, there is a parallel increase in the incidence of multiple pregnancies, many of which are unanticipated (Cohlen, 1998).
If this treatment is to be offered, it is essential to monitor the response with serial measurements of follicle growth by ultrasound. If the number of large follicles exceeds three, the cycle should be cancelled and the couple advised to abstain from intercourse. An alternative is to convert the cycle to IVF, in which a limited number of embryos can be transferred.
Even with careful monitoring it is impossible to completely avoid the risk of multiple pregnancies. The same applies to gamete intraFallopian transfer, in which the number of fertilized ova remains unknown and ovulation of follicles which have not been aspirated increases the potential for multiple pregnancy. In view of the difficulty of controlling the number of ovulatory follicles and the relative simplicity of IVF, it is doubtful that the benefits of these methods of treatment outweigh the hazards described in the two following paragraphs.
![]() |
Twin pregnancy: maternal risks and perinatal outcome |
---|
![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() |
---|
|
|
The most serious maternal complication is hypertension, with or without proteinuria. The prognosis is fairly good for pregnancy-induced hypertension with an accompanying substantial rise in plasma volume. Hypertension associated with haemoconcentration and proteinuria (so-called pre-eclampsia) has a much worse prognosis. In addition, multiple gestation appears to neutralize the protective effect of parity generally found in successive singleton pregnancies. Although some authors (e.g. Santema et al., 1995) report no differences between singleton and twin pregnancies in this respect, women expecting twins are at an increased risk for pre-eclampsia even when not in their first pregnancy. The incidence of anaemia is doubled in twin pregnancies. The dilutional anaemia associated with pregnancy is often aggravated by the iron deficiency generally accentuated in multiple gestations. Therefore, iron supplementation and stricter surveillance of haemoglobin concentration is required. The additional increased risk of postpartum haemorrhage in multiple gestation makes careful monitoring of haemoglobinaemia mandatory. The greater incidence of uterine atony and dystocia (both contributing factors towards postpartum haemorrhage) make the administration of i.v. oxytocin, i.m. methylergometrine and prostglandins highly advisable. Other obstetrical problems include increased operative delivery, uterine rupture, and preterm labour, all with consequent prolonged hospitalization.
Multiple gestations require more intensive clinical and ultrasonographic monitoring. The frequency of these investigations is largely determined by the number of chorions present. Serial ultrasonographic measurements are very important. These can greatly improve the outcome of twin pregnancies in various ways. Early determination of placentation (the ultrasonographic finding of the so-called lambda sign has a 100% sensitivity and 86% specificity in diagnosing a dichorionic placenta) (Wood et al., 1996) is fundamental and allows optimum subsequent clinical management. Monochorionic placentae are associated with highly increased and specific risks. Furthermore, ultrasound enables detection of fetal anomalies, evaluation of individual fetal growth and of possible growth discrepancies between the twins.
Intrauterine growth retardation (IUGR) is a frequent occurrence in twin pregnancy. Twin fetuses are generally lighter than singleton ones. However, this is not a good reason for using adapted normograms to evaluate twins' growth. The utilization of such charts can lead to an underestimation of IUGR. Growth-retarded fetuses are frequent among twins. Monochorionic placentae are more commonly linked with growth retardation as compared to dichorionic ones. Basing their results on the observation of more than 1000 twins, Ananth et al. (1998), found that monochorionic twins weighed significantly less on average than dichorionic ones. Hill et al. (1994) showed the usefulness of abdominal circumference (AC) measurements in the evaluation of growth discrepancy between twins. This parameter has a lower sensitivity but a higher positive predictive value (PPV) than weight estimation. Fetal growth was assessed in a group of 247 twin pregnancies, by AC evaluation. AC [expressed as number of SD from the mean for gestation (AC Z-score)] was found to decrease progressively with increasing gestational age (P = 0.001) in both dichorionic and monochorionic twins. Moreover, the difference between the AC Z-scores of the twins did not change with gestational age, but seemed to be significantly correlated with chorionicity until 28 weeks (P < 0.05) and with sex-linked discordance at higher gestational ages (Bryan et al., 1997).
IUGR and prematurity are the most important factors that determine neonatal morbidity and mortality of twins. 42% of twins and only 8% of singletons are born before the end of the 37th week of gestation (Spellacy et al., 1990). In addition, twins are exposed to a 6.6-fold increased risk of dying before the end of the first year of life. Those who survive beyond this time still have a 1.4-fold greater probability of suffering from some debilitating condition (Luke and Keith, 1992
).
Finally, Spellacy et al. showed that fetal mortality for twins is lower than the mortality for singletons weighing <2500 g. Some protective factor seems to be involved in twins. However, twins weighing >2500 g have a higher death rate (Spellacy et al., 1990). Therefore, twins would appear to have an `ideal' birthweight. Careful monitoring should possibly also be aimed at planning to deliver them if such birthweight is exceeded.
![]() |
Twin pregnancy: long-term outcome |
---|
Children may suffer long-term consequences of these perinatal complications. Many forms of disability, especially cerebral palsy, are more common among multiples. There is a 37-fold increase in cerebral palsy amongst twins and over 10-fold in triplets (Petterson et al., 1990). Furthermore, a single survivor of a monochorionic twin fetal death is at the greatest risk.
Even when the babies are healthy, shared parental attention, triadic communication and their own intrapair relationship may all adversely affect their language development (Mittler, 1976) and those with delays in language development have a greater chance of developing reading difficulties later. Individuation may also be delayed and incomplete and intense interdependence or competition can lead to enduring behaviour problems.
Depression is more common in mothers of multiple birth than with singletons (Thorpe et al., 1991), as are behavioural problems among the siblings (Hay et al., 1988
).
If one twin dies, the parents face special problems and the surviving twin may suffer from the bereavement into adulthood (Bryan, 1995).
Guidelines for couples and doctors (Bryan et al., 1997)
To couples considering treatment for infertility
During the pregnancy
Birth and the neonatal period
The pre-school years
Encourage parents:
![]() |
Embryo reduction |
---|
The procedure is variably named: selective abortion, selective reduction, but the preferred term is now multifetal pregnancy reduction (MFPR). It is now recognized as a safe and effective method to improve outcome in multiple pregnancies, without doubt in quadruplet and higher order pregnancies, arguably in triplets. The procedure is performed under ultrasonic control at 11 or 12 weeks gestation by injecting into the chest of each fetus 12 ml of KCl, 2 N (Dechaud et al., 1998; Kadhel et al., 1998
; Mansour et al., 1999
). The embryos selected for reduction are those which are in the upper part of the uterus, those which have increased nuchal translucency or other ultrasonographic markers of risk. Some authors advocate karyotyping by chorionic villous sampling prior to MFPR, but this requires an additional procedure and, when the results of the karyotype become available, it is not always easy to be certain which embryo has a given karyotype. In addition, if two viable embryos are left, the risk of chromosomal abnormality per pregnancy is only twice that of the age of the woman and subsequent amniocentesis could be performed at a later stage, if the woman wishes to do so. The results of the international collaboration on MFPR published in January 1999 (Evans et al., 1999
) regarding 3037 completed pregnancies show that the fetal losses prior to 24 weeks gestation following MFPR have decreased from 13.2% in the period 19861990 to 7.5% in the period 199597 and that the percentage of women who delivered at term increased from 35.4% to 52.2% during the same time intervals (Evans et al., 1996
). These data underline the importance of increasing experience in reducing unfavourable outcomes following MFPR. Total loss rate is also related to the starting and finishing number of embryos. Table III
shows that for any starting number of embryos, the fewest losses occur for a pregnancy reduced to two fetuses.
|
![]() |
Economic consequences |
---|
The probability of a multiple gestation pregnancy (MGP) is higher in assisted reproductive treatment cycles because of the stimulation of excess follicles in HMG cycles and the transfer of more than one embryo in IVF or IVF/ICSI cycles. MGP can be a rewarding fulfilment for infertile couples, but there are also psychological, medical, social and financial concerns for parents. The financial concerns arise mainly from the direct costs of prematurity: neonatal intensive care, drug therapy, inhalation therapy, imaging and other diagnostic procedures. Higher costs also are incurred after discharge, due to chronic health problems, visual problems and developmental disabilities.
This assessment of the economic consequences of MGP is restricted to the cost of delivering twins, triplets and quadruplets. It does not encompass the cost of pregnancy losses with MGP, or cost of fetal reduction. The coverage is also limited to assisted reproductive treatment cycles.
Assessments of such broad topics are severely limited by the specific nature of the relevant literature. Thus, a number of assumptions are made about costs and trends in order to bridge gaps in the available data. To emphasize the tentative nature of the results, the computations are projected to the year 2000.
Probability of MGP in assisted reproductive treatment cycles
A 1999 report on a factorial trial of superovulation or no stimulation and intracervical or IUI found that 23.6% of 72 superovulation live births were twins, 5.6% were triplets and 4.2% were quadruplets (Guzick et al., 1999).
The Society for Assisted Reproductive Technology/American Society for Reproductive Medicine (SART/ASRM) registry reports for 1992 to 1995 noted that MGP accounted for 33.7%, 34.1%, 36.3% and 35.9% of live births after IVF in 1992, 1993, 1994 and 1995, respectively (AFS and SART, 1994; SART and ASRM, 1995; SART and ASRM, 1996). In 1992, 26.0% of deliveries were twins, 5.8% were triplets and 0.4% were higher order MGP (AFS and SART, 1994). In 1993, 27.5% of deliveries were twins, 5.4% were triplets and 0.4% were higher order MGP (SART and ASRM, 1995). In 1994, 28.3% of deliveries were twins, 5.9% were triplets and 0.6% were higher order MGP (SART and ASRM, 1996). In 1995, the last year reported, 29.6% of deliveries were twins, 6.4% were triplets and 0.6% were higher order MGP (SART and ASRM, 1998). The 1992 to 1995 trend was a 7% increase in number of procedures and a 9% increase in MGP.
A 1998 report from Cairo, Egypt, followed 702 pregnancies from 3500 oocyte retrievals to late gestation. The MGP rate was 27.9%, comprising 19.8% twins, 6.0% triplets and 2.1% higher order multiple gestations (Serour et al., 1998).
In the UK, among 29 262 transfers of three embryos, 1755 of 6091 deliveries (28.8%) were twins and 5.8% were triplets or more (Templeton and Morris, 1998). With more than two embryos available, transferring only two embryos reduced twinning to 26.1% and triplets occurred in only 0.4% of deliveries. Live births per transfer were 22% with two embryo transfers compared with 21% for three embryo transfers. Higher order births were 16.7-fold more likely (95% CI 8.1, 32.9) after three embryos than after two embryo transfers.
The SART/ASRM data are consistent with other reports and the 1992 to 1995 trend can be projected to model the likelihood of multiple gestation in the year 2000 as a percentage of live births in assisted reproductive treatment cycles. The same trend was applied to the data of Guzick et al. (1999) for HMG/IUI cycles.
For superovulation cycles in the year 2000, the projected MGP rate is 36.5%, including 25.8% twins, 6.1% triplets and 4.6% quadruplets or more.
For IVF and IVF/ICSI cycles in the year 2000, the projected MGP rate is 43.8%, including 35.4% twins, 7.4% triplets and 1.0% quadruplets or more.
Cost of MGP
Costs may be higher for assisted reproductive treatment MGP births than for non-assisted reproductive treatment MGP births because IVF twins are at higher risk for prematurity than non-IVF twins (Moise et al., 1998).
Neumann et al. (1994) assumed that the costs per family were US$20 000 for twins and US$200 000 for higher order MGP. They loaded these costs onto the initial IVF cycle cost (US$8000), adding US$480 to the initial cost of all cycles to cover the cost of twins and US$720 to cover costs of higher orders of multiple gestation.
Callahan et al. (1994) found that the average charges per family in 1991 were US$9845, 37 947 and 109 765, respectively, for singleton, twin and triplet deliveries. Assisted reproductive treatment procedures accounted for 35% of twin and 77% of higher order MGP. The estimates were based on 1125 twin and 85 higher order gestations.
Luke et al. (1996) estimated the 1991 costs for 111 non-assisted reproductive treatment twins born at mean gestation of 34.7 weeks. Infant costs and maternal costs averaged US$38 295 and US$15 245, compared with US$33 098 and US$11 252 for singletons matched for gestational age and US$2812 and US$6514 for singletons matched only for race and payer status.
Chelmow et al. (1995) estimated the 1993 costs for 20 sets of non-assisted reproductive treatment triplets born at a mean gestation of 30.2 weeks. Maternal costs and infant costs were US$27 981 and US$36 856 per family.
The Callahan et al. (1994) estimates are deemed to be sufficiently representative to use for modelling. The Callahan et al. (1994) twin cost estimate is similar to that of Luke et al. (1996). The Callahan et al. triplet cost estimate is intermediate between the estimates of Neumann et al. and Chelmow et al. Cost estimates for higher order quadruplet gestations are not available. Therefore the Callahan et al. twin and triplet costs have been extrapolated to quadruplet costs. An error in this projection would not be serious, given the small proportion of quadruplet pregnancies in the total assisted reproductive treatment experience. Note that the Callahan et al. estimate for the cost of a singleton delivery has not been deducted from the MGP delivery costs.
Allowing for inflation of health care cost (at 5% per year), the projected year 2000 costs of delivering multiple gestation pregnancies arising in assisted reproductive treatment cycles per family are as follows.
National cost of MGP in assisted reproductive treatment cycles
Number of cycles per annum and success rates
SART/ASRM 1992 to 1995 trends projected to 2000 suggest there will be 57 317 IVF and IVF/ICSI cycles per annum, with a delivery rate of 37.4 per cycle. It has been estimated that four HMG/IUI cycles are performed nationally for every three IVF cycles (Collins et al., 1997), projecting 76 423 HMG/IUI cycles in 2000. The HMG/IUI success rate used is that of Guzick et al. (1999) (8% in 1997) projected at the upward 9% trend in IVF/ICSI success rates to yield 10.4% deliveries per cycle (Guzick et al., 1999
).
Cost of cycles
The cost of HMG and IUI in 1995 was US$1800 (van Voorhis et al., 1995), projected to US$2309 in 2000. The average cost of IVF cycles in 1993 was US$6233 (Collins et al., 1995
), projected to US$8770 in 2000.
National cost of HMG/IUI and MGP in 2000
The national cost of HMG/IUI in 2000 is projected at US$176.5 million. The additional cost of MGP is US$303.6 million, comprising US$120.2, 81.8 and 101.5 million for twins, triplets and quadruplets, respectively. The trend in costs from 1992 to 2000 shows that MGP costs began to exceed HMG/IUI cycle costs in 1996. The additional cost for MGP in 2000 would be reduced to US$221.1 million if all quadruplets were reduced to twins (assuming a 10% pregnancy loss rate with the procedure).
National cost of IVF/ICSI and MGP in 2000
The national cost of IVF/ICSI in 2000 is projected at US$502.7 million. The additional cost of MGP is US$875.9 million, comprising US$503.3, 303.2 and 69.4 million for twins, triplets and quadruplets respectively. The trend from 1992 to 2000 shows that MGP costs began to exceed HMG/IUI cycle costs in 1996, when the procedures cost 312.0 million and the MGP costs were 319.8 million. The additional cost for MGP in 2000 would be reduced to US$526.1 million if only two embryos were transferred. Reducing health care cost inflation to 3% did not alter the estimates appreciably.
National cost of infertility diagnosis and treatment and MGP
Table IV shows the effect of the estimated costs of MGP on a previously published model of national spending for infertility diagnosis and management (Collins et al., 1997
). The projected cost of MGP (Canadian US$269.9 million) is nearly half the cost of all infertility diagnosis and treatment.
|
For other national settings a prudent interpretation of the economic modelling in this chapter would focus on the relative costs associated with single, twin, triplet and higher order deliveries. Compared to singleton deliveries the costs are approximately 4-, 11- and 18-fold greater for twins, triplets and higher order deliveries, respectively.
![]() |
Summary and conclusions |
---|
Spontaneous twinning
The origins and implications of multiple pregnancy have been studied more for spontaneous twinning than triplet or higher orders. Monozygotic twinning is less frequent and it is associated with a genetic increased liability of an embryo to divide. There is a higher perinatal mortality because of the greater likelihood of sharing a single gestational sac. Dizygotic twinning is associated with increased basal FSH concentrations and excessive follicular recruitment. Dizygotic twinning occurs in 1.2% of births compared with 0.4% for monozygotic twinning.
Iatrogenic twinning
Ovulation induction should be carried out when possible with specific treatments such as bromocriptine and pulsatile GnRH which tend to induce single ovulations. With normogonadotrophic anovulation the first line of treatment for ovulation induction should be clomiphene which is associated with twin rates up to 10% and triplet rates up to 1%. Gonadotrophin treatment should be reserved for clomiphene-resistant women because of the high incidence of multiple gestation. Low dose regimens should be encouraged.
IVF multiple pregnancy rates are increased by the transfer of excess embryos in the hope of improving pregnancy rates. Data from different countries suggests that success depends more on the number of embryos available for transfer than the number transferred. Preventing multiple gestation in assisted reproductive treatment cycles could be achieved by means of strict guidelines for embryo transfer. Improved technology will also help: better recognition of high quality embryos; delaying implantation to the blastocyst stage when fewer but healthier embryos are available; and advances in natural cycle IVF.
Superovulation in combination with IUI has a high incidence of multiple births including quintuplets and higher orders. Careful cycle monitoring is essential, but even that does not avoid the risk of multiple gestations. Conversion of cycles with excess follicles to IVF is another prevention technique. Superovulation in combination with IUI is not a very safe technique and its benefits may not outweigh the associated hazards of MGP.
Maternal risks and perinatal outcome
Maternal risks such as hypertension and maternal mortality are 3- to 7-fold higher in multiple pregnancies than in singleton pregnancies. Twin and triplet pregnancies are associated with increased rates of IUGR, early pregnancy losses and premature delivery. Perinatal mortality rates are at least 4-fold higher for twins and 6-fold higher for triplets.
Long-term outcomes for twins and higher order multiple births are compromised from the start by the problems of prematurity and monochorionic placentation. In addition many forms of disability including cerebral palsy are more common in multiple birth progeny. Even when the babies are healthy, numerous factors contribute to adversely effect their language development and chance of reading difficulties.
Information about multiple births and their consequences should be provided for couples considering treatment for infertility. For parents with multiple gestations, detailed information should be available throughout the pregnancy, birth and prenatal period and further into early childhood.
Economic consequences
The 19921995 US trends projected to the year 2000 predict more than 57 000 IVF/ICSI cycles initiated, a 38% delivery rate and 21 454 deliveries. The projected 43% multiple birth rate means that of the 32 865 babies born, 73% will be in a multiple birth. The family costs per multiple gestation delivery are approximately 4-, 11- and 18-fold greater for twins, triplets and higher order deliveries, respectively, compared to singleton deliveries.
The multiple birth cost multiplied by the projected multiple birth frequency brings the total cost of assisted reproductive treatment multiple births in the USA in the year 2000 to US$876 million, compared with US$503 million for all IVF and ICSI cycles. Thus the estimated annual cost of assisted reproductive treatment multiple pregnancy births is greater than the cost of all IVF and ICSI cycles performed.
Prevention of multiple births
As noted above, the principal means of reducing multiple births from assisted reproductive treatment cycles is primary prevention. The approaches include cancelling superovulation cycles as necessary and improved embryo transfer techniques in IVF/ICSI cycles. Research continues to evaluate single embryo transfer protocols. Secondary prevention by means of embryo reduction involves difficult decisions for many infertile couples and the procedure is not easily forgotten by other couples. While it should be available as a last resort, embryo reduction is not a feasible way to reduce the frequency of assisted reproductive treatment-associated multiple births.
At present all registries report total live birth rate, which includes the desired outcome (singletons) combined with the adverse outcome (multiple births). Reporting the combined singleton and multiple birth rates in this way obscures the true success rate. Prevention would be enhanced if the singleton live birth rate were the principal outcome of assisted reproductive treatment reported by registries and clinical services. Since multiple births are a complication of the procedure, twin and triplet rates should be reported separately.
Successful IVF is a historic development in the treatment of infertility. The increase in multiple births is a barrier to further progress because despite their human appeal, multiple births generate numerous health and social concerns. Infertility specialists in all countries should address the prevention of multiple births in the most appropriate way, by means of improved clinical procedures.
![]() |
Notes |
---|
![]() |
References |
---|
American Fertility Society/Society for Assisted Reproductive Technology (1994) Assisted reproductive technology in the United States and Canada: 1992 results generated from The American Fertility Society/Society for Assisted Reproductive Technology Registry. Fertil. Steril., 62, 11211128.[ISI][Medline]
Ananth, C.V., Vintzileos, A.M., Shen-Schwarz, S. et al. (1998) Standards of birth weight in twin gestations stratified by placental chorionicity. Obstet Gynecol., 91, 917924.
Baird, D.T. (1993) Ovulation induction. In Adashi, E.Y. and Leung, P.C.K. (eds), The Ovary. Raven Press, New York, pp. 529544.
Bortolus, R., Parazzini, F., Chatenoud, L. et al. (1999) The epidemiology of multiple births. Hum. Reprod. Update, 5, 179187.
Berkowitz, R.L., Lynch, L., Stone, J. and Alvarez, M. (1996) The current status of multifetal pregnancy reduction. Am. J. Obstet. Gynecol., 174, 12651272.[ISI][Medline]
Bryan, E.M. (1995) Perinatal bereavement after the loss of a twin. In Ward, R.H. and Whittle, M. (eds), Multiple Pregnancy. RCOG Press, London, pp. 186195.
Bryan, E., Denton, J. and Hallett, F. (1997) Guidelines for Professionals: Multiple Births and their Impact on Families. Multiple Births Foundation, London.
Callahan, T.L., Hall, J.E., Ettner, S.L. et al. (1994) The economic impact of multiple gestation pregnancies and the contribution of assisted reproduction. N. Engl. J. Med., 331, 244249.
Chelmow, D., Pensias, A., Kaufman, G. and Cetrulo, C. (1995) Costs of triplet pregnancy. Am. J. Obstet. Gynecol., 172, 677682.[ISI][Medline]
Cohlen, B.J. (1998) Intrauterine insemination and controlled ovarian hyperstimulation. In Templeton, A.A., Cooke, I. and O'Brien, P.M.S. (eds), Evidence-based Fertility Treatment. RCOG Press, London, pp. 205216.
Collins, J.A., Bustillo, M., Visscher, R.D. and Lawrence, L.D. (1995) An estimate of the cost of in vitro fertilization services in the United States in 1995. Fertil. Steril., 64, 538545.[ISI][Medline]
Collins, J.A., Feeny, D. and Gunby, J. (1997) The cost of infertility diagnosis and treatment in Canada in 1995. Hum. Reprod., 12, 951958.[ISI][Medline]
Dawood, M.Y. (1996) In vitro fertilization, gamete intrafallopian transfer and superovulation with intra-uterine insemination: efficacy and potential health hazards on babies delivered. Am. J. Obstet. Gynecol., 174, 12081217.[ISI][Medline]
Dechaud, H., Picot, M.C., Hedon, B. and Boulot, P. (1998) First trimester multifetal pregnancy reduction: evaluation of technical aspects and risks from 2.756 cases in the literature. Fetal Diagn. Ther., 13, 261265.[ISI][Medline]
Denton, J. (1997) Multiple Pregnancy and Birth For Patients Considering Treatment for Infertility. Multiple Births Foundation, London.
Derom, R., Vlietinck, R., Derom, C. et al. (1991) Perinatal mortality in the East Flanders Prospective Twin Survey. Eur. J. Obstet. Gynaecol. Reprod. Biol., 41, 2526.[ISI][Medline]
Diedrich, K., Diedrich, C., Santos, E. et al. (1994) Suppression of the endogenous luteinizing hormone surge by the gonadotrophin-releasing hormone antagonist Cetrorelix during ovarian stimulation. Hum. Reprod., 9, 788791.[Abstract]
Dionne, C.E., Soderstrom, M. and Schwartz, S.M. (1993) Seasonal variation of twin births in Washington State. Acta Genet. Med. Gemellol., 42, 141149.[Medline]
Edwards, R.G., Lobo, R. and Bouchard, P. (1996) Time to revolutionize ovarian stimulation. Hum. Reprod., 11, 917918.[ISI][Medline]
Evans, M.I., Dommergues, M., Wapner, R.J. et al. (1996) International collaborative experience of 1789 patients having multifetal pregnancy reduction: a plateauing of risks and outcomes. J. Soc. Gynecol. Invest., 3, 2326.[ISI][Medline]
Evans, M.I., Wapner, R., Carpenter, R. et al. (1999) International collaborative on multifetal pregnancy reduction (MFPR): dramatically improved outcomes with increased experience. Am. J. Obstet. Gynecol., 180, S28.
ESHRE Capri Workshop Group (1996) Guidelines to the Prevalence, Diagnosis, Treatment and Management of Infertility, 1996. Hum. Reprod., 11, 17751807.[ISI]
Fauser, B.C.J.M. and Van Heusden, A.M. (1997) Manipulation of human ovarian function: physiological concepts and clinical consequences. Endocr. Rev., 18, 71106.
FIVNAT (1996) Bilan FIVNAT 1995. Contracept. Fertil. Sex., 24, 694699.[ISI][Medline]
Guzick, D.S., Carson, S.A., Coutifaris, C. et al. (1999). Efficacy of superovulation and intrauterine insemination in the treatment of infertility. N. Engl. J. Med., 340, 177183.
Hay, D.A., McIndoe, R. and O'Brien, P.J. (1988) The older sibling of twins. Aust. J. Early Child., 13, 2528.
Hill, L.M., Guzick, D., Chenevey, P. et al. (1994) The sonographic assessment of twin growth discordancy. Obstet. Gynecol., 84, 501504.[Abstract]
Kadhel, P., Olivennes, F., Fernandez, H. et al. (1998) Are there still obstetric and perinatal benefits for selective embryo reduction of triplet pregnancies? Hum. Reprod., 13, 35553559.[Abstract]
Lambalk, C.B., Boomsma, D.I., De Boer, L. et al. (1998) Increased levels and pulsatility of FSH in mothers of hereditary DZ twins. J. Clin. Endocrinol. Metabol., 83, 481486.
Landy, H.J., Weiner, S., Corson, S.I. et al. (1986) The `vanishing twin': ultrasonographic assessment of fetal disappearance in the first trimester. Am. J. Obstet. Gynecol., 155, 1419.[ISI][Medline]
Lichtenstein, P., Olausson, P.O. and Kallen, A.J. (1996) Twin births to mothers who are twins. Br. Med. J., 12, 879881.
Lichtenstein, P., Kallen, B. and Koster, M. (1998) No paternal effect on MZ twinning in the Swedish twin Registry. Twin Res., 1, 212215.[Medline]
Luke, B. and Keith, L. (1992) The contribution of singletons, twins and triplets to low birth weight, infant mortality and handicap in the United States. J. Reprod. Med., 37, 661666.[ISI][Medline]
Luke, B., Bigger, H.R., Leurgans, S. and Sietsema, D. (1996) The cost of prematurity: a casecontrol study of twins vs singletons. Am. J. Public Health, 86, 809814.[Abstract]
Mansour, R.T., Aboulghar, M.A., Serour, G.I. et al. (1999) Multifetal pregnancy reduction: modification of the technique analysis of the outcome. Fertil. Steril., 71, 380384.[ISI][Medline]
Martin, N.G., Shanley, S., Butt, K. et al. (1991a) Excessive follicular recruitment and growth in mothers of spontaneous DZ twins. Acta Genet. Med. Gemellol., 40, 291301.[Medline]
Martin, N.G., Robertson, D.M., Chenevix-Trench, G. et al. (1991b) Elevation of follicular phase inhibin and LH levels in mothers of DZ twins suggests nonovarian control of human multiple ovulation. Fertil. Steril., 56, 469474.[ISI][Medline]
Maymon, R., Herman, A., Shulman, A. et al. (1995) First trimester embryo reduction: a medical solution to an iatrogenic problem. Hum. Reprod., 10, 668673.[Abstract]
Mittler, P. (1976) Language development in young twins: biological genetic and social aspects. Acta Genet. Med. Gemellol., 25, 359365.[Medline]
Moise, J., Laor, A., Armon, Y. et al. (1998) The outcome of twin pregnancies after IVF. Hum. Reprod., 13, 17021705.[Abstract]
Neumann P.J., Weinstein, M.C. and Gharib, S.D. (1994) The cost of a successful delivery with in vitro fertilization. N. Engl. J. Med., 331, 239243.
Nylander, P.P. (1981) The factors that influence twinning rates. Acta Genet. Med. Gemellol., 30, 189202.[Medline]
Petterson, B., Stanley, F. and Henderson, D. (1990) Cerebral palsy in multiple births in Western Australia. Am. J. Med. Genet., 37, 346351.[ISI][Medline]
Radestad, A., Bui, T.H., Nygren, K.G. et al. (1996) The utilization rate and pregnancy outcome of multifetal pregnancy reduction in the nordic countries. Acta Obstet. Gynecol. Scand., 75, 651653.[ISI][Medline]
Rijnders, P.M., Van Os, H.C. and Jansen, C.A.M. (1998) Increased incidence of MZ twinning following the transfer of blastocysts. Fertil. Steril., 70, S1516 (abstract).
Santema, J., Koppelaar, I. and Wallenburg, H. (1995) Hypertensive disorders in twin pregnancy. Eur. J. Obstet. Gynecol. Reprod. Biol., 58, 913.[ISI][Medline]
Senat, M.V., Ancel, P.Y., Bouvier-colle, M.H. and Bréart, G. (1998) How does multiple pregnancy affect maternal mortality and morbidity? Clin. Ostet. Gynecol., 1, 7983.
Serour, G.I., Aboulghar, M.A., Mansour, R.T. et al. (1998) Complications of medically assisted conception in 3,500 cycles. Fertil. Steril., 70, 638642.[ISI][Medline]
Society for Assisted Reproductive Technology/American Society for Reproductive Medicine (1995) Assisted reproductive technology in the United States and Canada: 1993 results generated from the American Society for Reproductive Medicine/Society for Assisted Reproductive Technology Registry. Fertil. Steril., 64, 1321.[ISI][Medline]
Society for Assisted Reproductive Technology/American Society for Reproductive Medicine (1998) Assisted reproductive technology in the United States and Canada: 1995 results generated from the American Society for Reproductive Medicine/Society for Assisted Reproductive Technology Registry. Fertil. Steril., 69, 389398.[ISI][Medline]
Society for Assisted Reproductive Technology, The American Society for Reproductive Medicine (1996) Assisted reproductive technology in the United States and Canada: 1994 results generated from the American society for Reproductive Medicine/Society for Assisted Reproductive Technology Registry. Fertil. Steril., 66, 697705.[ISI][Medline]
Spellacy, W.N., Handler, H. and Feare, C.D. (1990) A casecontrol study of 1253 twin pregnancies from a 19821987 perinatal data base. Obstet. Gynecol., 75, 168171.[Abstract]
Templeton, A. and Morris, J.K. (1998) Reducing the risk of multiple births by transfer of two embryos after in vitro fertilization. N. Engl. J. Med., 339, 573577.
Thorpe, K., Golding, J., MacGillivray, I. and Greenwood, R. (1991) Comparison of prevalence of depression in mothers of twins and mothers of singletons. Br. Med. J., 302, 875878.[ISI][Medline]
van Voorhis, B.J., Syrop, C.H., Allen, B.D. et al. (1995) The efficacy and cost effectiveness of embryo cryopreservation compared with other assisted reproductive techniques. Fertil. Steril., 64, 64706650.
Vilska, S., Tiitinen, A., Hydén-Granskog, C. and Hovatta, O. (1999) Elective transfer of one embryo results in an acceptable pregnancy rate and eliminates the risk of multiple birth. Hum. Reprod., 14, 23922395.
White, D.M., Polson, D.W. and Kiddy, D. (1996) Induction of ovulation with low-dose gonadotropins in polycystic ovary syndrome: an analysis of 109 pregnancies in 225 women. J. Clin. Endocrinol. Metab., 81, 38213824.[ISI][Medline]
Wood, S.L., Onge, R.S., Connors, G. and Elliot, P.D. (1996) Evaluation of the twin peak or lambda sign in determining chorionicity in multiple pregnancy. Obstet. Gynecol., 88, 69.